Japanese Language

Japanese (日本語, Nihongo) is the main language of Japan and the only country where it is the national language, with roughly 123 million speakers worldwide. It is the principal member of the Japonic family, which also includes the Ryukyuan languages. Its relationship to other languages remains unresolved: it is usually treated as a language family of its own, and no proposed relatives have gained broad acceptance. Its sound system is comparatively simple, with five vowels and a mora-based rhythm, but its writing system is one of the most complex in use, combining Chinese characters (kanji) with two syllabaries (hiragana and katakana) and, in modern practice, Latin letters.

The language is important for linguistics because it displays features that are rare or distinctive: an agglutinative morphology, a head-final sentence structure with frequent omission of subjects and objects, and an elaborate honorific system. Its history also shows deep and lasting contact with Chinese, which shaped its vocabulary, its writing, and its literary tradition. Japanese is also central to debates in sociolinguistics, since regional dialects, gendered speech, and standardization have all been closely tied to national identity.

This entry first covers the language's classification and origins, then its historical stages and its sound, grammatical, and lexical structure. It next examines the writing system in detail, followed by honorifics, dialects, and standard Japanese. The later sections address endangered Ryukyuan varieties, social variation, acquisition and education, global use, computational processing, and the history of research on the language.

Classification and origins

The Japonic family

Japanese belongs to the Japonic family, which includes the Ryukyuan languages of the Ryukyu Islands. According to the currently widely accepted phylogenetic classification, the Japonic family consists of three branches, namely Ryukyuan, Japanese, and Hachijo, although the phylogenetic position of Hachijo is debatable. The Ryukyuan branch is split into two groups: the Northern group comprises the Amami and Okinawa languages, whereas the Southern group comprises the Miyako, Yaeyama, and Yonaguni languages. Some classifications treat Hachijō as a third branch, while others regard it as a highly divergent variety of Eastern Japanese, so its status depends on which criteria a classification adopts.

Language isolate status and external proposals

Japanese is often described as a language isolate because no genetic relationship to another family has been demonstrated. Linguists traditionally considered the Japonic languages to belong to an independent family, and until Ryukyuan and eventually Hachijō were recognized as separate languages within a Japonic family, Japanese was considered a language isolate. Several external proposals have been made. The Altaic hypothesis rested on structural similarities such as agglutinative morphology, predominantly subject object verb order and certain parallels in phonology and verbal structure, but it is not accepted by most specialists. Other proposals have linked Japonic to Austronesian or Kra–Dai, and Paul K. Benedict maintained that Japanese is a Para-Austronesian language, proposing an Austro-Tai–Japanese grouping, but this grouping is not widely accepted by linguists. A proposed Korean connection has been pointed to by toponyms in today's southern Korea, which may be from the ancient isolated Gaya language or other scarcely attested languages. Overall, there are numerous hypotheses, none of which is generally accepted.

Proto-Japonic and its dating

The common ancestor of the family is called Proto-Japonic, a reconstruction rather than a attested language. It was reconstructed by using a combination of internal reconstruction from Old Japanese and by applying the comparative method to Old Japanese (both the central variety of the Nara area and Eastern Old Japanese dialects) and the Ryukyuan languages. Its date is less certain. Because Old Japanese displays several innovations that are not shared with Ryukyuan, the two branches must have separated before the 7th century. Earlier separation dates are harder to fix, and the sources disagree on how far back Proto-Japonic can be pushed. Some scholars argue that the split between Proto-Ryukyuan and the mainland line may have occurred around the Yayoi–Kofun transition, while others place it later.

Arrival on the islands

The arrival of Japonic in the archipelago is linked by many researchers to the spread of rice farming. Most historical linguists agree that Japonic first arrived in the Japanese archipelago during the Yayoi period. The Yayoi period is itself dated by archaeology: the introduction of rice agriculture into the Genkai Bay area of North Kyushu from around 1000 BC marks the beginning of the Yayoi period. One linguistic model places the first entry of Proto-Japonic on the Korean Peninsula, with the Proto-Japonic language first entering the Korean Peninsula and spreading to the Japanese archipelago at the beginning of the Yayoi period, around the ninth century BC, but this dating is contested.

Population genetics offers a parallel framework. A long-standing hypothesis posits that mainland Japanese populations derive dual ancestry from indigenous Jomon hunter-gatherer-fishers and succeeding Yayoi farmers. Later genomic work has added further layers, and one study identified a later influx of East Asian ancestry during the imperial Kofun period. Linguistic models often connect the Yayoi migrants with the language, but the match between genes, archaeology, and languages is imperfect. Researchers also disagree about the route, with one hypothesis proposing that Japonic originated in Shandong and was transmitted to Japan by migrating rice farmers following a route crossing the Bohai Sea to Liaodong. Others argue that the link between language spread and rice farming is weaker. A key weakness is evidence: written evidence appears relatively late, while the Japanese archipelago experienced several prehistoric population movements and long periods of interaction with the Korean peninsula and continental East Asia. Moreover, despite considerable hypothesizing about prehistoric language, the spread of historic languages and dialects through the islands is more accessible but difficult to relate to prehistory. The arrival of Japonic also does not necessarily mean that earlier languages disappeared at once, since after the arrival of Japonic, all Jōmon languages were not necessarily immediately replaced by the new language.

From Old Japanese to Modern Japanese

Old Japanese

Old Japanese is known only from writing produced in the Nara era, which ran from 710 to 794. It is the oldest attested stage of the Japanese language, recorded in documents from the Nara period (8th century). Most of the surviving material is verse. The texts were written in man'yōgana, a system that employs Chinese characters as syllabograms or (occasionally) logograms. The grammar already had features that later forms would partly abandon. Unlike later forms of Japanese, Old Japanese adjectives can be used uninflected to modify following nouns. Old Japanese verbs have a rich system of tense and aspect suffixes.

The sound system was also more finely divided. The earliest attested Old Japanese distinguished 88 syllables. Some of these contrasts were grouped into classes. OJ kept distinct so-called kô-rui (type A) and otsu-rui (type B) syllables, which merged in the transition to Early Middle Japanese. One reconstruction goes further and credits the language with about eight vowel distinctions. In that model, two sounds which later merged as a single e, two which merged as i, and two that merged as o. The exact pronunciation of these distinctions is still disputed, because the phonetic realization of these distinctions is uncertain.

Early and Late Middle Japanese

The Heian era, which ran from 794 to 1185, marks Early Middle Japanese. This was the period when the Chinese language had its most significant influence on the Japanese language. The sound system was reorganized during this time. Distinctions that appear in early texts were lost in later sources. For example, the distinction between mo1 and mo2 is found in the Kojiki (712), but had merged in later sources. Verb and adjective morphology changed in parallel, and the reorganization of verb conjugation is treated in morphology.

Late Middle Japanese covers 1185–1600. In this period, changes brought it closer to the modern language and the first appearance of European loanwords. The first European words entered in this phase, which is also when the language's sound system moved closer to the modern pattern.

Early Modern and Modern Japanese

The Edo period lasted from 1600 or 1603 and 1868. During these centuries, the standard dialect moved from that of the Kansai region to that of the Edo (modern Tokyo) region. Contact with Portuguese and Dutch traders added vocabulary as well. Contact with Portuguese and Dutch traders introduced loanwords such as パン (pan, bread) and ガラス (garasu, glass).

Modern Japanese is usually dated from the Meiji era, which marks the beginning of the Meiji period on October 23 under the reign of Emperor Meiji. The inflow of foreign vocabulary accelerated after the country's isolation ended. Following the end in 1853 of Japan's self-imposed isolation, the flow of loanwords from European languages increased significantly. English loanwords in particular have become frequent, and Japanese words from English roots have proliferated. Under the new government, the Tokyo-based Hyōjungo was promoted as the national standard taught in schools and used in public life.

Sound change and its consequences

The overall trend was toward simplification. Earlier syllable distinctions were reduced over time, and the written corpus itself shows this process. The syllable inventory in Table represents a system in change and the gradual break-down of the distinctions is evident through the OJ text corpus. The consequences reached the writing system as well. At its inception the kana orthography was syllabic since there was apparently no distinction between long and short syllables in Old Japanese (OJ). Subsequent developments introduced such a distinction however, and today the kana orthography would be better termed 'moraic.' This link between sound change and script is discussed further in writing-system and phoneme.

Chinese influence and the arrival of writing

Before contact with China, the Japanese archipelago had no script of its own. The first Chinese characters arrived largely as objects. Chinese characters first came to Japan on official seals, letters, swords, coins, mirrors, and other decorative items imported from China. The earliest known instance of such an import was the King of Na gold seal given by Emperor Guangwu of Han to a Yamato emissary in 57 AD. Chinese coins from the first century AD have been found in Yayoi-period archaeological sites, but the Japanese of that era probably had no comprehension of the script. The date of the King of Na seal is traditional, and the route by which writing entered is less certain. Between the 3rd and 6th centuries AD, cultural achievements from China's Han, Wei, Jin, and Northern and Southern Dynasties periods were introduced to Japan via the Korean Peninsula. The best-known textual account is the story of Wani, a scholar sent from the Korean kingdom of Baekje. According to the Nihon Shoki and Kojiki, a semi-legendary scholar called Wani was dispatched to Japan by the Kingdom of Baekje during the reign of Emperor Ōjin in the early fifth century, bringing with him knowledge of Confucianism and Chinese characters. Physical evidence is sparser. The Inariyama Sword, excavated in 1968, carries a gold-inlaid inscription, and its date is usually given as around 471 CE. Dating therefore rests on a mix of chronicles, imported objects, and inscriptions, and the scholarly estimates differ.

Writing in Japan was at first almost entirely Chinese. The earliest Japanese documents were probably written by bilingual Chinese or Korean officials employed at the Yamato court. The two oldest chronicles, the Kojiki and the Nihon Shoki, are written entirely in classical Chinese. In this sense classical Chinese functioned as Japan's first literary language, and the spoken Japanese of the court was recorded only indirectly. Official documents and education followed the Chinese model. During the Nara period (710-794), Japan established an educational system and examination system, and learning and using Chinese characters became an important subject and skill, producing a group of Japanese people who were proficient in Chinese characters.

Reading Chinese in a Japanese way led to kanbun, the Chinese text annotated for Japanese reading. Scribes developed an annotation and reading system, called Kundoku, by which they translated classical Chinese into Japanese. By the Nara and Heian periods, these reading practices used glosses and diacritical marks to render Sinitic texts with Japanese word order, particles, and inflections. Reversal marks placed beside characters indicate that a passage must be read out of the original order, which converts the Chinese SVO sentence into a Japanese SOV pattern. Yomigana, hiragana placed beside the characters, indicate the reading in Japanese. Kanbun thus served as a bridge: it kept the Chinese text intact while making it readable as Japanese, and it remains part of Japanese education today.

