This paper proposes a Tungusic Cluster-Based Phonological Reconstruction Method, a novel framework for recovering the original sounds of ancient Northeast Asian ethnonyms and toponyms that were historically transmitted only through Chinese logographic transcription. Because Chinese characters could represent only simplified CV syllables and lacked the ability to encode Tungusic final consonants, nasal codas, or cluster-rich phonotactics, nearly all names of Buyeo, Goguryeo, Sushen, Yilou, Mohe, Jurchen, Nanai, and early Japanese strata were distorted, normalized, or semantically overwritten in Chinese historiography. The proposed method treats the Chinese characters not as semantic indicators, but as imperfect phonographic approximations, and reconstructs original forms by identifying sound clusters (phonological groups) that regularly interchange across Tungusic, Mongolic, and early Japonic dialects. These clusters include: n / y / r / l, t / d / s, f / b / p / m / w, k / h / kh / x / g, and final -n / -m / -ŋ,along with flexible vowel alternations i/e – a – o/u – ŋ. Using this method, Sino-recorded regions such as Lelang 樂浪, Xuantu 玄菟, Lintun 臨屯, and Zhenfan 真藩 are naturally reconstructed as the Tungusic ethnonyms Nanai, Golid, Yakut, and Sibe, respectively. The analysis extends further to Japanese ancient names recorded in the Kojiki and Nihon Shoki, where elements like –mi / –mimi / –mikoto correspond to the Manchu verbal suffix –mbi. Remarkably, the name “Nippon / Jippon” (日本) itself aligns with the same Tungusic cluster as Sufēn (卒本), Zhenfan (真藩), Shiwei (室韋), and Sibe (錫伯)—suggesting that the ethnonym “Japan” may derive from the same Northeast Asian tribal-name pattern meaning “our land / our people.” This approach provides a coherent phonological framework for disentangling the multilayered linguistic substrate of the Japanese archipelago and for reinterpreting East Asian ancient records free from the distortions imposed by Chinese logographic conventions.
Konno Tetsuo (Mon,) studied this question.