Talk:Japanese conjugation/Archive 3

Archive 1

Archive 2

Archive 3

To-do: Negative progressive & Conjunctive as a conjunction

In the article are two blocks of commented-out code where I've started on adding relevant information.

Specifically, a small subsection of the Negative needs to be added to explain the negative progressive tense, since it's the only other inflectional suffix available to the mizenkei base.

Furthermore, the conjunctive form hasn't explained how it literally behaves as a conjunction yet. This needs to be explained, particularly the nuance in how it differs from the -te form.

I will probably get around to completing this, but I do invite the community to contribute if they're willing.

Cheers — JKVeganAbroad (talk) 15:04, 1 July 2021 (UTC)

DONE — JKVeganAbroad (talk) 11:12, 4 July 2021 (UTC)

The dictionary form is not an infinitive

This article keeps calling the dictionary form "infinitive" even though it is a finite verb form. It has all the grammatical information that Japanese verbs routinely conjugate for and it can head simple declarative sentences. Stibitzki (talk) 13:48, 6 July 2021 (UTC)

The definition of "infinitive", according to the Oxford dictionary, is "the basic form of a verb, without an inflection binding it to a particular subject or tense.". The dictionary form would appear to fit that definition, since it's neither bound to the present nor future tense. Furthermore, it has no inflection, and is the basic form of a verb.

As for the Wikipedia page you linked to on the "finite verb", it states: "Under this newer conception, finite verbs are often the locus of grammatical information of gender, person, number, tense, aspect, mood, and voice." None of this grammatical information pertains to the dictionary form, so it would appear to fail the definition of a "finite verb".

Furthermore, the Wikipedia page on Infinitives states: "Like other non-finite verb forms (like participles, converbs, gerunds and gerundives), infinitives do not generally have an expressed subject… Unlike finite verbs, infinitives are not usually inflected for tense, person, etc". This further appears to match the phenomena of the dictionary form.

I don't intend to argue, but I would like clarity why you believe the dictionary form fails the definition of "infinitive". This is important if you want to redefine the terminology used in the article.

Regarding your intent to change the phrase "infinitive" to "dictionary form" in all of the verb conjugation tables, I'm neutral on this part of your proposed edit (so if you were to change those things specifically, I wouldn't object). — JKVeganAbroad (talk) 14:15, 6 July 2021 (UTC)

Of course it isn't bound to the present or future tense, since neither of those exist in Japanese. It is bound to the non-past tense. And of course it isn't bound to person or number either because Japanese verbs don't inflect for those. But it is bound to and the locus for all the grammatical information that Japanese verbs inflect for, being the plain positive non-past form. And as per the second definition given in the article on finite verbs, it can act as the head of simple declarative sentences such as 彼は本を読む ("He reads books") and undisputably has a subject there. What dictionaries pick as their base form is arbitrary and has no bearing on whether a word is an infinitive, which is based on morphological and syntactic characteristics (and also some subjectivity on the part of the grammar writer, since languages are messy and not every aspect neatly fits into preconceived categories). Stibitzki (talk) 15:42, 6 July 2021 (UTC)

I haven't understood your explanation properly. "It's bound to the non-past" How is it "bound" exactly? The dictionary form can imply past behaviour, so it's not properly bound to the "non-past". e.g. 「果物を食べる」 ("I eat fruit") is an example where a past habit is strongly implied. But it could also mean "I will eat the fruit" (eventually) or "I will eat the fruit" (right now). This form isn't bound to any tense.

Can you give more concrete examples to explain? Sorry about the trouble. — JKVeganAbroad (talk) 16:08, 6 July 2021 (UTC)

Japanese has two morphological tenses, past and non-past. What they're called and what specific scenarios they're used in doesn't really matter, you could also just call them Tense I and Tense II. By "bound to the non-past tense" I just mean that it "is in" the non-past tense (or Tense I). The dictionary form is also bound to/expresses/contains all the grammatical information that any Japanese verb can express, thus making it a finite verb form. (Grammatical categories like person and number that simply don't exist in Japanese are not applicable.) Stibitzki (talk) 16:37, 6 July 2021 (UTC)

This is no different to English, which has no future tense like Japanese. English has infinitives (e.g. "to eat") which can be inflected to the past tense ("ate") in the same way Japanese has the infinitive 食べる which can be inflected (albeit with the 連用形 base) to the past tense 食べた. Can you, explain, with comparative examples, how the dictionary form is not fitting the English definition of "infinitive"? — JKVeganAbroad (talk) 05:59, 7 July 2021 (UTC)

I agree with changing infinitive to dictionary form. Looking at the literature (a book on Japanese linguistics by Shoichi Iwasaki and different scholarly articles), infinitive refers to the 連用形, so it should be changed per WP:VERIFY. TranquilHope (talk) 17:24, 6 July 2021 (UTC)

Calling the 連用形 (masu stem, for other readers) an infinitive seems contrary to the definition of infinitive, because the masu-stem is a gerund (a noun form of a verb). As a noun, it doesn't fit the definition of "basic form of a verb", because it's a noun. I'd call that a misnomer in literature, I believe.

