Whilst we tend to talk about the stories behind  stories on this channel, I thought I’d dive into  

something I thought was fascinating, based on  a paper written almost eight years ago now.  

And this is the discovery of ultra-conserved  words in our language, and what this means is  

that there are instances of words discovered  in separate languages that appear to be the  

same and which may have roots going back to  the last ice age, and so over 15,000 years old.  

So welcome to the story of the oldest words  of the Eurasians, and welcome to Crecganford.  

Language change over time, due to influences  such as migrations, war, or the coming together  

of cultures, and as such that the more  time goes on the more language changes,  

and the more difficult it is to trace a path back  to our oldest languages and our oldest words.  

To some extent we are lucky with the  Proto-Indo Europeans, when we have a  

culture with such a strong impact, that many who  encountered them wanted to interact with them,  

and that was a key in how the  language spread so successfully.  

And this has led us to understanding which stories  derived from the time this language was spoken,  

due to being able to track the language  similarities across the Indo-European geographic  

landscape. However, going back in time past the  Proto-Indo Europeans is incredibly challenging,  

as language and words older than 5,000 to 9,000  years are difficult to identify due to all the  

influences. Although that has not stopped some  researchers trying to understand if there was a  

common language before the Proto-Indo Europeans.  And this research into whether there was an  

older language was approached using math, or more  precisely statistics with quantitative modelling,  

which in brief is a method that  allows us to understand whether  

a difference between two things is significant.  So in very simplified terms, for this research,  

are two words statistically similar or not. And if  this modelling is applied to some ultra-conserved  

words, which, simply put, are words with common  properties between two different sources,  

then it may show that this commonality is actually  the result of a deeper relationship between the  

words, in effect, that they have a common  ancestor. So why do this? Why do researchers  

think that there was a common ancestor  language on the Eurasian continents before  

the Proto-Indo Europeans? Well, most language  evolution theory is based on cognate words,  

so words that show a structural evolution  into others through consistent adaptations  

by a culture or region. But such cognates and  their evolution can suffer from phonetic erosion,  

and over a period of time such as 9,000 years,  then that needs to be considered, as such erosion  

could change words completely. This makes the  statistical verification of any changes form a  

cognate perspective difficult to validate,  as the results would show a probability of  

being no better than a chance change, meaning  we can not be sure of the evolution of a  

word using cognates if it is over 9,000 years old.  However, these cognate challenges can be mitigated  

if the words sound the same, which is rare in  many words, with the average word being replaced  

by a new non-cognate word every 2,000 to  4,000 years, but some words, such as numerals,  

pronouns, and some adverbs, are replaced more  slowly because of their frequent use, and may  

only change once every 10,000 to 20,000 years.  Now the advantage of these later words is that  

they can be predicted independent of their sound,  and this is by using the frequency a word is used  

in everyday speech in Indo-European languages,  along with its part in a sentence construction,  

and then it can be predicted how rapidly it  would evolve, as frequency and there part in  

sentence construction is consistent, irrespective  of language, across the Eurasia, and even  

into North America. And this finding shows that  humans use a core set of words in everyday speech  

at the same frequency and show the possibility  that irrespective of the language or location,  

each type of word evolve at similar rates  across these. Then, using this, it provides  

us with a statistical framework for predicting  words likely to show a deep ancestry amongst  

languages whose relationships extend well beyond  9,000 years. And the result of research using this  

quantitative modelling is that the following  words we use today, were probably in use,  

in a similar form just after the last ice  age, around 14,500 to 15,000 years ago.  

There are the number words, but also these common  words; “thou”, “I”, “not”, ”that”, ”we”, ”who”, ”this”, ”what”,  

and ”ye”. Which we can link cognitively as well.  Then other frequently used words such as: “To say”,  

“day”, and “to know”. But also some infrequently used  words come back as statistically likely to be  

from the post-ice age period, including: “bark”,  “ashes”, “worm”. Now one of these may infer some  

significant importance in these words, and  this is because the word worm fascinates me,  

as it can also mean snake or serpent, and so  could be important as this could be related  

to serpent and the common dragon myths that  the Proto-Indo Europeans have. And whilst it is  

next to impossible to know for sure, to think that  the dragon myth may have been an important part  

of culture post-ice age, suggests that elements  of the creation myth maybe as 15,000 years old.  

But that is pure supposition at the moment, but  a fascinating one from my interest in creation  

myths. So what can we conclude from this study?  Well the fact is that it does seem probable  

that words we use today have originated from a  post-Ice Age language on the Eurasian continent,  

but where? Well the data from this research allows  us to build up a phylogenetic tree of Eurasiatic  

languages to see when language spread from a  source around 15,000 years ago. And at this time  

it looks that around the North West of  India and Pakistan, the language started  

to change to develop the first split known  as Proto-Dradivian around 15,000 years ago.  

Then at 12,500 years ago another branch,  the Proto-Kartvelian came into existence,  

followed by splits into Proto-Indo European  and Proto-Uralic at 12,000 years ago.  

The other stem from this split, at 10,500  years ago enough change happened to develop the  

Proto-Altaric branch of language, and the final  branch discovered, came from here as language  

moved across what is now Russia and across  into North America, around 10,000 years ago,  

which would coincide with a land bridge  of ice made by the Younger Dryas event  

about 11,500 years ago. And this developed  branches called the Proto-Inuit Yupik language,  

and the other Proto-Chukchi-Kamchatkan. And  from this we can infer that the language  

started from the area we know as Kazakhstan  today. And that is the story behind some of the  

oldest words we know today. I hope you found that  interesting, and I’ll put a link to a useful paper  

if you wish to read more about this in the  description below. And so until the next video,  

please stay safe, stay well.  And this was Crecganford.