Chinese vocabulary entered Japanese through the same channels. The state borrowed heavily from the continent: the Japanese government also borrowed from China's legal system and introduced many Chinese character vocabulary to express legal terminology. Buddhism, transmitted through Chinese translations of scriptures, added a religious and scholarly layer, and adopting these Chinese characters in the Japanese language led to adopting various Chinese words and vocabulary that were adapted to fit the Japanese pronunciation. The result was a learned, formal register in which Sino-Japanese words sat beside native vocabulary, a pattern discussed further in the vocabulary section.

The main structural problem lay in the mismatch between the two languages. Chinese characters carried meaning and, roughly, one syllable of Middle Chinese sound, while Japanese was a polysyllabic, agglutinative language whose grammar depended on particles and inflectional endings that Chinese script had no natural way to express. Many accounts treat this gap as the main pressure behind the invention of phonetic writing. One solution was man'yōgana, Chinese characters used as loangraphs during the Nara period for phonetic transcription using their Middle Chinese pronunciations. The man'yōgana was composed of approximately 970 characters that were used to represent the ~90 different syllables used in 8th-century Japanese. This multiplicity made reading difficult. Given the large number of homophones and the fact that man'yōgana was combined with kun writing, it was almost impossible to establish a single correct reading of a text. The system was nonetheless used in verse, and this system was already in use in the verse parts of the Kojiki (712) and the Nihon Shoki (720).

The later kana grew out of this experiment. Around 650 CE, a writing system called man'yōgana, used in the ancient poetry anthology Man'yōshū, evolved to use a number of Chinese characters for their sound rather than their meaning. Man'yōgana written in cursive style evolved into hiragana, while katakana also evolved from parts of man'yōgana by Buddhist monks who wanted to simplify it even further. The details of these scripts are treated in the kana section.

Phonology: sounds, syllables, and rhythm

Standard Japanese, based on the Tokyo variety, has five vowel phonemes, /a i u e o/, and each can be short or long. Length is contrastive, so the minimal pair 十 tō "ten" and 戸 to "door" differs only in how long the vowel is held. Likewise, ojisan "uncle" and ojiisan "grandfather" differ only in vowel length. Short /i/ and /u/ are often devoiced between voiceless consonants or at the end of an utterance, as in desu. This devoicing is a phonetic process that does not create new contrasts.

Most syllables are open, with the shape CV, and many consist of a vowel alone. Native words avoid consonant clusters and do not allow most consonants at the end of a syllable. The main exceptions are the moraic nasal /N/ (written ん) and the first half of a geminate obstruent (written っ, the sokuon), which lengthens the following consonant, as in kitte "stamp" versus kite "come". Palatalized sounds such as kya are written with a small ya, yu, or yo (拗音, yōon). They form a single syllable but take two kana. Because of these patterns, the syllable is a poor unit for counting sound in Japanese.

The basic timing unit is the mora (拍, haku). The word 東京 Tōkyō has two syllables but four morae: to-u-kyo-u. Long vowels, ん, and っ each occupy one mora. This mora-based rhythm explains why haiku are measured in on (音), with lines of five, seven, and five morae, rather than in syllables. It also underlies many word games and puns that depend on counting sound units. The kana chart reflects the same principle: most characters correspond to one mora, which is why the syllabary works so well for writing.

Standard Japanese has a lexical pitch accent rather than a stress accent. In Tokyo speech, an accented word has a single drop from high to low pitch after a particular mora, while unaccented words stay low and then rise after the first mora. The pitch pattern can distinguish words: 雨 ame "rain" has an accent on its first mora and falls, while 飴 ame "candy" is unaccented and rises. Accent systems differ across regions. The Kyoto-Osaka (Kansai) system organizes its contrasts differently from Tokyo, and several Kyushu varieties have their own patterns. Accent is therefore a point where dialects diverge clearly, and it is also a useful feature for distinguishing regional speech in media.

Japanese phonotactics restrict word-initial clusters and most final consonants, so loanwords are adjusted to fit native patterns. Usually a vowel is inserted to break up a cluster or to close a syllable. English strike becomes sutoraiku, glass becomes garasu, and Christmas becomes kurisumasu. Final consonants generally receive an extra vowel, as in keeki from "cake". Since the twentieth century, katakana combinations such as ティ (ti) and ファ (fa) have allowed loanwords to approximate English sounds more closely. Many speakers still adapt these forms in everyday speech, so the original sounds are only partly preserved.

Morphology and verbs

Japanese is an agglutinative language. Stems are extended by chains of suffixes and auxiliary elements, each with a fairly fixed meaning, so one verb form can carry negation, politeness, voice, and tense. The verb tabe- "eat" shows the pattern: tabe-mase-n-deshi-ta means "did not eat" in polite speech, with the polite suffix, the negative, the copula, and the past marker each in a set position. Causative and passive suffixes can stack in the same way (tabe-sase-rare-ru, "to be made to eat"). Japanese has no separate future tense. The non-past form in -ru or -u covers both present and future reference, while aspect is expressed through constructions such as -te iru. Each affix is a separate morpheme, and the analysis of how these units combine is a central task of morphology.

Verbs fall into two regular classes. Consonant-stem verbs, called godan ("five-row") verbs, change their final consonant across the stem, as in kak-u "write" and yom-u "read." Vowel-stem verbs, called ichidan ("one-row") verbs, add suffixes directly to the stem, as in tabe-ru "eat" and mi-ru "see." The class of a verb cannot always be read from its ending. Kaeru "return," for example, ends in -eru but is a godan verb. The small irregular set includes suru "do" and kuru "come," whose stems vary (shi-/sa-, ki-/ko-). The verb iku "go" is irregular only in its -te form, itte.

Each verb has several stems, which are used for different environments. Traditional grammar lists the irrealis stem for negation, passive, and causative; the continuative stem for polite -masu, desiderative -tai, and -te/-ta forms; the terminal stem for plain present; the attributive stem before nouns; the conditional stem for -ba; and the imperative. The -te and -ta forms also involve sound changes, so kaku "write" becomes kaita while yomu becomes yonda and shinu "die" becomes shinda. These alternations are conditioned by the stem's final sound and are a major source of difficulty for learners.

Adjectives split into two types. I-adjectives, such as takai "high," inflect on their own: takakunai (negative), takakatta (past), takaku (adverbial). Adjectives such as ii "good" have irregular forms (yokatta). Na-adjectives, such as shizuka "quiet," take the copula da or desu to predicate. They modify nouns through na (shizuka na heya, "a quiet room"), and the negative is formed with ja nai or de wa nai. The different treatment of the two classes reflects a basic difference in how each combines with tense and the copula.

Word formation relies mainly on derivation, compounding, and reduplication. Suffixes derive nouns from adjectives, as in taka-sa "height," and prefixes add emphasis or politeness. Compounds are very common and may mix native and Sino-Japanese elements. Reduplication produces forms such as hito-hito "people," iro-iro "various," and kuni-guni "countries," and it often involves sound-symbolic effects. Many verbs are also paired by transitivity, for instance aku (intransitive) and akeru (transitive) for "open," which are derived from one another through the choice of suffix.

Sound-symbolic words form a distinctive and productive lexical class. Traditional terms distinguish giongo (imitations of sounds, such as goro-goro "rumbling") from gitaigo (expressions of states or manners, such as doki-doki "pounding heart" or pika-pika "glittering"). Most are reduplicated, and they often take to as adverbs or combine with suru, as in doki-doki suru. Voicing and vowel choice can shift the impression of size, intensity, or hardness. New mimetic forms appear regularly in manga, advertising, and speech, which keeps this class open.

Syntax and sentence structure

Japanese is a verb-final language with a basic order of subject, object, and verb (SOV). The predicate closes a main clause, and most other constituents come before it. Grammatical roles are not shown mainly by word position. Instead, postpositional particles follow each noun phrase and mark its function, so the order of the middle of the sentence can vary more than it would in English. Because the verb arrives last, a listener must hold the earlier material in mind until the predicate resolves it, which is one reason syntax and semantics interact closely in Japanese analysis.

Particles and the wa/ga problem

The particles が (ga) and を (o) typically mark the subject and direct object, while に (ni) and で (de) mark targets, locations, and means. The particle は (wa) is different. It is usually described as marking a topic, a frame that the rest of the sentence comments on, and it often refers back to something already established. The subject particle が tends to appear when new information is introduced or when a question word is answered, so a reply to "who came?" will normally use が. The same event can therefore be described with either particle, and the choice changes what the speaker presents as foregrounded. Grammarians have debated how to analyze the pair for decades. Some treat wa as a marker of contrast and others as a marker of topic status, and the analysis affects how negation and embedded clauses are interpreted. This remains one of the most discussed questions in Japanese grammar.

Omitted arguments

Subjects and objects are frequently left unstated when the context makes them clear, so a sentence can be grammatically complete with only a predicate. Translators therefore often have to decide whether the missing participant is "I," "you," "he," or "they," and the answer may depend on discourse rather than on any overt marker. For parsers and language-processing systems, this means that identifying the referent of an unstated argument is a separate task from parsing the sentence itself, a difficulty discussed further under natural language processing.

Modifiers and embedding

Modifiers precede the words they modify. Adjectives, adverbs, and genitive phrases all stand before their heads, and so do relative clauses. Japanese has no relative pronoun; instead, a clause simply sits before the noun and the gap inside it is understood from the noun's role. In a phrase such as 昨日買った本 ("the book [I] bought yesterday"), the noun 本 is interpreted as the object of 買った even though nothing in the clause points to it explicitly. Such clauses can be stacked, producing long noun phrases that delay the head noun until the end of the phrase. This head-final pattern is a consistent feature of Japanese linguistic typology and is one reason that long sentences place heavy demands on memory.

Sentence-final particles

Sentence-final particles (終助詞, shūjoshi) sit after the predicate and express the speaker's attitude toward the proposition. か (ka) marks a question. よ (yo) is generally used to assert something or to give information the listener is assumed not to have, while ね (ne) asks the listener to share or confirm a belief. The combination よね (yo ne) is used to seek confirmation of something the speaker thinks is already accepted. Other particles, such as な (na) and の (no), carry emotional or informal nuances. Their use depends on pragmatics and on the speaker's relationship with the listener. Some particles have also been associated with gendered speech, and their distribution has changed over time, a point taken up under sociolinguistics. Because these particles interact with the polite forms です and ます, as well as with intonation, a single particle can shift a sentence from polite to blunt or from neutral to emotional.

