The Origin of な-Adjectives: History, Etymology, and Evolution

From nouns to adjectives: what な-adjectives really are

Category: History of Grammar

General overview of な-adjectives

In modern Japanese, adjectives are divided into two major categories: -i adjectives (い形容詞, i-keiyōshi) and -na adjectives (な形容詞, na-keiyōdōshi). The former end in the い sound in their base form and are “proper” adjectives, meaning they have their own independent inflection (they can conjugate to express past tense, negation, and so on), for example atsui 暑い (“hot”) or samui 寒い (“cold”).

The latter, on the other hand, keep an invariable form and require the morpheme “な” (-na) when they come before a noun. Common examples of -na adjectives are kirei 綺麗 (“beautiful, clean”) and yūmei 有名 (“famous”) – which become respectively kirei-na hana 綺麗な花 - (“a beautiful flower”) and yūmei-na haiyū 有名な俳優 - (“a famous actor”). In predicative position (that is, at the end of a sentence), these adjectives behave like nouns followed by the copula “to be” (da/desu): for example hana wa kirei desu その花はきれいです - (“that flower is beautiful”).

In essence, -na adjectives have a nominal nature: since they cannot conjugate on their own, they use forms of the verb “to be” as an auxiliary to express tense and negation (just as a noun would). In the following sections we will explore the origins of this adjectival category, the evolution of the morpheme -na from Old Japanese to modern Japanese, and the historical reasons for its development as a separate category from -i adjectives.

Historical origin of な-adjectives

The emergence of -na adjectives dates back to the classical period of the Japanese language. During the Heian period (794–1185), people felt the need to express qualities and concepts for which native adjectives (in -i) proved “insufficient.” The solution was to create new “nominal” adjectives from nouns by attaching to them a form of the verb “to be.” In particular, already in ancient texts we find constructions of the type [noun]+ にあり+ [noun], where ni ari literally means “to be in (a state).” For example, in the Man’yōshū (an eighth-century poetry anthology) we find the expression 「うつそみの人にある我れや」 - (utsusomi no hito ni aru ware ya) – literally “I, who am a person in the earthly world (that is, still alive)” – in which the nominal phrase utsusomi no hito (“a person of flesh and blood, a mortal”) is linked to ware (“I”) through ni ari. This use of にあり therefore served to connect a qualifying noun to the main noun, indicating a state or a quality. Later, this construction evolved into more compact forms.

During the Heian period, nouns carrying new descriptive qualities began to take the ending -nari (a fusion of ni + ari) or -tari (a fusion of to + ari) in order to form adjectival expressions. In practice, にあり first contracted into nari (terminal form) and naru (attributive form), creating the so-called “-nari adjectives,” and in parallel とあり became tari/taru, creating “-tari adjectives.” For example, the term shizuka 静か (“silence, quietness”), originally a noun, came to be used in phrases such as 犬は静かなり - (inu wa shizuka-nari, “the dog is quiet”) and 静かなる犬 - (shizuka-naru inu, “a quiet dog”). In the same way, a noun such as heizen 平然 - (“calmness, composure”) could yield 平然たり - (heizen-tari, “is calm”) and 平然たる犬 - (heizen-taru inu, “a calm dog”).

These -nari and -tari forms were the foundation of modern -na adjectives: in particular, adjectives with nari conjugation are the direct predecessors of modern -na adjectives, while those with tari conjugation have mostly disappeared or survive only in highly formal expressions (known as -taru adjectives).

Most classical -nari adjectives therefore merged into the -na adjectives of modern Japanese, while -tari forms disappeared or remain fossilized in literary constructions in -taru.

It should be noted that the introduction of -nari/-tari adjectives historically coincided with a lexical enrichment brought about by contact with Chinese civilization. Many descriptive words of Chinese origin (brought into Japan through kanji) were adopted and adapted as nominal adjectives during the Heian era.

In fact, -na adjectives have their roots both in native Japanese vocabulary and in Sino-Japanese vocabulary. On the one hand, there are -na adjectives formed from native Japanese words, often recognizable through hiragana suffixes such as -か (-ka) or -やか (-yaka) – for example shizuka 静か (“quiet, calm”) and nigiyaka 賑やか (“lively, animated”) derive from purely Japanese roots. On the other hand, a large number of -na adjectives come from Sino-Japanese terms (words written with kanji and originally imported from China), which did not fit into the inflectional system of -i adjectives and were therefore reworked as nominal adjectives.

