A major new linguistic study led by Dr Morten Christiansen, professor of psychology and director of Cornell's Cognitive Neuroscience Lab found that most languages use similar sounds for common objects, body parts or adjectives. The study analysed the 100 most commons words in 3700 languages (62% of the world's 6000 languages).
Morten Christiansen explains in the Cornell Chronicle that "these sound symbolic patterns show up again and again across the world, independent of the geographical dispersal of humans and independent of language lineage".
For example, in most languages, the word for “nose” is likely to include the sounds “neh” or the “oo” sound, as in “ooze.” The word for “tongue” is likely to have “l” (as in “langue” in French). “Leaf” is likely to include the sounds “b,” “p” or “l.” “Sand” will probably use the sound “s.” The words for “red” and “round” are likely to include the “r” sound. “It doesn’t mean all words have these sounds, but the relationship is much stronger than we’d expect by chance,” says Christiansen, noting that words for body parts in particular showed an unexpectedly high association between sounds and meaning.
"Perhaps these signals help nudge kids into acquiring language," he added: “Maybe it has something to do with the human mind or brain, our ways of interacting, or signals we use when we learn or process language. That's a key question for future research."
For example researchers have shown that words for small spiky objects in a variety of languages are likely to contain high-pitched sounds, while rounder shapes contain ‘ooo’ sounds, which is known as the ‘bouba/kiki’ effect.
The researchers considered the possibility that it could be the remnant of some form of "prehistoric protolanguage" that was once spoken by the earliest humans before the evolution of modern languages, but their own analysis suggests it's more likely that these similarities are rooted in human biology.
Analysis and comments
Personally, ever since I came across Richard Dawkins's concept of extended phenotype, I have wondered how much do our genes influence our language, including the choice and range of sounds, syntax, grammar, and so on. A phenotype is the expression of all the visible characteristics of DNA, such as morphology, development, biochemical or physiological processes. Our looks, temperament, immune reactions and so on our phenotype. The extended phenotype is everything that exists outside a life being's phenotype, but that also results from that being's gene expression. For example a bird's nest is the extended phenotype of a bird's DNA. The nest could not exist without the bird's DNA. Each species of bird makes its own distinctive kind of nest, and the characteristics of that nest are instinctive, coded in the DNA. In the same way, bird songs are unique to each species and coded in their DNA. You see where I am going with this. If a bird's vocalisation, including its frequency, pitch, but also the whole melody characteristic of a particular species is an expression of their DNA, why would it be any different for humans?
It is very likely that primitive human language evolved following the same instinctive sound patterns coded in our genes as with other species. The sound range, frequency and pitch may simply be determined by the shape of our mouth, throat, uvula and vocal cords. This may be why some ethnic groups, possessing different facial morphology could naturally produce some sounds that weren't found in other languages.
For example, all people with a minimum of Germanic ancestry, now speak a language where the schwa (ə) sound (short or long, as the a and i in 'a bird') is omnipresent. This is true of Germanic languages, but also of non-Germanic languages like French. More interestingly, the ə sound is only found in northern and eastern French dialects, where the proportion of Germanic ancestry is higher in the population. It is absent from Breton language, which is not surprising as Brittany was a Celtic stronghold that received little Germanic influx. Yet, it is present in Welsh, as well as Scottish and Irish Gaelic, even though the schwa wasn't found in pre-Germanic Old Irish. I couldn't find information about Old Welsh and Old Scottish Gaelic, but I expect it to be the same (i.e. lacking the schwa). In the Iberian peninsula, the schwa is only present in Catalan, arguably the region with the highest Germanic ancestry in of Iberia and Southwest France combined. It is also found in Romanian, a region with even higher Germanic ancestry (similar to eastern parts of France like Lorraine or Burgundy). I don't think it is a mere coincidence that that schwa sound is found among population with Germanic genes. There could be something in Germanic DNA that help people pronounce this sound, perhaps a different bucal morphology that naturally shifts clearer vowels toward the centre of the mouth. Once the medieval elite (with higher Germanic ancestry) picked up that pronunciation it became standard and the common people had to imitate it, even if they weren't naturally suited for that kind of sound.
This makes me wonder whether the ease with which we learn a language (our natural affinity) could be related to the genes we inherited that related to speech production or how our brain processes grammar (as their must be interpersonal as well as inter-ethnic differences too). Just look at Black people in the US. Even though the majority of them have been in the country for two, three or even four centuries (the first African slaves arrived in 1619), many still speak English with a distinctive African-American accent (e.g. inability to pronounce th and the final g in -ing words), and, for that matter, also vocabulary and grammar. Black people also have a characteristic voice, even if they speak with a standard accent and grammar. That is partly because of the differences in mouth morphology. Not all do, but it is now well known thanks to modern DNA tests that some African Americans have considerable European ancestry, even if it is not always clear from the skin tone alone. It would be interesting to compare the DNA of African Americans whose voice (not just accent, but also tone, pitch, etc.) resembled most that of Whites, regardless of their social class and education. It might be a way to isolate the genes that influence mouth morphology and speech production.
If our genes influence our ability to produce sounds or to favour some sounds rather than others, it is only natural that languages evolve together with genetic drift, and that the influx of foreign DNA by invaders almost inevitably leads to changes in pronunciation (as with the adoption of the schwa in French and Gaelic).
What this study shows is that there is undeniably some underlying genetic process that make humans naturally think of sounds associated with specific meanings. English is full of onomatopoeic words that fit in that category. Even non-English speakers could virtually hear and guess the meaning in words like hiss, buzz, flush, snip, bang, crunch, cough, pop, crash, smack, slap, groan, moan... To come up with these words, our imagination had to convert environmental sounds into vocaliseable (did I just make up that word?) sounds, and from there into actual words. What sounds are easily vocaliseable, or how we prefer to render them may depend on our bucal morphology.