I'm okay with changing the terminology in the article to use "dictionary form", since this isn't controversial. However, removing this line of text: "In Japanese, the imperfective form is also the infinitive (the basic form of a verb) and is used as the headword or lemma, without using any conjugations." is the main reason for this discussion. — JKVeganAbroad (talk) 01:12, 7 July 2021 (UTC)

Do you have any sources that support what you are saying to avoid WP:ORIGINAL?　TranquilHope (talk) 03:57, 7 July 2021 (UTC)

Iwasaki says, "Imperfective aspect in Japanese is divided into the two major types of "progressive" and "resultative." These aspectual meanings are signaled by the interaction between verb semantics and the morphological structure, -te-iru/-aru." TranquilHope (talk) 04:05, 7 July 2021 (UTC)

You've provided a quote without interpreting it. how does that quote clarify this discussion? As for the accusations about "verify" and "original research", if aligning with or contradicting the English dictionary's definition is not enough, I'll try to find some research that references the matter in a non-contradictory way (i.e. whete nouns are not infinitives). — JKVeganAbroad (talk) 05:59, 7 July 2021 (UTC)

The quote is what I see what imperfective is in reference to in Japanese—teiru and tearu and not the dictionary form. You need to find a source that supports the statement "the imperfective form is also the infinitive (the basic form of a verb)." TranquilHope (talk) 06:51, 7 July 2021 (UTC)

You can see why the continuative would be called infinitive when you look at a sentence like 彼らは泳ぎ始めた ("They began to swim"). "泳ぎ" just like "to swim" carries no inflectional information, that is all on "始めた" or "began" which are also the heads of their respective sentences. Just because a form is used as a gerund in some contexts doesn't mean it can't be an infinitive (or other word form) in other contexts. In English, "swimming" being a gerund in "swimming is fun" doesn't mean that it isn't a participle in "they are swimming". I also find it debatable whether the continuative form should be called a gerund since it isn't just straightforward nominalization as with の or こと. Stibitzki (talk) 08:28, 7 July 2021 (UTC)

References of the "infinitive" with regards to the dictionary form was part of the original contents of the article before it was extensively overhauled. I retained it in the article because I couldn't find contradictory evidence between the definitions of the word when comparing it to the thing it was labeling (i.e. the dictionary form).

To be honest, I'm still not convinced that the definitions of "infinitive" are exclusive of the Japanese dictionary form. However, at least 2 other Wikipedia authors disagree on this. Furthermore, I cannot find literature that links this word to the dictionary form. I've checked through many resources, not a single mention. This shouldn't be ignored: if literature aren't linking this word to this Japanese pattern, perhaps there's a valid explanation. I must concede that, somehow, what constitutes an "infinitive" is more nuanced than I can understand.

I'm no longer opposed to removing the terminology from the article. I'd like to thank Stibitzki and TranquilHope for their contributions to this discussion. @Stibitzki, please pardon me for intervening with your proposed edits. It would be good if you could re-apply them. Sorry for the inconvenience. — JKVeganAbroad (talk) 16:10, 7 July 2021 (UTC)

Thanks. No big deal. ;-) Stibitzki (talk) 16:19, 7 July 2021 (UTC)

Mistake in the text

It says "the shūshikei/rentaikei and meireikei bases do not conjugate with any inflectional suffixes." However, the shūshikei/rentaikei conjugate with べき/べし (must) and な (negative imperative). 193.198.8.213 (talk) 20:31, 6 July 2021 (UTC)

Those arent conjugations of inflectional suffixes, those are auxiliaries. — JKVeganAbroad (talk) 02:27, 7 July 2021 (UTC)

