Conjugable words (verbs, iâÂÂadjectives, and naâÂÂadjectives) are traditionally considered to have six possible . This article lists those from the , as well as the that was split off during the postâÂÂWWII spelling reforms.
The negative, apart from being plainly negative, can commonly express affirmation of the contrary but in an understated way, for example in rhetorical negative questions which imply affirmation, often for politeness, such as for softening hortatives.
The negative form is created by using the mizenkei base, followed by the suffix. This auxiliary arose in eastern dialects during Late Middle Japanese and displaced the western âÂÂn (see below) in standard Japanese, and while having been linked to another negative auxiliary, , it appears to be a grammaticalized version of the adjective .
The regular negative of aru would be aranai, but it is very rarely used, for example in , , etc. In Murakami Haruki's 2017 novel Killing Commendatore, the character "Commendatore", who is characterized as having "an odd way of speaking" that is "not the way ordinary people would speak", often, but not always, uses aranai in place of nai.
âÂÂN () is a negative auxiliary that was reduced from the earlier western and classical âÂÂnu (), but was largely displaced by the eastern âÂÂnai () in standard Japanese. âÂÂN(u) is still found in modern Japanese and thought of as "shortened" from âÂÂnai, although as the fact that is strictly dialectal compared to the standard shows, it is not really a "short" form. Some modern examples include , , . âÂÂN(u)s infinitive form, âÂÂzu, is still used in writing. is indeed shortened to especially by young people. âÂÂNai and âÂÂnu can be used in the same sentence where âÂÂnai is treated as conclusive and âÂÂnu as attributive: .
For the negatives of suru and its compounds, the general practice is to pair the eastern shiâ with the eastern âÂÂnai, and the western seâ with the western âÂÂn(u) and âÂÂzu for mainstream Japanese (only the eastern shinai is standard). Any of such combinations as sanai, sunai, shin(u), etc are considered dialectal. The writer Mori à Âgai, hailing from the western prefecture of Shimane, reportedly rigorously used senai, apparently in accordance with the classical (western) seâÂÂ, despite the dominance of sen in the west.
The current negative of âÂÂmasu is âÂÂmasen. The unique shapes of âÂÂmasen (rather than âÂÂmashinai) and the hortative âÂÂmashà  (rather than *âÂÂmashiyà Â) suggest their provenance in western polite speech. Compare . The easternized âÂÂmashinai was first recorded in its spoken form with vowel fusion as âÂÂmashinàin Edo Japanese. It has been attested in dialog from modern literature, for example in the speech of underclass, poor, peasant, servile or rural characters, even non-Japanese ones, notably the eye dialect of black slaves from the novel Gone with the Wind. The practice of using the stigmatized "unsophisticated" Tà Âhoku Japanese, in which âÂÂmashinai and âÂÂmashinàfigure, to translate the stereotypical speech of black Southern Americans, has become unpopular. The use of a low-class pseudo-dialect has been claimed to contribute to the churlish stereotype of Tà Âhoku Japanese, although it has also been argued that such use is increasingly perceived as mere fictional rural speech without necessarily any association with real-life dialects.
âÂÂN also happens to be a reduced form of âÂÂmu (), whose other reduced form, âÂÂu, is still used to make hortative forms (). âÂÂNu also happens to be an archaic perfective auxiliary, with a different conjugation from the negative âÂÂnu; it is equivalent to âÂÂta/âÂÂda in modern Japanese, and it is sometimes used elevatedly, as in .
The godan and ichidan-based negative endings âÂÂranai, âÂÂrinai or âÂÂrenai, specifically with the consonant r, can be reduced to âÂÂnnai and even âÂÂnnàin speech, as in , , .
The expressions comes from a special use of the subsidiary verb (see for examples).