Origin of the morpheme な and its evolution

As mentioned above, the morpheme na that characterizes these adjectives has a very specific historical origin. It descends from the ancient copula -nari (a contracted form of ni + ari, “to be in [a state]”). In classical Japanese, nari was the terminal (sentence-final) form of this copula, while -naru was its attributive form, used immediately before a noun. For example, in the Genji Monogatari and other classical works we can find phrases such as 静かなる家 - (shizuka-naru ie, “a quiet house”) or 花は綺麗なり - (hana wa kirei-nari, “the flowers are beautiful”).

With the evolution toward modern Japanese, the attributive form -naru gradually underwent phonetic erosion until it was reduced to -na. This development is historically documented: にあり ➜ なる ➜ な. Already in Middle Japanese (Muromachi period), examples begin to appear in which na is used on its own as an adjectival linker in place of naru.

Today, -na preserves exactly that attributive function: marking the adjective when it comes before a noun, taking over the role once carried by naru. In the same way, the predicative form だ (da) used with -na adjectives is the functional descendant of the old nari, although its etymological origin is slightly different (da most likely comes from de ar(i), a contraction of ni te ari, another copular construction that emerged in Middle Japanese).

In any case, both nari and de ari are related, since they are both derived from the verb ari (“to be”) combined with particles. The morpheme -na is therefore not simply a suffix without meaning, but the contracted form of an auxiliary verb. Functionally speaking, we can think of na as meaning something like “which is…,” linking the adjectival concept to the noun. For example, when we say 綺麗な人 (kirei-na hito), we are literally expressing “a person who is beautiful.” It is no coincidence that English-speaking linguists often refer to -na adjectives as “adjectival nouns” or “nominal adjectives”: the Japanese term keiyō-dōshi (“adjectival verb”) highlights precisely the fact that historically a verb (ari) was hidden behind that na. In modern Japanese, ari is no longer perceived as a separate element, but the presence of na and da still reveals the older construction underneath.

It is estimated that about two thirds of all -na adjectives are of Sino-Japanese origin, typically written entirely in kanji; examples include kirei 綺麗 (“beautiful, clean”), anzen 安全 (“safe”), and kenkō 健康 (“healthy”). In later periods, this mechanism remained productive: new adjectives entering Japanese from other languages – for example modern loanwords such as arudente アルデンテ (“al dente”) or yunikū ユニーク (“unique, original”) – are also treated as -na adjectives (often written in katakana), confirming that this category serves as an “entry point” for many descriptive foreign words.

Over time, the Japanese language gradually simplified the old -nari forms. Already in Middle Japanese (Kamakura–Edo period), the classical copula nari began to give way to newer copulas such as de arimasu (from which modern da developed) in spoken language. In modern Japanese, nari survives only indirectly: the attributive form has become the current morpheme “な” when the adjective comes before a noun, while the predicative form is expressed through da (plain form) or desu (polite form), which historically derive from the verb aru (“to be, to exist”).

For example, the classical adjective shizuka-naru - 静かなる (“quiet/calm” in attributive form) corresponds today to shizuka-na - 静かな, and shizuka-nari - 静かなり (“is calm”) has been replaced by shizuka da - 静かだ in contemporary Japanese. This etymological development is well illustrated by the term kirei 綺麗 (“beautiful, clean”): in a hypothetical backward reconstruction, the modern expression 綺麗な女 - (kirei-na onna, “a beautiful woman”) derives from the now old-fashioned literary form 綺麗なる女 - (kirei-naru onna), which in turn goes back to the original phrase 綺麗にある女 (kirei ni aru onna, literally “a woman who is in a state of beauty”).

Interestingly, the old form -naru is still occasionally used today in poetic or solemn contexts to give speech an archaic tone – for example 静かなる田舎 - (shizuka-naru inaka, “the quiet countryside”) deliberately uses naru instead of the normal na. In general, however, the distinction between -nari (predicative) and -naru (attributive) has disappeared from everyday usage, leaving behind the morpheme -na and the copula da as its direct heirs.

