Recent research landscape shown that substantial resources have been put into the development of large language models (LLMs) and explainable artificial intelligence (XAI). However, our insights into human intelligence, especially regarding language comprehension, have not benefited proportionately despite the extensive research efforts on fields that are such closely related. In this article, I aim to establish a novel structure of language understanding that provides philosophers and psychologists a way to directly benefit from the breakthroughs in current LLM and XAI research. Competing prior approaches generally face two primary challenges: either falling short of explaining the relationship among language, words, word definitions, meanings, and linguistic theories, or they do not address the concerns of whether word representations can accurately reflect human understandings of word meanings. In contrast, this approach addresses both problems by conceptualizes language as a high-dimensional vector space and systematically defining its constituent components. As a result, we show that this perspective leads us to research directions that can maximize the acceleration of the improvements in science.