News Release

The world's languages may be so similar because of how humans talk about language

Peer-Reviewed Publication

University of Helsinki

The languages of the world are all unique, but also share important similarities. These mostly lie in the grammatical elements each of the approximately 7000 human languages contain, the word parts and rules that speakers can use to build a sentence. Traditionally, linguists either assume that the explanation for these similarities is that all people are born with a blueprint for these grammatical categories in their genetics or that they emerge from other cognitive capacities (for example, all people have an understanding of time so human languages can develop past tense or future tense).

A problem with these accounts is that they can only explain part of the grammatical similarities of the world's languages and also that they do not explain where grammar came from.

In a new paper recently published in Frontiers in Communication, linguists Stef Spronck and Daniela Casartelli from the University of Helsinki propose a radical new theory: these grammatical similarities are due to the way in which humans talk about language. The authors noted that in many languages sentences that reflect people's speech or thought, known as 'reported speech', can develop new meanings that closely resemble grammatical categories. This means that the sentence 'He said: "I will go"' in some languages can become the main way to express meanings such as 'He was about to go', 'He might go', 'As for him, he will go'. Each of these interpretations have no clear connection with the meaning of reported speech, but use a sentence structure that derives from reported speech. The meanings associated with these non-speech interpretations of reported speech correspond to common grammatical categories in the languages of the world, which linguists call 'aspect', 'modality', 'topic' and others.

Similar examples include instances in which the ‘saying’ sentence on the left below is used as the main or only strategy to express the meaning on the right:

-    She said ‘I will go’ means She WANTS to go
-    He believes saying they will go is the only way to express He believes THAT they will go (so the word ‘say’ is interpreted as the complementiser ‘that’)
-    It says ‘I will go’ means It IS ABOUT TO go
-    We say ‘You will go’ is the only way to express We MAKE you go
-    You say ‘we will go’ means You FEAR we will go  
-    Saying me, I go is the only way to express AS FOR ME, I go

Dr Stef Spronck, one of the authors of the study, says: "Observations about expressions of saying being used to express a wide variety of meanings (including those above and others) have been made in the linguistic literature before for individual languages and areas, but our study shows that they in fact occur all around the world in languages that have not been in contact. We propose that such observations can be explained by the hypothesis that saying clauses (reported speech) are an important source for a wide, but also quite regular range of meanings that constitute core parts of grammar and the meaning of some verbs and thus have played a central role in the constitution of language."

"Humans talk about other people's thoughts and statements all the time, from the moment we first learn to speak. It determines our cultures, the way we see the world and who we trust. A phenomenon that is so fundamental to human existence likely leaves its trace on languages and our study shows that this goes far beyond simple sentences of reported speech. We propose that in the evolution of language talking about language was a way of forming some of the first complex language structures and that from these structures new types of grammar could develop. This would explain why we see reported speech with so many different functions in our world-wide sample, but at the same time also find the same functions over and over again. Many of these new 'non-speech' functions could be seen as a type of grammatical category. Our hypothesis is not meant to replace traditional cognitive explanations of grammar, but provides a new story for the emergence of grammatical categories, particularly those that are traditionally more difficult to explain."


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.