Vocabulary: native, Sino-Japanese, and loanwords

Three sources and a hybrid category

Japanese vocabulary is traditionally sorted by origin. Wago (和語), native words also called yamato kotoba, make up the inherited core, including much everyday vocabulary such as yama ("mountain") and umi ("sea"). Kango (漢語) are Sino-Japanese words, built from Chinese morphemes and borrowed in several waves over many centuries. Gairaigo (外来語, "words from outside") are loanwords from other languages, and in usage they are generally not counted as kango even when the Chinese origin is not obvious. Hybrids, called konshugo (混種語), combine elements from different sources, as in jūbako (重箱, a stacked food box), which pairs kango 重 with wago 箱.

The shares depend on how the lexicon is counted. Dictionary entries are heavily weighted toward kango, while everyday speech relies far more on native words. Estimates of kango in dictionaries are often given as about 60%, but the proportion in common conversation is much lower. Register matters as well: kango is typically more formal and common in writing, while wago is more casual and frequent in speech. Gairaigo are often associated with modern or stylish usage.

Sino-Japanese words and homophony

Kango entered Japanese through contact with Chinese learning, Buddhism, and the court bureaucracy, and Japanese speakers also later coined new compounds from Chinese elements. Because kango are built from a limited set of Chinese sound patterns, many of them are homophones. The reading kōka can correspond to 効果 ("effect"), 高価 ("expensive"), or 硬化 ("hardening"), among others. Such ambiguity is usually resolved by context, by the surrounding words, or by the choice of script in writing, since the kanji distinguish words that sound identical.

Loanwords from Portuguese, Dutch, and English

The first European loanwords came with Portuguese traders and Jesuit missionaries, who arrived in the 1540s. Words such as pan ("bread"), botan ("button"), and karuta ("playing cards") date from this period. Portuguese loans were numerous at the time, but after the expulsion of Portuguese missionaries and traders in the early 17th century, many fell out of use; one account estimates that only about 100 of roughly 4,000 once in circulation survive. The name igirisu ("England") is itself derived from Portuguese inglês.

During the Edo period, contact with the Dutch at Dejima in Nagasaki introduced terms such as garasu ("glass") and arukōru ("alcohol"). Meiji-era modernization brought large numbers of loanwords from German, French, and English. Over the 19th and 20th centuries English became the dominant source, and it now supplies most gairaigo. Gairaigo are written mainly in katakana, while many older Portuguese and Dutch borrowings are written in hiragana or kanji instead.

Wasei-eigo and ongoing change

Wasei-eigo (和製英語, "Japanese-made English") are expressions built from English material that either do not exist in standard English or carry different meanings there. Sarariiman ("salaried worker") and baikingu (a buffet, from "Viking") are classic examples, and erito shain ("elite employee") mixes an English word written in katakana with a kanji compound. Karaoke combines Japanese kara ("empty") with a clipped form of "orchestra." Some wasei-eigo are used as euphemisms; rabuho for "love hotel" is one example. A few have moved into English, most notably waifu. Many of these patterns became established in the Taishō period.

Vocabulary continues to change. New coinages, abbreviations, and technical terms enter through katakana, and some older loanwords are replaced by native or Sino-Japanese equivalents. Tracking these shifts is an active area of lexical study.

Kana: hiragana and katakana

Hiragana and katakana, collectively called kana, are the two syllabaries that Japan developed for writing its own language. Both took recognizable shape in the Heian period (794–1185). Their common ancestor was man'yōgana, a system that borrowed Chinese characters for their sound alone and ignored their meanings. Hiragana grew out of cursive, simplified versions of whole characters, while katakana was built from angular fragments of characters. Scholars disagree about the earliest dates. Some studies have proposed that the kana idea reached Japan through Buddhist texts from the Korean peninsula in the 8th century, but many researchers question this.

Hiragana became closely associated with women at the Heian court, who used it for poetry, diaries, and letters, and the Tale of Genji is the best-known work written in it. Katakana was traditionally linked with Buddhist scholars, who used it to add reading marks and glosses to Chinese texts. This learned role explains why katakana long remained a script for notes and specialist writing rather than for literature.

The basic chart, the gojūon, arranges about 46 characters in a grid. The columns follow the five vowels and the rows follow consonant groups. The moraic nasal ん does not fit the grid, because it was added long after the ordering was devised. Diacritics extend the chart: a dakuten (゛) voices a consonant, so that k, t, s and h become g, d, z and b, and a handakuten (゜) produces the p series. Small ya, yu and yo (ゃ, ゅ, ょ) combine with an i-row kana to form palatalized syllables such as きょ. Because the grid is organized by sound, the chart reflects the phonology of the language more directly than the older, more irregular spellings did.

Before 1946, kana spelling was based largely on the sounds of earlier stages of Japanese rather than on contemporary pronunciation, which is why the historical orthography is often called "old" spelling. Several characters were used that no longer matched any sound: ゐ and ゑ were written for wi and we, both of which had merged with i and e. The を character was used for the particle, while the same kana also appeared in some words that were pronounced o. Long vowels in yōon syllables were written with two or three full-size kana, since small kana were not part of the system, so きよ could be read either kiyo or kyo. Attempts at regularization had been made earlier, in the Kamakura and Edo periods, but they were not mandatory. After the war, a Cabinet order of 16 November 1946 introduced modern kana usage, which replaced the obsolete characters with i, e and o and brought spelling closer to speech. The system was amended in 1986.

In contemporary writing the division of labour is clear. Hiragana carries native grammatical elements, particles, and the okurigana that follow kanji, and it is the script of children's books and of the furigana readings printed beside difficult kanji. Katakana marks loanwords, onomatopoeia, foreign names, scientific and technical terms, and emphasis. Some words keep historical kana as a stylistic choice, such as ウヰスキー (whisky). For sorting and dictionaries, kana follow the gojūon order, although the older iroha sequence is still used for enumeration. On computers and phones, users typically type kana or romanized input, which is then converted into kanji and kana.

Kanji: readings, forms, and mixed script

Each kanji can carry two kinds of reading. The on-reading (音読み) is a Sino-Japanese pronunciation derived from Chinese sounds borrowed in different periods. The kun-reading (訓読み) attaches a native Japanese word of similar meaning to the character. The character 山 shows the split: in compounds it is usually read san, as in 登山 tozan ("mountain climbing"), while on its own it takes the native reading yama. Choosing between readings depends mainly on the compound a character belongs to and on the surrounding context. Some characters have several readings in each category, so readers often decide the pronunciation only after identifying the word.

The mixed script pairs kanji for content words with kana for grammatical elements. A verb or adjective stem is usually written in kanji and followed by kana endings: 食べる (taberu) is the stem 食 plus the okurigana べる. Okurigana conventions specify which kana attach to a kanji stem, and these rules have been revised over time, so the same word may be spelled differently in different publications. Furigana are small kana printed beside a kanji to show its reading. They appear in children's books, in texts for learners, and beside uncommon words in print, so kanji and kana glossing work together in everyday reading.

Simplified character forms (shinjitai) were introduced after 1945 and are used in the official lists, for example 國 became 国 and 學 became 学. The older traditional forms (kyūjitai) remain in some personal names and in older texts. The Jinmeiyō list still includes traditional characters for names, so a name may be written with a form that differs from the one in ordinary modern printing.

The official lists are central to this system. The 1946 tōyō kanji list set 1,850 characters for general use. It was replaced in 1981 by the jōyō list of 1,945 characters. A 2010 revision added 196 characters and removed five, bringing the total to 2,136, and it also changed the readings of some existing characters. The jōyō list serves as a guideline for official documents, newspapers, and broadcasting rather than a strict prohibition. A separate jinmeiyō list governs which characters may be used in registered personal names. It began in 1951 with 90 characters, and it now contains roughly 860 characters, although published counts differ by year and by how the lists are compared.

The number of characters to teach remains debated. Supporters of a fixed core argue that a defined set makes schooling and publishing more predictable and reduces the burden of learning. Critics argue that a limited list leaves out characters needed for names, place names, and technical vocabulary, and that the restriction can force writers into kana or alternative spellings. Decisions about the list also affect school textbooks, font design, and the character sets that software must support. The choice of how many characters to use therefore has consequences for education and for the typography of printed and digital text.

Romanization, Braille, and other notations

Romanization, the rendering of Japanese in Latin letters, serves readers who cannot read kana or kanji. It appears on street signs, in passports, and in dictionaries and textbooks for learners, and Latin-script writing about Japan in fields such as linguistics and history relies on it heavily (Romanized Japanese is used wherever the target readers cannot read kanji or kana, and it also transliterates Japanese terms in English-language writing on Japanese topics.) Three main systems are in use: Hepburn, Kunrei-shiki, and Nihon-shiki (Hepburn, Kunrei-shiki (ISO 3602) and Nihon-shiki (ISO 3602 Strict)).

The three systems compared

Hepburn uses English-style spelling to approximate how syllables sound to an English reader, so the syllable し is written shi and ちゃ is written cha (the system uses English orthography to phonetically transcribe sounds, reflecting pronunciation in English). The traditional form was codified in James Curtis Hepburn's dictionary of 1886, and a modified version followed in 1908 (the 1886 third edition codified the traditional form, and a revised version was published in 1908). Nihon-shiki was devised by a native speaker for Japanese writers, and it follows the kana syllabary strictly without adjusting for pronunciation changes (it strictly follows the syllabary, with no adjustments for changes in pronunciation). Kunrei-shiki modifies Nihon-shiki so that kana which have merged in modern speech receive the same spelling; for instance づ and ず are both written zu (it eliminates differences between the kana syllabary and modern pronunciation, so づ and ず are represented the same way). The same word can therefore look quite different: 新聞 is shinbun in Hepburn but sinbun in Kunrei (新聞 is shinbun in Hepburn (Kunrei: sinbun)).

Official status and the debate

The government adopted Kunrei-shiki by cabinet order in 1937, and it was reaffirmed in 1954 (it was set by government order in 1937 and reaffirmed in 1954, and it is the romanization taught in elementary schools). Its official status, however, never displaced Hepburn in practice. Hepburn remained the primary romanization system used in Japanese government and by other groups in Japan, and the 1954 notice itself was described by the Agency for Cultural Affairs as non-binding guidance ("is not mandatory and only meant as a government guideline"). Advocates of Hepburn stress readability for non-Japanese readers and the fact that it is already the common usage. Advocates of Kunrei argue that it better reflects the internal structure of Japanese, including word formation and regular sound alternation (Hepburn is modeled on Italian vowel and English consonant values, whereas Kunrei is more faithful to the inner logic of Japanese, including etymology and regular sound shifts). In January 2024 the Agency for Cultural Affairs proposed making Hepburn the standard, and on December 16, 2025 the government decided to issue a cabinet notification revising the rules, replacing Kunrei-shiki with Hepburn (the government decided to issue a cabinet notification revising national rules on romanization for the first time in about 70 years, making Hepburn the standard). Some observers have also noted that the change matters for schools, where textbooks had to teach Kunrei even though pupils met Hepburn everywhere else.

