Linguistic diversity as a testing ground for theories of language

Every language is a variation on the universal theme of human communication.
EFL explores this diversity as a living laboratory for testing, refining, and transcending major linguistic theories. Far from reducing languages to exceptions or anomalies, this program considers diversity to be a prerequisite for scientific rigor.

Why is linguistic diversity crucial to the science of language?

Too many dominant linguistic theories have been developed based on a limited corpus of so-called “major” languages—often European, written, and standardized. This bias impoverishes our understanding of what human language is.

EFL takes the opposite approach by asking a fundamental question: how different can human languages be while still remaining languages? In other words, what are the structural, functional, and cognitive limits of language as a universal capacity?

Scientific questions on the topic

What are the invariants and variables in the languages of the world?

• Are there universal syntactic, phonological, or semantic constants?
• Which grammatical systems challenge dominant theories (e.g., ergativity, polysynthesis, verb-less languages)?

How far can linguistic diversity go without leaving the field of language?

• Study of rare, marginal, or marginalized structures.
• Exploration of cognitive and perceptual limits (tonal, sign, whistled languages, etc.).

How can theories be empirically tested on the basis of real diversity?

• Integration of data from poorly documented, minority, or endangered languages.
• Computational modeling of typologically extreme phenomena.

What is the place of the word in languages?

• In some languages, the word is not a central unit (e.g., polysynthetic languages).
• This challenges the classical conception of the word as the universal basis of linguistic analysis.

How do language structures interact with other forms of cognition?

• Grammar and memory, categorization, perception of time or space.

Methods used

EFL relies on a combination of empirical and analytical methods:

• Linguistic typology: comparative study of hundreds of languages around the world.
• Field linguistics: documentation of poorly described languages with native speakers.
• Multilingual corpora and structured databases (e.g., typological atlases).
• Formal and computational modeling: simulations of linguistic variations and transformations.
• Experimental approaches: perception and processing of rare structures.

Epistemological and societal issues

• Science open to complexity: it is not a question of explaining a language, but language in its possible forms.
• Decolonization of linguistic knowledge: active inclusion of minority languages in theoretical construction.
• Preservation and promotion of global linguistic heritage.
• Contribution to unbiased linguistic AI: training models on a diversity of grammatical structures avoids performance and design biases.

Partnerships

– Collaboration between laboratories specializing in world languages: LaCiTO, Llacan, SeDyL, CRLAO.

– Strong support from the CNRS, INALCO, and funded field projects.

– Integration with other areas of the project to combine diversity, acquisition, aging, and artificial intelligence.

In summary

The diversity of human languages is more than just a subject of study: it is a critical tool for advancing the science of language. By exploring it in all its breadth, EFL aims to build theories that are more robust, more inclusive, and more faithful to the reality of linguistic humanity.