Morten Christiansen explains in the Cornell Chronicle that "these sound symbolic patterns show up again and again across the world, independent of the geographical dispersal of humans and independent of language lineage".
For example, in most languages, the word for “nose” is likely to include the sounds “neh” or the “oo” sound, as in “ooze.” The word for “tongue” is likely to have “l” (as in “langue” in French). “Leaf” is likely to include the sounds “b,” “p” or “l.” “Sand” will probably use the sound “s.” The words for “red” and “round” are likely to include the “r” sound. “It doesn’t mean all words have these sounds, but the relationship is much stronger than we’d expect by chance,” says Christiansen, noting that words for body parts in particular showed an unexpectedly high association between sounds and meaning.
"Perhaps these signals help nudge kids into acquiring language," he added: “Maybe it has something to do with the human mind or brain, our ways of interacting, or signals we use when we learn or process language. That's a key question for future research."
For example researchers have shown that words for small spiky objects in a variety of languages are likely to contain high-pitched sounds, while rounder shapes contain ‘ooo’ sounds, which is known as the ‘bouba/kiki’ effect.
The researchers considered the possibility that it could be the remnant of some form of "prehistoric protolanguage" that was once spoken by the earliest humans before the evolution of modern languages, but their own analysis suggests it's more likely that these similarities are rooted in human biology.
Analysis and comments
Personally, ever since I came across Richard Dawkins's concept of extended phenotype, I have wondered how much do our genes influence our language, including the choice and range of sounds, syntax, grammar, and so on. A phenotype is the expression of all the visible characteristics of DNA, such as morphology, development, biochemical or physiological processes. Our looks, temperament, immune reactions and so on our phenotype. The extended phenotype is everything that exists outside a life being's phenotype, but that also results from that being's gene expression. For example a bird's nest is the extended phenotype of a bird's DNA. The nest could not exist without the bird's DNA. Each species of bird makes its own distinctive kind of nest, and the characteristics of that nest are instinctive, coded in the DNA. In the same way, bird songs are unique to each species and coded in their DNA. You see where I am going with this. If a bird's vocalisation, including its frequency, pitch, but also the whole melody characteristic of a particular species is an expression of their DNA, why would it be any different for humans?
It is very likely that primitive human language evolved following the same instinctive sound patterns coded in our genes as with other species. The sound range, frequency and pitch may simply be determined by the shape of our mouth, throat, uvula and vocal cords. This may be why some ethnic groups, possessing different facial morphology could naturally produce some sounds that weren't found in other languages.
For example, all people with a minimum of Germanic ancestry, now speak a language where the schwa (ə) sound (short or long, as the a and i in 'a bird') is omnipresent. This is true of Germanic languages, but also of non-Germanic languages like French. More interestingly, the ə sound is only found in northern and eastern French dialects, where the proportion of Germanic ancestry is higher in the population. It is absent from Breton language, which is not surprising as Brittany was a Celtic stronghold that received little Germanic influx. Yet, it is present in Welsh, as well as Scottish and Irish Gaelic, even though the schwa wasn't found in pre-Germanic Old Irish. I couldn't find information about Old Welsh and Old Scottish Gaelic, but I expect it to be the same (i.e. lacking the schwa). In the Iberian peninsula, the schwa is only present in Catalan, arguably the region with the highest Germanic ancestry in of Iberia and Southwest France combined. It is also found in Romanian, a region with even higher Germanic ancestry (similar to eastern parts of France like Lorraine or Burgundy). I don't think it is a mere coincidence that that schwa sound is found among population with Germanic genes. There could be something in Germanic DNA that help people pronounce this sound, perhaps a different bucal morphology that naturally shifts clearer vowels toward the centre of the mouth. Once the medieval elite (with higher Germanic ancestry) picked up that pronunciation it became standard and the common people had to imitate it, even if they weren't naturally suited for that kind of sound.
This makes me wonder whether the ease with which we learn a language (our natural affinity) could be related to the genes we inherited that related to speech production or how our brain processes grammar (as their must be interpersonal as well as inter-ethnic differences too). Just look at Black people in the US. Even though the majority of them have been in the country for two, three or even four centuries (the first African slaves arrived in 1619), many still speak English with a distinctive African-American accent (e.g. inability to pronounce th and the final g in -ing words), and, for that matter, also vocabulary and grammar. Black people also have a characteristic voice, even if they speak with a standard accent and grammar. That is partly because of the differences in mouth morphology. Not all do, but it is now well known thanks to modern DNA tests that some African Americans have considerable European ancestry, even if it is not always clear from the skin tone alone. It would be interesting to compare the DNA of African Americans whose voice (not just accent, but also tone, pitch, etc.) resembled most that of Whites, regardless of their social class and education. It might be a way to isolate the genes that influence mouth morphology and speech production.
If our genes influence our ability to produce sounds or to favour some sounds rather than others, it is only natural that languages evolve together with genetic drift, and that the influx of foreign DNA by invaders almost inevitably leads to changes in pronunciation (as with the adoption of the schwa in French and Gaelic).
What this study shows is that there is undeniably some underlying genetic process that make humans naturally think of sounds associated with specific meanings. English is full of onomatopoeic words that fit in that category. Even non-English speakers could virtually hear and guess the meaning in words like hiss, buzz, flush, snip, bang, crunch, cough, pop, crash, smack, slap, groan, moan... To come up with these words, our imagination had to convert environmental sounds into vocaliseable (did I just make up that word?) sounds, and from there into actual words. What sounds are easily vocaliseable, or how we prefer to render them may depend on our bucal morphology.