Braille

Japanese Braille, known in Japanese as tenji, transcribes the language in kana, without any means of writing kanji (it is known as tenji, literally "dot characters", and transcribes Japanese much as it would be written in hiragana or katakana). It is structured as a vowel-based abugida, in which the vowel dots occupy the upper part of the cell and the consonant dots occupy the lower part, and a consonant cannot appear alone (it is a vowel-based abugida; the vowels sit in the upper-left corner and may be used alone, while the consonants sit in the lower-right corner and cannot occur alone). Because kana-based braille can be read directly from kana text, it is the form used for ordinary education and publishing. Converting a standard mixed kana-kanji text into braille, however, requires preprocessing: text must first be converted to kana, and word boundaries must be inserted, which makes the process cumbersome (the two-step process is inefficient because speech-to-braille translation passes through kana-kanji conversion).

To encode kanji directly, Tai'ichi Kawakami, a teacher at the Osaka School for the Blind, devised Kantenji in 1969 (devised in 1969 by Tai'ichi Kawakami, a teacher at the Osaka School for the Blind, and still being revised in 1991). It uses an eight-dot cell whose lower six dots match standard Japanese Braille, while the two upper dots indicate the components of a kanji (the lower six dots correspond to standard Japanese Braille, and the upper two dots indicate the constituent parts of the kanji). Its status is unofficial, and it has remained a specialist system.

Vertical and horizontal typography

Traditional Japanese is written in columns from top to bottom, with the columns running from right to left. Horizontal writing from left to right, which spread with Western-style printing, is now standard in scientific texts, business documents, signage, and on screens, while vertical writing remains common in novels, poetry, and some formal and decorative settings. The two layouts affect punctuation and numerals. In vertical text, punctuation marks such as the ideographic full stop 。 and comma 、 are placed in the lower right of their cells, and kanji numerals such as 一, 二, and 三 are customary. In horizontal text, Arabic numerals are widely used, and it is common to find mixed systems in one document, with Arabic numerals for dates and quantities and kanji numerals in set phrases or formal documents. Latin letters and Arabic numerals also appear inline in horizontal text, for example in product codes, abbreviations, and brand names.

Honorifics and politeness

Japanese marks social relations in its grammar and vocabulary to a degree that many languages do not. The main system is keigo (honorific language), which changes verbs, nouns, and sentence endings according to who is acting, who is being addressed, and who is being talked about. Keigo is usually described in three broad categories: sonkeigo, which raises the actions of the other person; kenjōgo, which lowers the speaker's own actions or those of the speaker's group; and teineigo, a general polite register built on the endings desu and masu. Standard textbooks often illustrate the difference with paired verbs. Taberu ("to eat") becomes meshiagaru when the other person eats and itadaku when the speaker receives or eats something, while iku ("to go") becomes irassharu for the other person and mairu for oneself.

In 2007 the Agency for Cultural Affairs issued guidance, published as Keigo no shishin, that replaced the traditional three-way scheme with five categories. Humble language was split into kenjōgo I and kenjōgo II (also called teichōgo, a politeness-oriented humble form), and polite language was divided into teineigo and bikago, the beautifying use of o- and go-. Many textbooks and courses still teach only the older three-part model, so learners often encounter the two systems side by side. Beyond verbs, honorific and polite morphology operates through prefixes such as o- and go-, which attach to words for goods, relationships, and actions to mark courtesy, and through titles and address forms. The suffix -san is the most common general-purpose form of address, while other suffixes indicate narrower relations or ages.

A central organizing principle is the uchi–soto distinction between in-group and out-group. Uchi ("inside") covers one's family, workplace, or other circle, and soto ("outside") covers everyone else. When speaking to an outsider, a speaker humbles their own group and elevates the outsider's group, even when a member of the in-group outranks the speaker internally. A typical example is an employee who refers to their own manager in humble terms when speaking to a client, which may seem counterintuitive to learners from cultures where hierarchy is expressed differently. Because membership is relative to context, the same person may be in-group in one conversation and out-group in another. Kinship terms show this clearly: a speaker uses a humble term for their own father when speaking to an outsider, but an honorific term when addressing the father directly or referring to someone else's father.

Register also matters. Speakers switch between plain forms, used among close friends, family, and subordinates, and polite or honorific forms used with strangers, superiors, and customers. Learners often find the boundaries hard to judge, since the same situation may call for different forms depending on the relationship, the setting, and the topic. Many learners therefore begin with teineigo, which covers most everyday and workplace situations, and move to fuller keigo only when needed.

Honorific usage has changed over time. Elaborate honorific systems were developed in the courts and elite circles of earlier periods and gradually spread to broader society, and modern usage has been shaped by education, business manuals, and media. Public discussion has repeatedly focused on whether keigo is used correctly. Commonly criticized patterns include the overuse of certain constructions, such as the causative-humble form used to express obligatory or polite requests, and the use of respectful forms for one's own actions. Some commentators argue that keigo has become too complicated for everyday use, while others defend it as a necessary expression of social harmony. Reform proposals, including the 2007 guidance, have tried to clarify categories and usage without eliminating the system.

Regional dialects

Japanese dialects (方言, hōgen) are usually sorted into two main clades: an Eastern group that includes Tokyo and a Western group that includes Kyoto, the former capital.The dialects fall into two primary clades, Eastern (including modern capital Tokyo) and Western (including old capital Kyoto). Some classifications set Kyushu and Hachijō apart as additional branches, and the Hachijō varieties may be the most divergent of all. An earlier three-part scheme by Misao Tōjō divided the mainland varieties into Eastern, Western, and Kyūshū groups.Misao Tōjō classified mainland Japanese dialects into three groups: Eastern, Western and Kyūshū dialects. Kagoshima speech is so distinctive that some scholars count it as a fourth branch beside the Eastern, Western, and rest-of-Kyushu groups. The Ryukyuan languages of Okinawa and the southern islands are a separate branch of Japonic and are not dialects of Japanese, even though they are often loosely called that.

Regional variation is older than the modern dialect map. Early poetry already shows it: the Man'yōshū contains poems written in the speech of the capital and of eastern Japan, but other regional forms went unrecorded. Eastern features were rarely passed on to later dialects, with Hachijō as a notable exception. Records became more detailed in the Late Middle Japanese period, when Christian missionaries described regional features. Those missionaries treated the speech of Kyoto court nobles as the true colloquial standard, and the features they noted resemble what is found in modern dialects.

The most audible boundary is pitch. West of the main dividing line, speakers use a Kansai-type pitch accent, while east of it the simpler Tokyo-type system prevails. Tokyo-type accents do reappear further west, beyond the Kansai area. Kansai pitch follows the Keihan pattern, in which the starting pitch of a word matters as much as where the pitch falls. The result can reverse Tokyo patterns: hashi can be pronounced with a low-to-high contour for "bridge" and a high-to-low contour for "chopsticks", almost the opposite of the Tokyo standard.Using the same word, hashi, in many parts of Kansai, 'bridge' may be pronounced with a low-to-high pattern (ha-SHI), while 'chopsticks' is high-to-low (HA-shi)—almost the reverse of the Tokyo standard.

Kansai, the most prominent Western variety, owes its status to history. When Nara and Kyoto were the capitals, its forms became prestigious, and the region kept that influence after the seat of government moved to Edo in 1603.The early capitals of Nara and Kyoto lay within the western area, and their Kansai dialect retained its prestige and influence long after the capital was moved to Edo (modern Tokyo) in 1603. Kansai influence also runs in the other direction: the Tokyo dialect has several western features not found in other eastern dialects. Kansai vocabulary such as meccha ("very"), honma ("really"), and akan ("no good") is common in everyday speech. Its grammar differs as well. The copula da or desu becomes ya or yade in Kansai speech, and the past of the verb harau ("to pay") appears as harōta where standard Japanese has haratta. Kansai idioms have also passed into other dialects and sometimes into standard Japanese.The Kansai dialect's idioms are sometimes introduced into other dialects and even standard Japanese.

Kansai's cultural profile is largely a product of entertainment. Osaka comedy has a long tradition, and since the Taishō period, the manzai form of Japanese comedy has been developed in Osaka, and a large number of Osaka-based comedians have appeared in Japanese media with Osaka dialect. As a result, Kansai speakers are widely perceived as funnier or more talkative, and Tokyo speakers sometimes imitate Kansai speech for comic effect. Kansai itself is not uniform: Osaka speech is generally seen as brisk and comic, while Kyoto speech is considered gentler and more refined.

Grammatical variation also exists within the Western and Kyushu groups. In the Chikuzen variety of northern Kyushu, the distinction between progressive and perfect aspect, which Eastern dialects and standard Japanese lack, is marked by separate suffixes.Like Western Japanese and other Kyushu dialects and unlike Eastern Japanese dialects (including standard Japanese), there is a distinction between the progressive aspect and perfect aspect in the Chikuzen dialect. -Yoru (-よる) or -you (-よう) is used for the progressive aspect, whilst -toru (-とる) or -tou (-とう) is used for the perfective aspect. Negation patterns also differ by generation, with older Chikuzen speakers using forms that younger speakers have partly replaced.

Hachijō is the most isolated case. Its speakers live on the Izu Islands, and the variety is considered a descendant of Eastern Old Japanese, keeping features recorded in eighth-century eastern poetry.Hachijō is considered to be a descendant of Eastern Old Japanese, retaining several unique grammatical and phonetic features recorded in dialect poems from Eastern Japan ("Azuma") in the 8th-century Man'yōshū and the Fudoki of Hitachi Province. Its origins remain unclear, since its shared vocabulary with Kyushu and Ryukyuan speech could reflect settlement, later loans, or independent survival.Hachijō is a moribund language with a small and dwindling population of primarily elderly speakers. Because of this, the Hachijō varieties are both a classification problem and an endangerment case. Mutual intelligibility between dialects and standard Japanese varies by region and by generation, and mass media and migration have steadily narrowed the gap between local speech and the standard.