The godan verb , whose kanji represents the Chinese word for "know", is often translated as "know", but a more accurate translation would be "learn", "find out", or "get to know". In order to say "I know", the construction is used instead (see for the usage of âÂÂte iru). This is because shiru is imbued with active recognition, which relates to the archaic meaning of taking physical command or possession of somebody or something, reflected by the spellings with the kanji for the Chinese word for "rule", "govern" or "control", and for "lead". To "know", therefore, is to take psychological command or possession of outside information, hence shiru ("I take command of this information"), and maintain it, hence shitte iru ("I took command of this information and it's still in my knowledge"). While "I know" ("I take command of it and still have it") is shitte iru, "I don't know" is actually , which is morphologically the negative of shiru, but semantically the negative of both shitte iru and shiru. On the other hand, implies perpetuation of ignorance, for example in . There have been analyses on precise cases where shitte inai is interchangeable with or even preferable to shiranai, for example when paired with shitte iru in the same clause, as in ; or in cases involving the perfect aspect, only shitte inai, not shiranai, can be used, as in , Nevertheless, the question remains as to why shiranai is uniquely more common than shitte inai as the negative of shitte iru to begin with, a phenomenon not yet observable in other verbs. A recorded conversation with a young child shows that shiranai is not inherently intuitive in and of itself: when his father asked him a question with papa [...] shiranai ("I don't know [...]"), he mirrored his prompt with shiranai ("I don't know"); yet, when his father asked him with shitte 'ru deshà  ("you know, don't you?"), he erroneously responded with *shitte 'ru nai ("I don't know", or more "accurately", shitte 'nai, though the expected response was still shiranai). In western dialects, there are , , , and .
Unlike verbs, adjectives do not have mizenkei, and instead their negatives are formed with their infinitives followed by the negative adjective nai, which is used to expand adjectives' negative conjugation similarly to how its opposite, the verb aru, is used to expand their affirmative conjugation. Since nai is of itself an adjective, it is possible to have a double negative, as in , , , etc.
The western negative auxiliaries âÂÂn(u) and âÂÂzu cannot attach directly to adjectival infinitives, and thus require the help of aru (which does have a mizenkei, araâÂÂ), hence âÂÂkaran(u) (â âÂÂku aran(u)), âÂÂkarazu (â âÂÂku arazu), âÂÂkarazaru (â âÂÂku arazaru), etc. These are primarily used in writing, seldom in speech.
The eastern negative âÂÂnai generally works like an adjective, although its âÂÂnai de is more common than the adjectival âÂÂnakute. The western gerund is âÂÂn(u) de.
The imperfect negative can be followed by the noun , whereby it means "while it hasn't happened" or "before it happens." Contradictorily, when it is followed by the nouns and , it can also mean the same thing, even though the literal meaning would suggest "before it doesn't happen." Thus, would mean the same as , and roughly as the affirmative which, however, may be used when the speaker has positive conviction that one will in fact come.
The classical, and thus more literary than colloquial, âÂÂzu is originally the of âÂÂn(u), but it is often used conclusively with a copula. It is often optionally followed by the copular infinitive particle ni as in âÂÂzu ni which works like a sort of adverbial gerund that means "without doing." It is possible to distinguish between coordination in and subordination in .
The passive or passive potential turns the patient (or target) or victim of an action into a subject, which can be marked with the nominative particle ga.
The pure passive simply expresses what act is done by the agent (A) to the patient (P), thus becomes . The agent is typically marked with ni, but kara can sometimes be used instead, especially if ni might be ambiguous; for example in , it is not clear which between A and B the agent is, so kara can be used instead as in .
The adversative, "victimizing" or affective passive expresses how a victim (V) is affected by, or suffers from, if the original verb is intransitive, the act being done by an agent, as in ; or, if the original verb is transitive, the act being done by an agent to a patient, as in . The terms adversative and victimizing are broadly correct, although the "victim" can sometimes be positively or even desirably affected by the action, so affective would be a more general term; for example, , , etc. Sometimes what seems to be positive may turn out to be negative with additional context, such as . English has some similar constructions with intransitive verbs, though not necessarily with a negative connotation, such as "I got rained on" or "I got talked to".