Why a category distinct from -i adjectives?

Why did Japanese develop a separate class of -na adjectives instead of incorporating new adjectives into the existing -i adjective system? The reasons are both historical and morphological:

• First of all, -i adjectives formed a relatively closed and mostly native group: they were the earliest adjectival type to appear in Japanese and derived almost exclusively from indigenous vocabulary, not from Chinese loanwords. This means that many concepts introduced later (especially through Chinese) had no already existing adjectival equivalent in -i. During the period of strongest Chinese influence (Heian), rather than altering the morphology of existing adjectives, Japanese preferred to rely on a syntactic construction: turning new nouns into descriptive elements through nari/ari, as described above. In other words, Sino-Japanese words expressing qualities or states (for example 静寂 (seijaku) - “silence”, 健康 (kenkō)- “health”, 安全 (anzen) - “safety”, were integrated not as i-keiyōshi but as nominal forms, linking them to the noun they modified through an auxiliary verb (“to be”) rather than through inflection. This led to the birth of a distinct grammatical category. We can imagine that if Old Japanese had already possessed -i adjectives for concepts such as “quiet” or “safe,” perhaps -na adjectives would not have spread so widely; but because such adjectives were lacking, the language filled that gap through a different mechanism, converting nouns into adjectives by means of nari/なり.

• Another factor is the linguistic structure of Sino-Japanese loanwords. These terms were often nominal by nature (in Chinese, root words do not change form depending on grammatical function). When they were adopted into Japanese, they retained this morphological stability and therefore required a linking element when used attributively. Japanese already had the particle no の to connect one qualifying noun to another noun (similar to “of” in English, e.g. Nihon no kotoba = “the language of Japan”).

However, using no would have kept the first term within the noun category. In our case, by adding -nari (and later -na), those nominal terms acquired an adjectival verbal structure and fully entered the system of predicates and adjectives. It is worth noting that the only syntactic difference between a common noun and a -na adjective lies precisely in this attributive marking: a noun qualifying another noun uses no, while a -na adjective uses na. For example, hito no kodomo 人の子供 (“a person’s child”) uses no, but yūmei na hito 有名な人 (“a famous person”) requires na.

This no/na distinction marks the boundary between a simple nominal compound and a true adjective. In the classical period, the use of nari specifically signaled that the first element, although etymologically a noun, was functioning as an adjective. Over time, this gave legitimacy to a separate class of “nominal” adjectives precisely because they represented a productive way to import new descriptive vocabulary without disrupting the native morphology of -i adjectives.

From a grammatical point of view, the existence of two adjectival categories reflects two different inflectional mechanisms: one synthetic and one analytic. -i adjectives use the synthetic inflection inherited from archaic Japanese (where adjectives had endings such as -ku, -shi, -ki, and so on); for example akai 赤い (“red”) can inflect as akaku (the -ku form, for example in akaku naru, “to become red”), akakatta (past form, “was red”), akaku nai (“is not red”), and so on.

-na adjectives, by contrast, follow an analytic model: they do not modify their own ending, but combine the nominal stem with the appropriate forms of the copula da/desu. For example, hen 変 - (“strange”) in adjectival function remains hen but requires “な” before a noun (henna hito 変な人, “a strange person”)* and “da” to state the quality (hito ga hen da 人が変だ, “the person is strange”). The various verbal forms of “to be” provide tense and negation: hen da 変だ (“is strange”), hen datta 変だった (“was strange”), hen dewa nai 変ではない (“is not strange”), hen dewa nakatta 変ではなかった (“was not strange”), and so on. In short, while -i adjectives conjugate the adjective itself, -na adjectives conjugate the verb “to be.”

This explains why in Japanese they are called 形容動詞 (keiyō-dōshi), literally “adjectival verbs”: historically they contained the root of the verb ari (“to be”) in the attributive form, and still today they rely on da/desu in order to function. One advantage of this dual system is the flexibility in the use of modifiers and sentence structure.