Standard Japanese: formation and language policy

Standard Japanese was not invented from scratch. It grew out of the speech of Tokyo, which rose to prominence over several centuries. Kyoto had long been the cultural model, and its speech strongly shaped the Edo dialect in the early Edo period. As Edo gained political and economic weight, that dominance weakened. By the late eighteenth century, glossaries of local expressions were using Edo forms as the reference point, which suggests that Edo speech was already treated as the standard before 1868.

The Meiji government then turned this regional speech into a national norm. After Edo was renamed Tokyo in 1868 and became the capital the following year, officials chose the educated speech of the Yamanote districts as the model, not the rougher Shitamachi speech of merchants and craftspeople, even though the Shitamachi variety is the one most closely tied to the old city. Standardization proceeded gradually over almost fifty years. A Ministry of Education body began work on the question in 1903, the same year a government-approved elementary reader using middle-class Tokyo speech appeared, and in 1913 the body's guidelines settled on Tokyo speech as the common language. The Kansai dialect, by contrast, was fixed as a provincial variety after the court moved to Tokyo, and it has kept its prominence as the most widely known non-standard dialect.

Written language was reformed in parallel through the genbun itchi movement, whose name means the unification of speech and writing. Its origins lay in an 1866 petition by Maejima Hisoka calling for script reform, and the term itself was popularized in 1885 by Kanda Kōhei. Writers such as Futabatei Shimei, whose novel Ukigumo (1887–1889) is often cited as the first genbun itchi novel, struggled with archaic conventions even as they moved toward vernacular prose. The movement gained institutional force when school textbooks adopted the style from 1903, and during the early twentieth century archaic forms largely disappeared from print, although they lingered in government documents until the end of the Second World War.

The postwar period brought the most sweeping changes to standard writing. On 16 November 1946, the cabinet promulgated a list of 1,850 characters for general use (tōyō kanji) and a new kana orthography, which was amended in 1986. The character reform used simplified forms (shinjitai) for many kanji, and Japan adopted fewer and less drastic simplifications than mainland China did. Older forms remained in print into the 1950s because of the time needed to change typesetting equipment. The list was later replaced in 1981 by a more flexible system. These reforms were controversial. Some critics later argued that the moderate changes were unnecessary, while defenders of the traditional system accepted the 1946 list as a compromise that closed the debate over abolishing kanji.

Standardization also carried social costs. Ministry textbooks presented Tokyo speech as the model, and critics have argued that this encouraged speakers of other dialects to feel their own speech was inferior. Dialect use was discouraged in schools, and regional speech came to be seen as a marker of rural or provincial status. Standard Japanese remains the variety used in education, broadcasting, and official settings, although its dominance has weakened since the Second World War, and many regions continue to maintain their own everyday speech.

Ryukyuan languages and endangered varieties

The Ryukyuan languages form the second main branch of the Japonic family, alongside Japanese. Their speakers live on the Amami, Okinawa, Miyako, and Yaeyama island groups, which are also known collectively as Shimakutuba. The six Ryukyuan languages are often divided into Northern and Southern groups. The Northern group includes Amami, Okinawan, and Kunigami, while the Southern group includes Miyako, Yaeyama, and Yonaguni. The number of languages depends on the method. A handbook of the field counts five, while the UNESCO atlas, which adds Kikai, gives six. Within Japan, the Ryukyuan varieties have often been classified as dialects of Japanese for socio-political reasons, a view most mainland Japanese have held since World War II. Linguists now treat them as separate languages, since they are not mutually intelligible with Japanese or with each other.

Their sound systems and grammar differ clearly from standard Japanese. Southern Ryukyuan languages have a number of syllabic consonants, including unvoiced syllabic fricatives. Okinawan in particular is a language distinct both from Japanese and the other five Ryukyuan languages, so the group is internally diverse as well as distant from the mainland standard. For the wider picture of how such features are described, see phonology and dialect.

Degrees of endangerment

UNESCO's 2009 atlas placed the Ryukyuan languages in its endangered category, but its ratings differ between sources. One widely cited summary says UNESCO labels four of the languages "definitely endangered" and two others "severely endangered." Another local guide classes Amami, Okinawan, Kunigami, and Miyako as definitely endangered, and Yaeyama and Yonaguni as severely endangered. Individual varieties are often even smaller. The Kurima variety of Miyako is spoken natively only by elderly people, and the Yonaguni language has been estimated at around 400 speakers.

The loss of speakers follows a generational pattern. Proficient speakers are limited to old generations, typically in their 50's and older, while most of the younger generation are monolingual in Japanese. Japanese has taken over almost all daily uses, so all six languages are highly endangered today. Projections in the literature have been stark: UNESCO estimates that all Ryūkyūan languages face extinction as early as 2050, a figure that should be read as a projection rather than a fixed date.

Hachijō and the classification problem

The Hachijō varieties, spoken on the southern Izu Islands and on the Daitō Islands, show both problems at once. Based on the criterion of mutual intelligibility, Hachijō may be considered a distinct Japonic language, rather than a dialect of Japanese. Its place within the family is less settled: its position within the family is unclear. Hachijō is also not mutually intelligible with standard Japanese, and retains grammatical and phonetic features from Eastern Old Japanese. Each village has its own variety, with each village where Hachijō is spoken has its own local dialect.

Speaker figures vary widely. One source gives maybe a few hundred speakers of Hachijō, most of whom are elderly. Another reports fewer than 1,000 native speakers (2011), and a municipal-level account states that the language has 0 native speakers being children or young adults. Hachijō is also taught in schools, but few younger people are learning the language at home.

Historical causes of language shift

The decline has a documented history tied to state institutions. After the Ryukyu Domain became Okinawa Prefecture in 1879, Vice Minister of Education Tanaka Fujimaro was dispatched to the archipelago with the objective of developing and implementing an educational policy for the prefecture. A bilingual textbook, used from 1880 onwards in all schools of the prefecture, was part of this effort. In 1907, Ryukyuan languages were banned from schools in the so-called Ordinance to Regulate the Dialect. Schools also used a system of "dialect tags" that punished children for speaking local languages. During the war, speaking Ryukyuan was officially illegal, although in practice the older generation was still monolingual. Transmission was further disrupted when natural intergenerational language transmission in the Ryukyuan languages was interrupted from the early years of the US occupation.

Revitalization efforts

Revitalization is pursued by communities, prefectural bodies, and researchers. Okinawa is the most active case: the situation in Okinawa, which is at the center of this chapter, is the most favorable. The prefecture set up a Center for Spreading Community Languages in 2017, which organizes instructor training and speech contests, and local governments have published readers of community languages that can be used in elementary and junior high schools. Surveys suggest public interest: some 70% of those surveyed state that they want their local languages to be taught at school. However, this remains impossible at present because school curricula are entirely decided by the Japanese central government. Student-centred efforts such as extracurricular dialect clubs have existed since 1989, but participation has remained small. Elsewhere the picture is more limited, as the situation is less encouraging, in particular, in Amami and in Yonaguni. Efforts continue to be debated, and the current status of individual programmes varies by island and by year.

Sociolinguistic variation: gender, age, and register

Gendered speech in Japanese has been described mainly through two kinds of marker: sentence-final particles and first-person pronouns. Traditional accounts link the particles wa and kashira with women's speech and ze and zo with men's, while ne and yo are usually treated as gender-neutral. Men's speech has been described as using pronouns such as ore and boku, the informal copula da, and particles such as ze, zo, and kana, with fewer polite forms and less frequent honorific prefixes. These descriptions come largely from prescriptive guides and from men's reports of their own speech, so they tell us as much about ideology as about everyday practice.

Actual usage often diverges from these norms, and the norms themselves have shifted. A large corpus study of about 408 million Japanese tweets from 2015 to 2019 found gendered particles and pronouns in roughly 6% of the posts, which suggests the features are not confined to speech. A study of casual conversation among college students found that male speakers used the "feminine" particles wa and no more often than female speakers did in its sample. Analyses of manga have reported that Japanese may be becoming less overtly gendered than its traditional description suggests, with manga itself sometimes cited as an influence. Ethnographic work at a junior high school found girls using the plain masculine pronoun boku as a marker of a confident, non-traditional identity, while boys who used it were judged differently. A study of a reality television program found a gap between how speakers actually used these features and the stereotyped image that media presented.

Age-based variation is most visible in youth vocabulary, which changes quickly. The word yabai, originally meaning "dangerous," has broadened into a general intensifier with both positive and negative senses, and the character 草 (kusa, "grass") is used as a shorthand for laughter because of its visual resemblance to the written form of laughing. Verbs are formed from trendy loanwords by adding the suffix ru, as in tapiru, derived from tapioca and meaning to drink bubble tea. Shortened forms such as rii, used as a quick acknowledgement in text messages, are typical of digital communication. Generational change also operates in pronunciation and in regional vocabulary. A dialect survey of the Shōnai district in Yamagata, comparing data gathered in 1950 with a second survey in 2018, showed that many older regional words were kept only by elderly speakers, while some forms that matched Tokyo colloquial speech survived among younger people.

Register is a further axis of variation. Classical and colloquial styles, formal written prose, and casual spoken language differ in grammar, vocabulary, and politeness, and digital communication has added abbreviations, emoticons, and playful kana and kanji usage that do not fit neatly into either category. Group-specific speech adds another layer. Occupational jargon, subcultural vocabulary, and fan language such as oshikatsu (推し活, activities centred on supporting a favourite performer) show how groups create their own registers. The 2025 "new words" selection by the publisher Sanseidō, for instance, featured words that were tied to particular communities and online spaces.

Media both reflects and shapes these styles. Manga and anime have long used stylised speech to mark characters, such as samurai or ninja who use archaic copulas and first-person forms, and female idol song lyrics have been analysed as constructing gendered personae, including cross-gender performances. Such representations can reinforce stereotypes of how men and women, or people from particular regions, are expected to speak, and they can also be borrowed by ordinary speakers. Research in this area therefore often examines the distance between what speakers do, what audiences expect, and what media portray, an approach that connects sociolinguistics with pragmatics and with the study of social norms and socialization.

Acquisition, literacy, and education

Children usually acquire spoken Japanese at home before school age, learning the core sound system, common verb endings, and the basic particles through everyday exchange. Politeness registers and honorific forms are typically mastered later, often in school or through contact with adults outside the family. Written Japanese is taught separately and gradually. Children first learn hiragana, then katakana, and then begin the kanji sequence in primary school.

