Language consists of multiple elements that interact through speaker–listener communication, producing linguistic phenomena that exemplify a complex system in which global patterns emerge from local interactions. Photo: TUCHONG

The history of linguistics shows that both generative grammar and cognitive linguistics emerged in close connection with advances in other disciplines. In recent years, a methodological shift in scientific research from “reductionism” to “holism” has increasingly highlighted the significance of complex systems science. Linguistics should also seize this opportunity to broaden its research approaches.

Language as a complex system

Traditional linguistics often views language as a static system governed by fixed grammatical rules. Yet vocabulary and expression are far from immutable—they continually evolve alongside social, cultural, and technological change. Language is composed of multiple elements—vocabulary, grammar, phonology, and semantics—that interact through communication between speaker and listener to generate linguistic phenomena. This interaction exemplifies a defining feature of complex systems: system-level patterns arising from local interactions among individual components. Language exhibits several key properties that reveal its dynamic and interconnected nature.

Openness of language: Language systems interact with other systems, including the objective world and human cognition. Social environments, historical contexts, and technological progress all exert influence on the evolution and development of language systems through cultural communication and political interaction.

Rule-based nature of language: The internal workings of language systems are governed by rules. Human communication is made possible through shared grammatical conventions. Language embodies four types of natural laws: deductive, statistical, generative, and functional.

Asymmetry of language: Vocabulary is asymmetrical in terms of both meaning and grammatical function. Likewise, the mechanisms behind grammatical markers also exhibit asymmetry. At any given time, no language possesses all possible grammatical devices, nor is any single grammatical device indispensable.

Adaptability of language: Language adjusts its grammar, vocabulary, and forms of expression in response to changes in social culture, technology, and media. In everyday use, it does not rigidly adhere to fixed rules but flexibly adapts to specific communicative needs.

Evolutionary nature of language: Given their adaptability, language systems are inherently evolutionary. This manifests in two ways: obsolete grammatical forms are phased out, while new grammatical devices emerge.

Complex systems science enriching linguistic research perspectives

Complex systems science not only helps illuminate how language spreads within social networks but also offers new methods for understanding linguistic phenomena.

First, “network theory” derived from complex systems provides new insight into the transmission and diffusion of language. Traditional linguistics often describes language change as a linear process spreading from the center to the periphery. Network theory, however, reveals its nonlinear and complex nature. Within social networks, language spreads not through a single channel but across multilayered, multidimensional webs of interaction. With the rise of social media, for example, neologisms, buzzwords, and even changes in grammatical structures often spread through user interaction, spontaneous imitation, and diffusion—rather than through traditional linguistic inheritance—rapidly achieving widespread acceptance.

Second, the theories of self-organization and emergence inspired by complex systems provide new insights into natural language processing. Self-organization refers to the spontaneous formation of a system-wide order through local interactions among the system’s components without external intervention. Emergence refers to a system exhibiting properties or behaviors not present in its individual components, arising from interactions among those components. In the field of artificial intelligence, large language models (LLMs) are trained on massive datasets to simulate the structure, grammar, and semantic relationships of human language. Through interactions among their learning units, these models can generate text outputs that conform to human linguistic conventions without explicit programming.

Third, complex systems science provides a fresh perspective on the language behavior of bilingual speakers, whose choice and frequency of language switching depend not only on their language proficiency but also on various factors such as social contexts, the linguistic background of their interlocutors, and the purpose of communication. Whether at home, in the workplace, or in public spaces, the language behavior of bilingual speakers displays a high degree of flexibility and adaptability, manifesting the dynamic changes and self-adaptive capabilities characteristic of complex systems.

Liu Feng is an associate professor from the School of Foreign Languages at Jishou University.