The passive can also have no passive meaning, but is merely a more honorific way to exalt the subject. Some verbs that come with the particle ni may be ambiguous when used this way, as it might not be clear whether it is in the passive or the honorific without context. For example, since the verb comes with ni whether or not it is in the passive, in the line , out of context, the king could be glad either if the prince is defeated by his princess (affective passive), or if he defeats her (subject exaltation); in context, it is the subject-exalting interpretation, hence "We would be delighted if ye defeat our princess."
Actual sentences may lack one of the said components (P, V or A) and therefore can become ambiguous (purely passive, adversatively passive or honorific) without additional context, for example . Intransitive verbs, such as and , do not have the pure passive meaning, only the other two, as in and .
Historically, the "passive" construction has also had potential use, but except for ichidan verbs and kuru, this use now has a dialectal or old-fashioned literary flavor. It has been argued that the various meanings of the passive arose from the spontaneous potential (see #Potential for more), which spawned both the regular potential and the regular passive, the latter of which came to be used additionally for subject exaltation. It has also been suggested that the pure passive came to be in its current state due to translations from the English passive.
The passive form is created by using the mizenkei base, followed by the suffix. For ichidan verbs and , the passive form and the potential form have an identical conjugation pattern with the same suffix. This makes it impossible to distinguish whether an ichidan verb adopts a passive or potential function without contextual information.
Arareru is historically attested with potential uses, but it is now primarily the more honorific way of saying and .
Honorific verbs can be made even more polite, as in , , etc. Excessively honorific verbs have been proscribed by textbooks, but they seem somewhat tolerable by speakers, even though they are still not as frequent with options without âÂÂreru. There are historical precedents of such double honorifics dating back to the Edo period.
Sareru is said to be shortened from serareru, the latter of which is said to be "pseudo-literary" (meant to emulate the writing style of classical Japanese). Shirareru is rare for Group-A verbs, and is not to be confused with the segmentally and accentually homophonous , which may also be spelt in hiragana. The three forms sareru, serareru and shirareru can sometimes be used interchangeably within the same text: .
A passive potential verb is an ichidan verb, and can be conjugated further mostly in the same way an ichidan verb can be, although its imperative is uncommon.
describes the possibility for an action to happen, whether it involves an intentional acting agent or not:
The patient of an action is customarily marked with the nominative particle ga (rather than the accusative o), which has the same appearance as the passive voice:
It is not impossible to use instead in the passive potential above, but it is less preferable in this case. However, o is more preferable to ga in the presence of a motion verb (whose pure passive does not naturally occur):
The rise of the particle ga as a marker of the patient of the action can be attributed to the , whereby the patient or the action simply comes about, seemingly in a vacuum, without a voluntary or intentional agent acting on it. This may also relate to the verb (see the phrase koto ga dekiru mentioned below) which means "come about; come into existence; materialize; form; take shape; be done/made/born/produced â be doable/feasible/possible", as in , , , etc. Thus, such phrase as , spoken in the past tense by an author who does not want to mention themself as an agent, can be translated as "the novel wrote itself" (= the novel came about in written form), rather than "I was able to write the novel". Other examples without voluntary agents include , , , , etc.
Potential verbs are unique to godan-based passive potential verbs, which have been said to have been shortened by removing ar from areru: â ; â . Other hypotheses of derivation include contractions from infinitives followed by , presumably as in â kakeru; or conversions of nidan attributives into ichidan potentials, such as , . Short potential verbs of this type are conventional in Tokyo Japanese, while long verbs have become largely obsolescent, elevated or non-Tokyo.
The hypothetical short potential verb * is not used. However, the non-potential intransitive , the adversative passive and causative are acceptable.
Non-godan-based verbs such as remain ambiguous. Such ambiguity can be resolved, at least colloquially, by a process dubbed , thus distinguishing the short for the potential and the long for the passive. This process was originally dialectal, but has been increasingly adopted by Tokyo speakers. Preference polls have shown that even among raâÂÂnuki kotoba users, the likelihood of usage significantly decreases as the mora count in the verb stem exceeds two; in other words, and are highly likely, but , , are much less likely.