Kanōkei base

I don't understand the recent edits to the verb base formation table in regards to the kanōkei base.
When it got grouped under the kateikei base instead of the mizenkei base, the base formation for the kanōkei base turns out to be identical to the kateikei base with the exception of する. Later, it was changed such that the kanōkei base of 来る is こ instead of くれ because the potential form of 来る is こられる and not *くれられる or *くれる. And then it got changed again, with the logic being that there is no kanōkei base for 来る and that its potential form uses the mizenkei base. However, won't this logic also apply to ichidan verbs? After all, the potential form of ichidan verbs also matches with te passive form.
When I had worked with JKVeganAbroad on the verb table, I initially planned on considering the potential form of verbs as using the kateikei base for godan verbs and mizenkei base for ichidan verbs and 来る. The kanōkei base later came into discussion and it was decided as a solution for a unified verb base that that potential form uses. But now that idea appears to be rejected, so I would like to know what's gonna be going on with the kanōkei base heading forward. Should we abandon the kanōkei base and say that the potential form uses the kateikei base for some verbs and mizenkei base for the other, or should we continue with the kanōkei base as the sole verb base that the potential form uses? Because otherwise, there is no point in considering the kanōkei base if it results in the potential form still being considered as using more than one verb base. Excusememoi (talk) 00:56, 9 July 2021 (UTC)

@Excusememoi: Various points here, let me try to address them all.

Nota bene: Below, I ignore the syntactic (sentence-structure) definition for "transitivity" that usually applies to English, for example -- "I eat." is regarded as an intransitive statement, because the syntax here lacks any explicitly stated object. "Transitivity" in the context of Japanese grammar is instead usually treated as a property of the semantics (meaning) of a verb. For example, 「私は食べる。」 has no stated, syntactic object, but the verb is still regarded as a 他動詞 (literally "other-acting word", often glossed as "transitive") because the action is regarded as inherently requiring an object, whether stated or omitted.

• The initial grouping of the kanōkei under the mizenkei was probably based on a misunderstanding of the Miyake paper included there as a reference. That paper makes a strong case for the kanōkei not having any direct relationship at all with the mizenkei, and instead evolving from a shift in valency wherein a yodan transitive verb is used intransitively, with a concurrent change in conjugation pattern to nidan, as required to make this valency change in sense clear and to distinguish the transitive and intransitive uses.

One example given is the transitive verb 知る (shiru, "to know something; to come into the knowledge of something"), which developed an intransitive stative use somewhat similar to the passive "to be known", but where certain usage indicated more of a potential sense "to be capable of knowing or being known". The intransitive sense was made clear by applying the nidan conjugation pattern -- a conjugation distinction still seen in modern Japanese, where various verbs of the godan paradigm have ichidan counterparts of opposite valence, such as 付く (tsuku, "to attach to something", intransitive, godan) and 付ける (tsukeru, "to attach something to something else", transitive, ichidan), or 開く (aku, "to open", intransitive, godan) and 開ける (akeru, "to open something", transitive, ichidan).

• The verb 来る (kuru, "to come") has no dedicated kanōkei, and indeed the kanōkei is only found in godan transitive verbs. This should probably be explicitly explained somewhere.

Niggle: Some grammars might describe the so-called "ra-nuki" form as a kind of kanōkei. Semantically, that's exactly what it is. For instance, 食べられる (taberareru) is both the passive and the potential for taberu ("to eat"). Some speakers (in my, admittedly limited, direct experience, more common in the Tōhoku region than in Tokyo or further southwest) drop the -ra- portion of the passive ending to create the "ra-nuki" (literally, "ra-removed") form 食べれる (tabereru), which is unambiguously used only to express the potential, and not the passive.

However, in most contexts (that I'm aware of, anyway), such as the context of the Miyake paper, the term 可能形 (kanōkei) is used to refer exclusively to the intransitive ichidan counterparts to godan transitive verbs, where the intransitive forms are used to express potential.

• I confess I don't understand the intended point of the sentence, "The kanōkei base later came into discussion and it was decided as a solution for a unified verb base that that potential form uses."

Morphophonemically, and in terms of historical sound shifts, the modern kanōkei cannot be from the mizenkei. Consider 読む (yomu, "to read", transitive, godan). The mizenkei for this verb has always been yoma-. The potential for this verb first appeared somewhere around the Muromachi period, with the 下二段活用 (shimo nidan katsuyō, lower bigrade conjugation) pattern, with a terminal form of yomu, an attributive of yomuru, and with all other conjugation bases manifesting as yome-. There is no clear mechanism for yoma- to become yome-: this kind of /a/ → /e/ sound shift in historical Japanese only occurs when there is some kind of vowel fusion going on, such as すごい (sugoi, "awesome, terrific; awful, terrible") → すげえ (sugee, slang, emphatic), or 高い (takai, "high, tall; expensive") → 高え (takee, slang, emphatic) -- but there is no following vowel for either verb base, and no possibility of vowel fusion.

• Regarding whether to include the kanōkei at all, I think it's useful to include, provided that we also clearly explain the constraints on this -- it only exists for transitive godan verbs, and since it is expressed using the ichidan conjugation pattern, the various conjugation forms are identical to those for any other ichidan verbs in general.