The kanji curriculum is organized around the Jōyō kanji list, which was revised to 2,136 characters in 2010. Its earlier forms date to 1946, when a list of 1,850 characters was adopted, and to 1981, when it was expanded to 1,945. Of the 2,136 characters, 1,026 are the kyōiku kanji taught in the six years of elementary school, and the other 1,110 are taught in junior high school. In elementary school, the load is set by grade: 80 characters in grade 1, 160 in grade 2, 200 in grade 3, 202 in grade 4, 193 in grade 5, and 191 in grade 6. Students therefore reach a cumulative total of roughly 240 characters after the second year and about 835 after the fifth. Because kanji readings vary with context, this graded sequence is also a long process of learning compounds and their pronunciations, not only shapes. This heavy load is one reason the education system is often described as central to Japanese literacy.

Literacy in Japan is widely reported at around 99 percent of adults, but this figure is less secure than it appears. One UNESCO-linked database entry notes that no national literacy survey has been carried out since the figure was established, so the 99 percent consensus is unconfirmed. The same source points to lower literacy among some groups, including Burakumin communities and some Ainu people, and it notes that many non-Japanese residents lack sufficient Japanese for daily life, a gap that is addressed largely through voluntary local programs rather than national adult education policy. The OECD Survey of Adult Skills (2023) offers a more direct measurement for adults aged 16 to 65. It finds that Japanese adults score above the OECD average in literacy, with about 10 percent scoring at Level 1 or below, compared with an OECD average of about 26 percent.

Japanese taught as a foreign language is assessed mainly through the Japanese-Language Proficiency Test (JLPT), which has five levels from N5 (beginner) to N1 (advanced). In Japan, it is administered by the Japan Educational Exchanges and Services (JEES), while in the United States it is run by the American Association of Teachers of Japanese (AATJ). Demand is high: registration for the December 2026 test in Japan reportedly reached capacity within days, and one U.S. university reported record numbers of examinees for the sixth consecutive year, with a notable rise at the lower levels N3 to N5. Sources differ on how often the test is offered in the United States; some describe a single December session, while others mention both July and December. Learners commonly find the kanji reading system, the particles, politeness levels, and omitted subjects most difficult, and many programs combine textbooks with graded kanji study and conversation practice.

Heritage speakers, such as children born abroad to Japanese parents or those who return to Japan after years overseas, present a different pattern. They often understand and speak Japanese well but may read and write less, depending on how much schooling in Japanese they receive. Weekend Japanese schools and supplementary programs have traditionally helped families maintain the language, and the balance between home use and the majority language is a major factor in whether it persists across generations.

Teaching methods, textbooks, and assessment tools have changed in recent decades, with more emphasis on communicative skills and digital resources. Online courses and remote tutoring expanded considerably during the COVID-19 pandemic, and many learners now combine self-study apps with live classes, though the effectiveness of these approaches for kanji and speaking practice remains an area of ongoing discussion.

Japanese around the world

Estimates of how many people speak Japanese cluster around 123 million, but published totals differ with the year and the counting method. Some sources report figures near 122 million for first-language speakers, while others include second-language learners and reach 126 to 130 million. Counts also differ on whether they include heritage speakers who understand Japanese without using it daily. Most native speakers, roughly 99 percent by one 2018 estimate, live in Japan, the only country where Japanese is the national language. Most of the remainder are in the United States (especially California and Hawaii) and in Brazil (mainly São Paulo and Paraná). Japanese therefore has a large population of speakers in one country and a much smaller spread of communities abroad.

The diaspora

Japanese emigration began in the late 19th century, when large numbers of workers went to Hawaii, the mainland United States, and later Brazil, where the economic conditions of early 20th-century Japan pushed many families to leave. Brazil holds the largest population of people of Japanese descent outside Japan, with estimates around 1.5 million, but descent does not equal fluency. Many second-, third-, and later-generation Brazilians of Japanese origin speak Portuguese as their main language, and Japanese survives mainly among older speakers and in community institutions. In the United States, Japanese-speaking communities have been most visible in California and Hawaii, and wartime incarceration and assimilation pressures in the 1940s weakened transmission to children. Smaller communities exist in Canada, Australia, Southeast Asia, and elsewhere, often tied to business postings and short-term residence rather than long-term settlement.

Angaur and the language's official status

Japanese holds official status in one place outside Japan: Angaur, a small state in the Republic of Palau. Palau came under Japanese administration in the period before the Second World War, and the Japanese presence there, including in the mining and administrative life of Angaur, left a lasting imprint. The state constitution, drafted in 1982, names Palauan, English, and Japanese as its official languages. This status is largely symbolic. The Angaur community is very small, and fluent Japanese speakers are now mostly older residents, while younger people use English or Palauan. Japanese remains recognized at the state level within a national system in which Palauan and English are the official languages.

Colonial-era spread

Between 1895 and 1945, Japan ruled Taiwan, and from 1910 to 1945 it ruled Korea. In both territories, the government treated Japanese as the national language, known as kokugo, and made it central to education. In Korea, the 1911 imperial education decree designated Japanese as the national language, and a large share of school hours went to Japanese instruction. Schools were organized on a dual basis, with separate tracks for Japanese residents and for colonial subjects. Taiwan also had national language training schools for both children and adults, and the government encouraged families to use Japanese at home. Historians generally agree that these policies aimed at assimilation and that their results were uneven. Legal distinctions, such as the household registration system that separated colonizers from colonized, meant that proficiency in Japanese did not bring equal status. Many older people in these territories learned Japanese at this time, and that generation's knowledge is part of the legacy that persisted after 1945. The same era also shaped language policy in Palau and other Pacific territories.

Learning Japanese abroad

Interest in learning Japanese dates back to the 19th century, but it grew markedly after Japan's economic boom in the 1980s and, from the 1990s, with the global spread of anime, manga, and video games. Surveys reported more than 3.6 million learners worldwide in 2015, concentrated in East and Southeast Asia. China, South Korea, Indonesia, and Australia have some of the largest numbers of students in schools and universities, although sources disagree on which country has the most learners. Many universities worldwide offer Japanese courses, and a growing number of secondary and primary schools include it in their curricula. The popularity of pop culture has also made Japanese a common choice for self-study, which is harder to count than enrollment in formal programs.

Japanese in computing and language technology

Japanese text has been carried by computers through several overlapping encoding families. The JIS standards, published by the Japanese Standards Association, were the basis for most early systems, and there are three JIS encodings (Shift JIS, EUC, ISO-2022-JP) and three Unicode encodings (UTF-8, UTF-16, UTF-32) in widespread use. Shift JIS was developed in the early 1980s by ASCII Corporation with Microsoft, and was defined as Appendix 1 to JIS X 0208:1997 in 1997. Its place on Windows and on early web pages made it the de facto standard for years. EUC-JP, by contrast, became the usual choice on Unix-derived systems, and ISO-2022-JP was mainly used for e-mail, where the original Internet e-mail standards only support 7-bit transfer protocols. Unicode has steadily taken over, but legacy files and older devices keep converters in demand.

Unicode also raises a specific issue for Japanese. Under Han unification, ideographs judged to share a historical origin across Chinese, Japanese and Korean receive a single code point even when their regional glyph shapes differ. As a result, the same character may look different depending on the font or on the language tag applied to the text.

Most users enter Japanese by typing kana or romanized readings, and a kana-to-kanji conversion engine, known as an input method, offers a list of candidate words for the typed reading. Because many kanji share the same reading, the engine ranks candidates by surrounding context and by usage frequency, and the user selects the intended form. The same mechanism affects the phones and computers that most people now write on every day.

Natural language processing faces a basic obstacle in Japanese. In non-segmented text like Japanese or Chinese, however, word boundaries are implicit rather than explicit. A program therefore cannot split the text on whitespace as it can for English. Experiments that delete the spaces from space-delimited corpora show how much these markers matter, since accuracy drops sharply once they are removed. Japanese analysers, such as MeCab, combine large dictionaries with statistical models to find word boundaries and assign parts of speech, a process usually called morphological analysis. Defining the "word unit" is itself contested, because the same string may be one entry in one dictionary and two in another. One dictionary-free approach reported recall of up to 98.3%, but it relied on transitions between character types and hiragana frequencies, and it was tested on non-kanji strings, so its performance on kanji-heavy text is a separate question.

Machine translation and speech recognition face other language-specific problems. Subjects and objects are often omitted, so translating into English requires inferring pronouns the Japanese sentence never states. Honorific forms encode the relationship between speaker, listener and referent, and a translation system must either preserve that information or lose it. In speech recognition, a single spoken sequence such as kōka may correspond to many different words written with different kanji, so the system must choose by context.

Tokenization is another area where Japanese poses distinctive questions for large language models. Tokenizers are usually trained on data dominated by English; one report on Llama 3 stated that 95% of the training dataset consisted of English and code. In a study of the Mistral tokenizer, the limited vocabulary size was not optimized specifically to parse the kanji characters, which causes the tokenizer to encode a large proportion of them as a sequence of bytes. The result was a low character-per-token rate, which significantly reduces the length of Japanese text that can be processed by the model. The researchers responded by adding 16k Japanese-oriented tokens to the vocabulary. Efficiency still varies by text type and tokenizer, and one 2026 comparison found that the largest spread for Japanese reached 60%.

History of research on Japanese

The first systematic descriptions of Japanese were written by Christian missionaries. Jesuits working in Japan from the 1540s needed the language for their work, and they compiled dictionaries and grammars to train newcomers.Christian missionary work in Japan began in the 1540s, necessitating the learning of its language. Missionaries created dictionaries and grammars. The most influential product of this effort was the work of João Rodrigues, a Portuguese Jesuit who had arrived in Japan as a teenager and who worked as an interpreter for leading political figures of the period. His grammar, the Arte da Lingoa de Iapam, was printed at Nagasaki between 1604 and 1608 and remains the oldest complete grammar of Japanese.It was compiled by João Rodrigues, a Portuguese Jesuit missionary. It is the oldest fully extant Japanese grammar and is a valuable reference for the late middle period of the Japanese language. Rodrigues's contemporary dictionary, the Nippo Jisho, was published in 1603 and was the first dictionary of Japanese to a European language.The Nippo Jisho is a Japanese-to-Portuguese dictionary compiled by Jesuit missionaries and published in Nagasaki, Japan, in 1603. Containing entries for 32,293 Japanese words with explanations in Portuguese, it was the first dictionary of Japanese to a European language. These works were written in Latin letters, so they record the pronunciation of their time in a form that later scholars use as evidence for the Late Middle Japanese period.