Shortening passive verbs for potential uses is not universal among dialects, some of which may only use long verbs instead.
Apart from the dedicated potential verbs, the less ambiguous but more circumlocutory phrase can be used, as in , , etc. Depending on usage, a particle such as wa, mo or no can be used instead of ga. Some potential verbs, such as , , etc., have meanings similar to those of âÂÂable adjectives in English, such as drinkable/potable, eatable/edible, etc., which describe the patient's quality to be enjoyed by the agent, not the agent's capability of enjoying the patient; these verbs are not completely interchangeable with koto ga dekiru; compare and .
Most contemporary âÂÂsuru verbs do not really have underlying potential verbs and must use suru koto ga dekiru, which can be shortened to dekiru only for Group-A verbs, as in , , ; but , , etc. Historically, the passive potential sareru, serareru and shirareru were seldom used for Group-A verbs, but they have been supplanted by (suru koto ga) dekiru, which figured in a trend towards disambiguating the polysemous passive potential during the Meiji era. Non-Group-A verbs, which have become more like godan (partially or completely), do have potential verbs, such as (short)/ (long), /, etc; or more like ichidan, although probably without ra-nuki kotoba, as in , or with it, as in .
Writers can make use of various ways to convey potentiality, such as how Natsume SÃ Âseki used a short potential verb (aisezu), a long passive potential verb (irarenai), koto + particle + dekiru (hirogete dakishimeru koto no dekinai) and eru/uru (aishiuru) in the same sentence in the novel Kokoro: ; or how used koto + particle + dekiru (shinu koto no dekinai, nete iru koto no dekinai), and both the long passive potential (shinareru) and short potential (shinenai) of the same verb in his translation of Kierkegaard's Either/Or:
A potential verb is an ichidan verb, and can be conjugated further mostly in the same way an ichidan verb can be, although it may resist the imperative due to semantic difficulty.
expresses how an instigator causes an agent to do something, whether by making or letting the agent (A) do it. The instigator (I) is marked with the nominative ga, while the agent is customarily marked with either ni or o depending on the original verb's transitivity:
For some speakers, the agent can be marked with ni even with an intransitive verb, which is possibly intended to mean "I lets A come" rather than "I makes A come"; or marked with o even with a transitive verb, despite the presence of the other o marking the patient (P). This means that the presence of the particle o does not guarantee clear meaning, for example in , . In more elevated text, the agent can be marked with o shite instead of just o.
The causative form is created by using the mizenkei base, followed by the ichidan suffix. Colloquially, the shorter godan can be used, but this may cause confusion among verbs with virtually identical forms, especially those that happen to end in âÂÂsu already. Compare the causatives of the intransitive and its historically derived transitive , the latter of which happens to resemble the godan causative ugokasu, as well as their (passive) potentials:
In classical Japanese, âÂÂ(sa)su was the nidan ancestor of the modern ichidan âÂÂ(sa)seru; it became yodan sometime during Late Middle Japanese. A survey found that respondents were more likely to use âÂÂ(sa)su, which is now godan, over âÂÂ(sa)seru, if the original verb was already godan.
One of the negative forms of , , as in , has been attested. has also been used.
The causatives of honorific verbs do not seem to occur, although at least one author has artificially used in their literal translations of Amdo Tibetan honorific causatives.
has been said to be shortened from , or even .
The pseudo-classical causative makes use of instead of âÂÂ(sa)seru as shown above. It has ichidan conjugation, and it is meant to emulate the true classical causative with âÂÂshimu which has nidan conjugation. As a pseudo-classical auxiliary, âÂÂshimeru combines only with classical irrealis forms, which in most cases are not different from modern ones; but in the case of suru which has three irrealis forms, only the classical seâ is used as in seshimeru, not *sashimeru nor *shishimeru. There exists a causative form for the taru (â to aru, ) that is tarashimeru, as in .
A causative verb is either an ichidan or godan verb, and can be conjugated further in the same way those types of verbs can be.