Because -na adjectives behave grammatically in a way similar to nouns, they can be preceded by degree adverbs (very, fairly, and so on), which is normally not possible with a pure noun. For example, it is correct to say kanari mujihi-na kōi - かなり無慈悲な行為 (“a rather merciless act”), where kanari かなり (“fairly, rather”) modifies mujihi-na 無慈悲な (“merciless”).

By contrast, with a common noun + da, a similar construction would sound unnatural. This shows that -na adjectives, although derived from nouns, function as full adjectives in every respect in modern syntax. In the same way, they can form adverbs with -ni (相当 ni): for example shizuka ni hanasu - 静かに話す – “to speak quietly” – where shizuka ni is the adverbial form of shizuka-na (“quiet, calm”).

-i adjectives form corresponding adverbs by changing -i to -ku (e.g. hayai 速い, “fast” → hayaku 速く, “quickly”), while -na adjectives add ni after the adjectival stem (e.g. shizuka 静か, “quiet” → shizuka-ni 静かに, “quietly”). This difference goes back to the copular origin of -na adjectives: -ni is none other than the original particle that linked the noun to the verb ari, preserved today as the marker of adverbial function.

Finally, it is interesting to note that from a linguistic point of view the boundary between -na adjectives and nouns remains blurred. For some scholars, -na adjectives are simply nouns followed by the copula, and therefore should not be considered a category of their own. For example, the Japanese dictionary Kōjien does not list “形容動詞” as an independent part of speech, treating words such as seiren 清廉 (“integrity, [moral] purity”) or haran-banjo 波乱万丈 (“full of dramatic events”) directly as substantivized nouns rather than adjectives. In the past it was in fact common to use such terms with no instead of na – for example 清廉の人 (“a person of integrity”) – and only later did the adjectival form 清廉な人 (“an upright person”) become established.

Nevertheless, Japanese school grammar (derived from the studies of Hashimoto Shinkichi and other twentieth-century linguists) continues to formally distinguish -na adjectives as a separate part of speech, precisely because they display a fixed conjugational pattern (the verb “to be”) that sets them apart from pure nouns. In practice, for teaching and descriptive purposes, it is considered useful to separate -na adjectives from nouns, highlighting their attributive and predicative functions analogous to those of -i adjectives, while still recognizing that structurally they remain linked to nominal forms.

In conclusion, the emergence of a distinct category of -na adjectives depended both on the need to incorporate new descriptive concepts (often of external origin) and on the nominal nature of those concepts, which encouraged a grammatical treatment different from native adjectives. This led Japanese to develop two parallel ways of expressing qualities: one through inflected adjectives (-i) and one through nominal adjectives (-na), each with its own characteristics but cooperating to give the language greater expressive richness.

To summarize this historical development, we can break down one example:

• Ancient (8th century): 綺麗にあり女 – kirei ni ari onna (“a woman who exists in [a state of] beauty”)

• Classical (10th century): 綺麗なる女 – kirei naru onna (“a beautiful woman,” with attributive naru)


• Modern: 綺麗な女 – kirei na onna (“a beautiful woman,” the current form)


As we can see, にあり → なる → な: the na morpheme used today is the direct heir of that chain. Modern na is now fully grammaticalized, meaning that speakers no longer perceive it as an independent word (“to be”), but as an integral part of the adjective.

Even so, its origin explains many of its peculiar features: for example why -na adjectives use に to form adverbs (静かに “quietly”) – this reflects the old ni ari – or why they require da to assert attribution. Even the name of the category, 形容動詞 – “adjectival verb” – becomes more meaningful once we know that na derives from an auxiliary verb. Ultimately, that small -na carries with it a piece of the history of the Japanese language: it is the visible trace of how, more than a thousand years ago, Japanese speakers ingeniously fused nouns and verbs in order to create new adjectives and expand their expressive range.

Sources

• Shogakukan Kokugo Dai Jiten (辞典) and other classical Japanese dictionaries
• Man’yōshū (万葉集), poem 165
• 日本語文法書 (Japanese grammar references)
• Wikipedia (Japanese edition), article on 形容動詞
• Wikipedia (English edition), Japanese adjectives
• Nipponikai.it – Japanese adjectives: simple explanation, grammar rules, and examples
• StackExchange Japanese, Historical precursor to な? (discussion on the historical development of -na adjectives)