Within Japan, the kokugaku ("national learning") movement of the Edo period approached the language through philology. Its scholars concluded that the old texts had become hard to read because of changes in the language itself.Chinese influence had altered the Japanese language so that the ancient texts could no longer be understood without careful linguistic analysis. Kamo no Mabuchi advised Motoori Norinaga to first study the Man'yōshū poetry in order to understand the man'yōgana system of writing, rather than starting with the Kojiki commentary he had requested.Instead, Mabuchi recommended that he first analyze Manyoshu (ancient poetry) in order to gain an understanding of "manyogana," ancient Japanese language. Motoori's main achievement was his commentary on the Kojiki, which he worked on for about thirty-five years.Norinaga's most important works include the Kojiki-den (Commentaries on the Kojiki), written over a period of 35 years, and his annotations on the Tale of the Genji. Modern historians of linguistics regard his decoding of the eighth-century orthography of the Kojiki as an important step toward the study of Old Japanese.An accomplished philologist, he helped decipher the idiosyncratic eighth-century orthography of the Japanese chronicle of history and myth, the Kojiki (Records of Ancient Matters).

From the late nineteenth century, Japanese grammar was reorganized under the influence of Western linguistics. In the modernising era of the late 19th and early 20th centuries, grammatical studies were advanced under the rubric of kokugogaku (国語学), influenced by Western grammar. The discovery of the Jōdai tokushu kanazukai by Hashimoto Shinkichi (1882–1945) was especially consequential. It showed that Old Japanese distinguished more syllables than later stages, which led him to propose that Old Japanese had eight vowels, where modern Japanese has five.Hashimoto is especially noted for the discovery of Jōdai Tokushu Kanazukai, which makes it clear that Old Japanese made more syllabic distinctions than later periods of the language. This discovery led him to hypothesize that Old Japanese had eight vowels, while modern Japanese has only five. Hashimoto also proposed the bunsetsu as a grammatical unit in 1933.Hashimoto (1933) set up a unit called the 文節 bunsetsu, literally 'sentence segment', which he equated to Matsushita's 詞 shi. His school grammar was widely adopted in Japanese education.Hashimoto's analysis, adopted in the Ministry of Education's middle school grammar, Chut6 bumpo, has become widely accepted throughout the country.

Tokieda Motoki (1900–1967) developed a rival framework. He criticized Hashimoto's approach and built his own theory on the basis of a "process theory of language" (gengo katei setsu).He is noted for developing the Process Theory of Language (言語過程説, gengo katei setsu) and his criticism of Ferdinand de Saussure. His proposals were also distinct from those of Mikami Akira (1903–1971), who tried to describe Japanese grammar from within the language and without much regard for the schools of Western theory.Notons à ce propos qu'un autre linguiste, MIKAMI Akira (1903-1971), a tenté, lui aussi, de décrire la grammaire du japonais de l'intérieur sans trop se soucier des écoles Tokieda's ideas gained only partial acceptance among scholars.The distinction thus set up between the shi and the ji seems to solve the problem of the difference between the inflected and uninflected suffixes, whose precise classification had bothered Hashimoto, but Tokieda's proposals have won only partial acceptance in the scholarly world.

Research since the mid-twentieth century has been shaped by Western linguistic theories, including structuralism and generative grammar, as well as by the native tradition. Dialect surveys, historical phonology, and studies of written texts continued to develop alongside these approaches. More recently, researchers have used large text corpora, instrumental phonetics, and computational models. These methods have opened new questions about the analysis of particles, auxiliaries, and word classes, and they have not settled the debates. The disagreement between Japanese-centred and Western-based frameworks, and between formal and functional accounts of grammar, remains an active part of the field.

Origins debate and unresolved questions

Most of what is known about Proto-Japonic is a reconstruction. Scholars have built it by combining internal reconstruction from Old Japanese, the comparative method applied to Old Japanese dialects and the Ryukyuan languages, and the 20th-century work of Samuel Elmo Martin and Shirō Hattori. Their proposals still differ on the consonant inventory, the syllable shapes allowed in roots, and the reconstruction of many individual words. Because the reconstructed ancestor is only as firm as its evidence, the early stages of the family remain open to revision.

Old Japanese presents a separate problem. Its eight-vowel reading and its kō-otsu distinction, in which a set of syllables appears in two groups of characters that seem to have contrasted in sound, are attested only through the orthography. The early writing system used Chinese characters for their sound value, and it did not mark pronunciation directly. Scholars therefore propose different phonetic values, varying in vowel quality, palatalization, or the consonant before the vowel, and the choice among them shapes how the early language is reconstructed. The texts themselves, including the Kojiki (712), the Nihon Shoki (720), and the Man'yōshū, were composed in a script designed for Chinese, and compilers' purposes shaped what was written down. They therefore record a learned, partly standardized variety, which limits what they can show about everyday speech.

The date of the split between Japonic and related languages, and of its arrival on the islands, is also disputed. A 2011 Bayesian phylogenetic study by Sean Lee and Toshikazu Hasegawa placed the root of the family about 2,200 years ago and linked it to rice farming. Most historical linguists associate the arrival of Japonic with the Yayoi period, but they disagree on whether the spread happened in the protohistoric or the historical phase. In 2020, Chaubey and van Driem proposed that Japonic was already present in parts of the archipelago during the Jōmon period, linked to a southwestern Jōmon population, and then expanded during the Yayoi period through assimilation. Archaeologist Kazuo Miyamoto, writing in 2022, argued instead that Proto-Japonic passed through the Korean Peninsula and reached Japan at the start of the Yayoi period, around the ninth century BCE. The proposed dates therefore differ by roughly a millennium.

The competing hypotheses about origins can be grouped into three broad types. The population-movement model ties Japonic to Yayoi farmers from the Korean Peninsula, with later migrations also considered. The Korean-link model rests on lexical correspondences, such as the forms for "three" and "seven" that Lee Kim-moon compared between Koguryo and Old Japanese, and on archaeological continuity. Ancient DNA studies have reported West Liao River-related ancestry in both Japanese and Korean populations, which some read as support for a Transeurasian background, although the connection between genes and language is indirect. The substrate model proposes that an Austronesian layer lies beneath Japanese. Ōno Susumu and others have argued for it, and Holmer (2010) and Yoshizu (2011) claimed that morphology and basic vocabulary show resemblances to Austronesian. Mixed-language accounts, such as Ono's 1970 Austronesian-Altaic proposal, have a long history in Japanese scholarship. Critics note that the lexical evidence depends on which word lists and criteria are chosen, and that researchers often disagree on which correspondences count. Jared Hudson's 2022 review also points out that little clear evidence shows Jōmon-language influence on Proto-Japonic, and that some mixed-language theories were popular when the archaeology was poorly known.

Several questions remain open. Researchers have not agreed on whether Ainuic was the only language family in Jōmon Hokkaido, nor on how far the Yayoi migrants contributed to the genetic makeup of later populations. The active lines of work include new archaeological excavations, larger ancient DNA datasets, and computational methods for reconstructing ancestral languages and testing competing phylogenies. Each of these may narrow the range of dates and origins, but none has yet settled the question.