After conjugating into the causative form, the verbs become ichidan verbs. They can therefore be further conjugated according to any ichidan pattern. For instance, a causative verb (e.g. ) can conjugate using the ichidan pattern for the to join sequential statements (), or the infinitive form to append the polite auxiliary verb ().
It has been suggested that the double causative is possible, as in , although it is doubtful whether speakers would find it comfortable.
The passive can be added on top of the causative to create the passivized causative:
The passivized causative results from the passive built on the causative, but not the opposite. It can also be used on auxiliaries that append to infinitives or gerunds, for example in , , although such examples, while grammatically plausible, is probably not used.
The hortative or volitional expresses the speaker's or speakers' personal or collective volition ("I think I will do it", "we think we will do it"), or invitation to others ("let's do it"), to do something. The same form, otherwise known as the tentative, conjectural or presumptive, expresses subjective speculation ("I think/presume that's the case") or supposition ("perhaps that's the case", "that could be the case"). The tentative meanings are increasingly outdated as later generations of speakers favor using darà Â, deshà Â, or de gozaimashà  as less ambiguous tentative markers (more in the subsection below).
The so-called "hortative form" is actually a combination of the mizenkei and the auxiliary , which underwent systemic historical sound changes that affected all the examples below. These sound changes motivated the term godan ("five grade"), which replaced yodan ("four grade") (see ). For non-godan verbs, during the final stage as shown in the table below, the vowels of the original mizenkei, i, e and o, were reintroduced in some dialects, especially eastern ones, which yielded the suffix âÂÂyà Â, as in iyà Â, eyà  and koyà Â. Other dialects, especially western ones, did not undergo this development. Verbs with the vowel i in their stems may retain the forms ending in âÂÂyà « (â âÂÂiu) in some dialects, hence okyu(u); or gained the normalized âÂÂyà  in other dialects, hence okiyo(o), o(k)kyo(o). Yet other dialects have iro(o), ero(o) and koro(o). For suru in particular, the eastern vowel i is used rather than the western e, hence shiyà  rather than seyà Â, the latter of which is found in some Chà «bu dialects. The distributions of forms where the original mizenkei vowels were fused with u (okyà «/okyà Â, akyà Â, nyà Â, kà Â, shà Â) and forms where the vowels were added back in (okiyà Â, akeyà Â, neyà Â, koyà Â, shiyà Â/seyà Â) are uneven among western dialects, where ichidan-based forms with fused vowels are less widespread; notably, in Kyoto and Osaka Prefectures, the former capitals with historically prestigious dialects, the fused forms kà  and shà  are used for the irregular verbs, while the forms with the original vowels are used for ichidan verbs. Compare the three hortatives, only the first of which is distinctly western, in the following example:
For âÂÂmasu and desu, the unique shapes of âÂÂmashà  and deshà Â, as well as of the negative âÂÂmasen, suggest their provenance in western polite speech. While âÂÂmasen undoubtedly has eastern counterparts, âÂÂmashinai/âÂÂmashinà(albeit uncommon, archaic and stereotypical), and do not seem to have any, in spite of such historical spellings as and , which appear to be pronounced *âÂÂmashiyà  and *deshiyà Â, but were rather variant spellings of and , and in some older publications these spellings were inconsistently interchangeable. In contrast, suru, âÂÂmasus conjugational relative, has both western and eastern forms, as in sen/shinai/shinÃÂ, and shà Â/shiyà Â.
Due to the said historical sound changes, all hortative/tentative forms contain the long vowel à Â, but it is susceptible to shortening into an o in speech, especially in dialects. In ichidan verbs, shortening results in the hortative/tentative and the imperative sharing the same segments (consonants and vowels), although they can still be distinguished by accent. All hortatives and tentatives are currently accentuated on the start of the à Â, as in , , , , etc. Thus the hortative ichidan , are accentually distinct from the imperative ichidan , . On the other hand, since all hortatives are accentuated on the à Â, ambiguity among themselves can arise, for example in and . The western hortative kà  ("let's come") also resembles the eastern imperative kà  ("come!").
