From language to meteorology: kinesis in weather events and weather verbs across Sinitic languages

The impact of humans on the environment and vice-versa presents a significant challenge to scholarly pursuits. Traditional Ecological Knowledge (TEK) offers a valuable alternative by leveraging centuries or millennia of accumulated experience. While TEK integration has proven beneficial in various research areas, especially those involving languages with primarily oral traditions, large-scale text-based analyses of Sinitic languages remain under-explored. Previous linguistic studies on weather expressions have focused on linguistic relativism and typology, often centering on debates regarding the number of snow terms in Inuit, without sufficient integration of meteorological knowledge. This study aims to bridge this gap by analyzing weather verb selection in Sinitic languages, a family exhibiting significant diversity across climates and possessing over 3000 years of continuous textual documentation. The study leverages the concept of 'kinesis,' encompassing motion, force, and energy, to connect meteorological events (ideally modeled as movements of objects subject to external forces) with their linguistic encoding. By integrating various definitions of kinesis from Aristotle to modern physics, the study hypothesizes that the perception of physical properties (mass and speed/acceleration) underpins the linguistic encoding of weather events.

Existing typological research on weather and language, exemplified by Eriksen et al.'s (2010, 2012) framework, offers a dual module view classifying weather constructions formally (predicate, argument, argument-predicate types) and semantically (dynamic/static). While this framework accounts for formal encoding variety, it doesn't explicitly consider meteorological knowledge. Studies on Sinitic weather expressions, such as Dong's (2018, 2019) work, reveal the diversity of verbs used, challenging previous typological predictions. The Typology of Meteorological Events (TyME, Dong et al., 2020a) classifies events using [±Process] and [Material] features, but lacks the integration of meteorological knowledge. Research on the noun-verb dichotomy explores the cognitive basis of these categories. Hopper and Thompson (1980, 1985) introduced kinesis as a cognitive basis for verbs, emphasizing movement, visibility, and effectiveness. This study integrates these perspectives by hypothesizing that observable elements of kinesis play a central role in weather word encoding.

This study adopts a methodology inspired by Wilhelm Dilthey's notion of human sciences, integrating empirical evidence from a structured context encompassing human experience. To address Galton's problem (avoiding confounding factors while examining complex human behavior), the study leverages the well-documented typology of Sinitic languages, focusing on regional variations that deviate from general north-south dichotomies. This approach minimizes the influence of socio-cultural factors, allowing for a focus on the correlation between meteorological events and linguistic variations. An ontology-lexicon interface approach links traditional and scientific knowledge systems, providing a structured context for interpretation and hypothesis verification. The study analyzes weather expressions from 221 Sinitic languages/dialects, focusing on nine common weather phenomena. The transitivity of selected verbs is analyzed in relation to the kinesis of their corresponding weather events. A perception experiment using novel weather nouns investigates the role of mass and speed in verb selection, employing ordinal logistic regression to analyze the data. Finally, geographical distributions of weather expressions are compared with meteorological maps to identify correlations between linguistic choices and weather patterns.

Analysis of weather expressions across 221 Sinitic languages/dialects reveals a correlation between verb transitivity and the kinesis (mass and speed) of weather events. Precipitation events (rain, snow, hail) predominantly use low-transitivity verbs ('to fall'), while other phenomena show varied patterns. High-transitivity verbs are more common for weather events with larger substances and/or faster processes (frost, lightning, strong wind, heavy rain). The experiment on verb perception confirms that native Sinitic speakers perceive differences in mass and speed for various verbs, supporting the link between kinesis and verb selection. Geographical analysis shows that regional variations in verb selection often correspond to variations in meteorological patterns, particularly rainfall intensity and frost damage. For example, the use of the verb 'luò' ('to fall, to drop') for rain in areas north of the Yangtze River, contrary to the typical northern pattern of using 'xià', correlates with areas having over 1000 mm of yearly rainfall. Similarly, the use of high-transitivity verbs for frost correlates with areas experiencing significant frost damage. The study of hail nouns shows that more recent forms correlate with areas experiencing more hail damage. Finally, the analysis of the word 'mái' (smog) in Archaic versus Modern Chinese demonstrates how a change in the referred weather phenomenon (dust storm vs. smog) correlates with a shift in verb choice reflecting the difference in kinetic energy.

The findings strongly support the hypothesis that kinesis, particularly as reflected in the mass and speed of weather events, plays a crucial role in shaping the linguistic encoding of these events in Sinitic languages. The consistent correlation between verb transitivity and the perceived kinesis across various weather phenomena, supported by both linguistic analysis and experimental data, highlights the cognitive link between physical experience and linguistic representation. The geographical analysis further strengthens this conclusion by demonstrating the correlation between regional linguistic variations and meteorological patterns, controlling for historical factors. The kinetic energy model proposed provides a unifying framework for understanding the selection of verbs in weather expressions, suggesting that events with higher kinetic energy are more likely to be associated with high-transitivity verbs. This study contributes to our understanding of the interplay between language, cognition, and environment, offering a novel perspective on how human experience shapes linguistic structures.

This study demonstrates the significant role of kinesis in shaping the linguistic encoding of meteorological events across Sinitic languages. The consistent correlation between verb transitivity, perceived kinesis (mass and speed), and meteorological patterns strongly supports the proposed kinetic energy model. This interdisciplinary approach offers a valuable framework for understanding the dynamic relationship between language, cognition, and the environment. Future research could explore micro-scale meteorological events and extend this model to other language families.

While the study utilizes a large dataset of Sinitic languages, the analysis focuses primarily on meso-scale to large-scale meteorological events. Further investigation into micro-scale events is needed to fully assess the generalizability of the findings. The geographical analysis relies on existing meteorological and linguistic data, which might have limitations in spatial resolution and data density. Although the study attempts to mitigate confounding factors, the complexity of historical and cultural influences on language evolution cannot be entirely ruled out.