References

1. Reconstruction of Ryukyuan tone classes of Middle Japanese Class 2.4 and 2.5 nounsdegruyterbrill.com
2. Classification of the Japonic languagesen.wikipedia.org
3. Japanese Language: History, Writing System, AI Challengesblog.pangeanic.com
4. Japonic languagesstatic.hlt.bme.hu
5. Proto-Japonic languageen.wikipedia.org
6. Re-thinking Jōmon and Ainu in Japanese History - Asia-Pacific Journal: Japan Focusapjjf.org
7. Japan considered from the hypothesis of farmer/language spreadncbi.nlm.nih.gov
8. The emergence of ‘Transeurasian’ language families in Northeast Asia as viewed from archaeological evidencencbi.nlm.nih.gov
9. Ancient genomics reveals tripartite origins of Japanese populationsncbi.nlm.nih.gov
10. Japonic languagesen.wikipedia.org
11. Insular Japonic languagesen.wikipedia.org
12. Old Japaneseen.wikipedia.org
13. 2 The Phonology of OJconf.ling.cornell.edu
14. (Early) Old Japanese Phonologywakapoetry.net
15. Heian perioden.wikipedia.org
16. The History of Japanese Languages - JapanesePod101.com Blogjapanesepod101.com
17. 0% found this document useful (0 votes)scribd.com
18. Edo perioden.wikipedia.org
19. Japanese language historypreply.com
20. 1868 in Japanen.wikipedia.org
21. The Fascinating History of the Japanese Language: From Ancient Roots to Modern Times - Rocket Languagesrocketlanguages.com
22. R. Edward Smith: Natural Phonology of Japanesetrussel.com
23. Kanjiultimatepopculture.fandom.com
24. The fusion and conflicts between the Japanese language ...maajournal.com
25. Translation via Annotation: A Computational Study of Translating Classical Chinese into Japanesearxiv.org
26. Kanjien.wikipedia.org
27. Kanbun-LM: Reading and Translating Classical Chinese in Japanese Methods by Language Modelsarxiv.org
28. The History of Kanji 漢字の歴史 - It's Japan Timeitsjapantime.com
29. Japanese writing systemen.wikipedia.org
30. Why Does the Japanese Language Have Three Alphabets?ai.glossika.com
31. Man’yōganabritannica.com
32. A brief history of the Japanese writing system < Skritter Blogblog.skritter.com
33. List of Chinese%E2%80%93Japanese false friendsen.wikipedia.org
34. The Origins of Japanese Writing: Understanding Man'yōganabriefpedia.org
35. Word Order & Particles — I can use basic word order and...varsitytutors.com
36. An algebraic approach to translating Japanesearxiv.org
37. Japanese Grammar for Beginners: The Essential Guide · Shinobi Japaneseshinobi-japanese.com
38. Japanese Sentence Structure & Word Order - JapanesePod101.com Blogjapanesepod101.com
39. What Are the Sentence-Final Particles in Japanese? - JapanesePod101japanesepod101.com
40. Sentence-Ending Particles — I can use common sentence-ending...varsitytutors.com
41. Language and an expression of identities: Japanese sentence-final particles ne and naresearchgate.net
42. Wagoen.wikipedia.org
43. Sino-Japanese vocabularyen.wikipedia.org
44. Loanwords in Japaneseen.wikipedia.org
45. List of Japanese words of Portuguese originen.wikipedia.org
46. List of gairaigo and wasei-eigo termsen.wikipedia.org
47. Wasei-eigoen.wikipedia.org
48. Wasei-kangoen.wikipedia.org
49. Mentonasianstudies.org
50. Examining the Phenomenon of Wasei Eigo: English Pseudo-Loanwords in Japanesedl.tufts.edu
51. Wasei Eigo, the Japanese words that reinvent Englishyahoo.com
52. Japanglish: 26+ Japanese English Words You Can Easily Learnlingopie.com
53. These Loanwords in Japanese Didn't Come From Englishcotoacademy.com
54. Japanese Language Basics: Katakanakcpinternational.com
55. Kana Definition - History of Japan Key Termfiveable.me
56. The Origin and Development of Hiragana and Katakana (PDF)academia.edu
57. Introduction to Historical Kana Orthographybyvoid.com
58. Kana - New World Encyclopedianewworldencyclopedia.org
59. Historical kana orthographyen.wikipedia.org
60. Modern kana usageen.wikipedia.org
61. Wo (kana)en.wikipedia.org
62. Wi (kana)en.wikipedia.org
63. Goj%C5%ABonen.wikipedia.org
64. J%C5%8Dy%C5%8D kanjien.wikipedia.org
65. The Jōyō Kanji List (常用漢字): The 2,136-Character Set Explainedj-compass.com
66. Jinmeiy%C5%8D kanjien.wikipedia.org
67. Ky%C5%ABjitaien.wikipedia.org
68. Romanization of Japaneseen.wikipedia.org
69. Hepburn romanizationen.wikipedia.org
70. Romanization Systems: Hepburn, Kunrei, Nihon-shikielon.io
71. Kunrei-shikien.wikipedia.org
72. “Sushi” Or “Susi”? Japan to Rethink Longstanding Romanization Rulesnippon.com
73. Romanization rules are changing. Why Kunrei won’t be missed. - The Japan Timesjapantimes.co.jp
74. Japanese Brailleen.wikipedia.org
75. (PDF) The Translation System from Japanese into Braille by Using MeCabresearchgate.net
76. Braille kanjien.wikipedia.org
77. Nihon-shikien.wikipedia.org
78. Japanese Keigo Guide: Sonkeigo, Kenjougo, Teineigo & Baito Keigo Explainedmy-senpai.com
79. 敬語の5分類｜The 5 Types of keigotumblr.com
80. Keigo: The Japanese Art of Honorific Languagejapaneserituals.com
81. Guide to Keigo: How to Correctly Use Japanese Honorificscotoacademy.com
82. Uchi%E2%80%93sotoen.wikipedia.org
83. Uchi-sotomanga.fandom.com
84. How to Address Family Members in Japanese - Learn languages with italkiitalki.com
85. Honorific speech in Japaneseen.wikipedia.org
86. Japanese dialectsen.wikipedia.org
87. Understanding the Unique Intonation and Expressiveness of Osaka’s Kansai Dialect (Kansai-ben)osakaa.net
88. Kansai dialecten.wikipedia.org
89. Chikuzen dialecten.wikipedia.org
90. Hachij%C5%8D languageen.wikipedia.org
91. Japanese dialectsstatic.hlt.bme.hu
92. Japanese pitch accenten.wikipedia.org
93. The Kansai Dialectsenseijapanese.com
94. Discovering Japanese Dialects: Kansai-ben, Kanto-ben, Tsugaru-ben, and Moretcj-education.com
95. Tokyo dialect — Wikipedia Republished // WIKI 2wiki2.org
96. Japanese Writing: The Kanji Debate of the Meiji Period - Unseen Japanunseen-japan.com
97. The Genbun Itchi Movementoxfordre.com
98. Genbun itchien.wikipedia.org
99. The Genbun Itchi Movementouci.dntb.gov.ua
100. T%C5%8Dy%C5%8D kanjien.wikipedia.org
101. Japanese script reformen.wikipedia.org
102. Chapter 1 of Literacy and Script Reform in Occupation Japanu.osu.edu
103. 1. The Standard Japanese VS the Kansai dialectweb.mit.edu
104. Shimakutuba and Uchinaaguchi (Ryukyuan Languages and Dialects) - Okinawa Studies Research Guide - Research Guides at George Washington Universitylibguides.gwu.edu
105. Ryukyuan languages (琉球諸語) — Edgepediaedgechat.ai
106. Ryukyuan languagesen.wikipedia.org
107. Use them or lose them: There's more at stake than language in reviving Ryukyuan tongues - The Japan Timesjapantimes.co.jp
108. Kurima dialecten.wikipedia.org
109. Ryukyuan Languages of Japan2m.com.au
110. UNESCO Recognizes Ryukyuan Languages as Endangered - Language Conflict Encyclopedialanguageconflict.org
111. Hachijō languageomniglot.com
112. Hachij%C5%8D Subprefectureen.wikipedia.org
113. Language Loss and Revitalization in the Ryukyu Islands - Asia-Pacific Journal: Japan Focusapjjf.org
114. Language Loss and Revitalization in the Ryukyu Islandscambridge.org
115. Okinawans in Japan - Language Conflict Encyclopedialanguageconflict.org
116. (PDF) Ryukyuan Languages in Japanresearchgate.net
117. 25 25 2 RYUKYUAN SOCIOLINGUISTICSiris.unive.it
118. Yonaguni languageen.wikipedia.org
119. Miyakoan languageen.wikipedia.org
120. (PDF) Language Shift in the Ryukyu Islandsresearchgate.net
121. Focusing on Ryukyuan/Okinohchr.org
122. Hachijō-jimaen.wikipedia.org
123. [PDF] The Hachijo Language of Japan: Phonology and Historical Developmentsemanticscholar.org
124. Is Japanese gendered language used on Twitter ? A large scale studyarxiv.org
125. Online Journal of Communication and Media Technologies, 2020, 10(4), e202024ojcmt.net
126. Unconventional Usage of Gender-Based Japanese Sentence-Final Particles: A Study of wa and no in Youth Conversationsmdpi.com
127. Distribution of sentence-final particles by gender and genre (F =...researchgate.net
128. Japanese first-person singular pronouns revisited: A semantic and cultural interpretationresearchgate.net
129. University of Kentucky University of Kentucky UKnowledge UKnowledgeuknowledge.uky.edu
130. Gender differences in Japaneseen.wikipedia.org
131. Dialectologia 27 (2021), 97-160. ISSN: 2013-2247 97 Received 24 January 2020.raco.cat
132. 16 Top Gen Z Japanese Slang and What They Meancotoacademy.com
133. Japanese Youth Slang 2026: 40+ Gen Z & Modern Japanese Words｜Why language workskodack.net
134. Modern Japanese Slang Terms You'll Actually Use in 2026migaku.com
135. Complete Guide to the 2,136 Joyo Kanji for Japanese Learnersmigaku.com
136. What are grades?kanjisho.com
137. KANJI-Link: Learn Japanese kanji by school gradekanji-link.com
138. Osaka-Literacy and Language Classes in Community Centers, Japan - Effective Literacy and Numeracy Practices Databaseuil.unesco.org
139. Adult literacy rateversus.com
140. Survey of Adults Skills 2023: Japanoecd.org
141. Japanese Program Expects Another Record for Students Taking Language Proficiency Testnews.uark.edu
142. How to Apply for 2026 JLPT Online: Step-by-Step Guidecotoacademy.com
143. JLPT (Japanese Language Proficiency Test) in 2026 (12/6/2026; Registration in August 2026) - The Japan Studies Program at LaGuardia Community Collegechrysanthemum.commons.gc.cuny.edu
144. JLPT Registration USA (2026): Dates, Locations & Guidenihongopath.com
145. Registration for the JLPT Japanese Language Proficiency Testinalco.fr
146. Japanese languageen.wikipedia.org
147. | Spread of the Japanese Languageworldmapper.org
148. Japanese as a foreign languageen.wikipedia.org
149. Where Is Japanese Spoken? 7+ Unexpected Places You’ll Hear Japaneselingopie.com
150. Japanese-speaking Populationthetranslationcompany.com
151. Japanese speaking countries - Languagesworlddata.info
152. The Island That Speaks Japanese - Japan Hub Notesjapanhubnotes.substack.com
153. Angauren.wikipedia.org
154. Palauen.wikipedia.org
155. Japan%E2%80%93Palau relationsen.wikipedia.org
156. Japanese settlement in Palauen.wikipedia.org
157. Learning “Japanese” versus “Kokugo” - by Kat Woodmankyattogallery.substack.com
158. Japanese Language Spread in Japan, Taiwan and Koreahasp.ub.uni-heidelberg.de
159. Full article: The construction of different educational ideologies for Korean students during the Japanese colonial eratandfonline.com
160. Taiwanese Girls' Education, 1897-1945: An Analysis of ...monash.edu
161. Encodings of Japanese - sci.lang.japan FAQsljfaq.org
162. Shift JIS — Grokipediagrokipedia.com
163. 100% found this document useful (2 votes)scribd.com
164. Method for segmenting non-segmented text using syntactic parseimage-ppubs.uspto.gov
165. Tokenization efficiency of current foundational large language models for the Ukrainian languagencbi.nlm.nih.gov
166. RakutenAI-7B: Extending Large Language Models for Japanesearxiv.org
167. Universal Word Segmentation: Implementation and Interpretationarxiv.org
168. Gendered Cultural Discourse in Japan across the Prewar-Postwar Transition: Evidence from Historical Word Embeddingsarxiv.org
169. Word Segmentation Standard in Chinese, Japanese and Korean — Lacunalacuna.tiptreesystems.com
170. A Probabilistic Algorithm for Segmenting Non-Kanji Japanese Strings — Lacunalacuna.tiptreesystems.com
171. Language Model Tokenizers Introduce Unfairness Between Languagesarxiv.org
172. Tokenization Tax 2026: LLM Tokenizer Comparisonellamind.com
173. Let’s Build the GPT Tokenizer: A Complete Guide to Tokenization in LLMsfast.ai
174. Arte da Lingoa de Iapamen.wikipedia.org
175. Nippo Jishoen.wikipedia.org
176. Motoori Norinaga - New World Encyclopedianewworldencyclopedia.org
177. Motoori Norinaga (1730–1801) - Routledge Encyclopedia of Philosophyrep.routledge.com
178. Universal dependencies: Japanesecjvlang.com
179. Shinkichi Hashimotoen.wikipedia.org
180. Japanese language studies in the Shōwa period : a guide to Japanese reference and research materials / edited by Joseph K. Yamagiwa.quod.lib.umich.edu
181. Motoki Tokiedaen.wikipedia.org
182. (PDF) Théories du langage au Japonresearchgate.net
183. Motoori Norinagaen.wikipedia.org
184. Basics about the Japanese people - Mediumtakahashi2saito0.medium.com
185. On the Japanese homelandlanguagesoftheworld.info
186. Dienekes’ Anthropology Blog: Dating the origin of Japanese languages with Bayesian phylogeneticsdienekes.blogspot.com
187. Triangulation supports agricultural spread of the Transeurasian languagesnature.com
188. Origin Theories of the Korean Languagelinguistics.byu.edu
189. Yayoi peopleen.wikipedia.org