The common form of the hortative/tentative ends in , but occasionally a classical (thus more elevated) alternative ending in or turns up in modern writing, for example in . Both and derive from the earlier , but through different mechanisms: is from the loss of a consonant in , ; while is from the loss of a vowel, . Since also happens to represent the western negative âÂÂn and the nominalizing particle no, the spelling represents either or , and represents not only or but also the noun phrase , the last of which contains the Sino-Japanese noun . Accentually, verbs whose dictionary forms are accentless can be distinguished, as in for "not go" and for "let's go"; but verbs whose dictionary forms are accented are ambiguous and context-dependent, as in () for both "not write" and "let's write." Some idioms derived from the classical hortative/tentative include , , , , etc.
Most verbs have hortative meanings, as in , although this can be interpreted as self-tentative ("I'll probably do it"). To express tentativity unambiguously, darà Â/jarà Â, de arà  or deshà  (or hyper-politely, de gozaimashà Â), which is unambiguously tentative, is added, as in . In some cases where the subject lacks human agency, the tentative meaning is more plausible, for example, means "it'll probably be cloudy" tentatively, not *"let's be cloudy" hortatively; means "it'll probably rain", not *"let's rain"; means "probably can" not *"let's be able". These have been increasingly replaced by , and . The rise of the darà Â/deshà Â-appended tentative, as in , as a distinction in form from the original polysemous tentative/hortative, is attested from the early 19th century; the use of the original tentative, as in , has become obsolescent or elevated.
Arà Â/arimashà  and their derivatives tend to be tentative, and can be replaced with aru darà  / aru deshà  / arimasu deshà Â. The same applies to de arà  / de arimashà Â, which can be replaced with de aru darà  / de aru deshà  / de arimasu deshà Â. Adjectival tentatives such as with a built-in arà  can be replaced with . The eastern adjectival negative tentative âÂÂnakarà Â, as in , and are to be replaced with . The past tentative âÂÂta/âÂÂda darà Â/deshà  are preferred to âÂÂtarà Â/âÂÂdarà Â, deshita deshà  to deshitarà Â, and âÂÂmashita deshà  to âÂÂmashitarà Â. However, âÂÂta/âÂÂda darà  can be shortened back to âÂÂta/âÂÂda 'rà Â, albeit with different accent patterns; compare for âÂÂtarà Â, and or for âÂÂta darà  â âÂÂta 'rà Â. This contraction is transferable to the polite âÂÂmashita 'rà Â, which is still not quite the same as âÂÂmashitarà Â, but which does suggest an unabbreviated form, âÂÂmashita darà Â, with a mismatch in politeness, compared to the well-formed polite âÂÂmashita deshà Â. doubted the plausibility of âÂÂmashita darà Â, but it is not impossible.
The western negative âÂÂn(u) and âÂÂzu take the classical âÂÂji for occasional elevated use, mostly followed by the quotative particle to, as in , as well as in some cliches such as , , , etc. The negative tentative/hortative has been expressed with the followed by ; in the particular case of âÂÂmasen, there is âÂÂmasu mai. In many non-Tokyo eastern dialects, mai (â majii) is the negative counterpart of the affirmative be(e) (â bei), the regularly modernized adjectival form of , which the Tokyo dialect does not use for tentative or hortative meanings. âÂÂN darà Â/jarà Â/de arà Â/deshà  and âÂÂmasen deshà  are also used.
The hortative can be quoted with the quotative particle to, with in particular often being used to unambiguously convey volition (as opposed to invitation to act together), similarly to the with âÂÂtai/âÂÂtagaru; compare and . The extender conveys imminent realization of volition, even for actions by inanimate agents.
The tentative can combine with ga or to (mo) to form the tentative concessive, with meanings similar to the gerundive concessive âÂÂte mo, the conditional concessive âÂÂedo and the imperative concessive ni shiro/seyo and de are.
The classical tentative is often followed by the particle bakari ("almost").