I'm not sure what you mean in your mention that "the potential form uses the kateikei base for some verbs and mizenkei base for the other" -- there is no dedicated kanōkei for any verbs other than godan transitive verbs. What other verbs might you be thinking about? If you're talking about the passive / spontaneous / potential form like される (sareru, "to be done") (since it's really all three at once), that's a different phenomenon. Modern godan verbs can be conjugated into the passive / spontaneous / potential form as well. 読まれる (yomareru), depending on context, could mean "to be read", or "able to read", or it might even be just the politer form using the mechanism of indirection that is commonly applied in pretty much all languages around the world.

• Regarding the overall structure of the table, and whether the kanōkei belongs as a "sub-base" of the kateikei, I'm certainly open to other suggestions. Phonologically, the kanōkei is most similar to the kateikei, and that is largely why I moved the kanōkei into its current position, but that is also not a very strong argument in favor of the current structure. Derivationally, the kanōkei doesn't have much to do with the kateikei: as described in the Miyake paper, the kanōkei was a valency shift + conjugational paradigm shift in yodan transitive verbs, starting from the shūshikei lemma form. (There might be an argument to be made that the realis mood expressed in the Old Japanese 已然形 (izenkei, realis form) that became the modern kateikei has a semantic role to play in the stative sense expressed in the modern kanōkei, but that is a potentially very deep rabbit hole, and probably not very germane here.) Moreover, the kateikei is much older, already in evidence in the Man'yōshū of 759, whereas the kanōkei doesn't emerge until the late Muromachi period, I think more specifically around the 1500s.

There's certainly no lack of things to think about here.  :) I hope the above is at least somewhat useful. ‑‑ Eiríkr Útlendi │^{Tala við mig} 22:09, 9 July 2021 (UTC)

You've really provided quite comprehensive knowledge on this matter, and although I can't really wrap my head around the idea that the kanōkei is only used in transitive godan verbs, I guess the history of it makes sense. But now that gets me thinking about the ishikei and onbinkei, which are also only there because they only differ from regular mizenkei and ren'yōkei (respectively) in godan verbs. Would you say that ichidan and irregular verbs lack the ishikei and onbinkei and that the te form of these verbs uses the ren'yokei and the volitional form uses the mizenkei? If so, I think it would be better to make a separate table if we're going with the idea that the derivative bases are only applicable to godan verbs.

To clarify, the logic that the article was going with until recently is that, although these derivative bases only affect godan verbs, they can be projected onto every verb. This is so that for example, one can say that the volitional form uses the ishikei for every verb, even though the ishikei only differs from the mizenkei in godan verbs. Going with this logic is advantageous because of two reasons:

• Every conjugated verb form can associated with a single verb base (e.g volitional form: ishikei, as opposed to volitional form: ishikei for godan verbs and mizenkei for other verbs);
• Every verb can be considered to carry a full set of nine verb bases, with minor exceptions such as する whose potential form is a suppletive verb and hence lack a kanōkei.

But now you've provided your point that this cannot the case for the kanōkei, I'm afraid this logic can no longer apply, especially when the other two derivative bases are taken into consideration. I would definitely like to know what you (and perhaps others) think about this. Excusememoi (talk) 03:01, 10 July 2021 (UTC)

Since I was mentioned, I just want to explain that I researched for quite some time trying to find the origins of the kanōkei base. I was unable to find literary analysis in English however I did read through and find answers in Japanese resources: Nakano 2008, and Miyake 2016. They both mentioned theorised links to the mizenkei base (Miyake 2016 Section 2A, Nakano 2008 Section 2.2). It also seems unlikely that the potential form of the irregular verb 来る happens to coincide with the unique pattern derived from the mizenkei base (rather than coincidence, it's probably derivative). It all seems to make sense that the kanōkei is linked to the mizenkei form both through the consistent pattern of the ichidan verbs and the irregular 来る verb, but it's also a documented observation in literature.

Having said that, I am not a professional translator of the Japanese language, whilst Eiríkr Útlendi is (it's mentioned in their user page). I'm delighted that they were able to notice my misinterpretation of Miyake's conclusion, and this is a wonderful testament to the importance of including citations for verifiability in general. So I guess the 来る to こ resembling the mizenkei base really was a peculiar coincidence after all.

However, I do believe the table of bases should include 来る → こ, at least in parenthesis with a footnote, since its absence from the table might imply the potential form doesn't exist for 来る. The article should add clarity, not confusion. I've already changed the article to reflect this, I hope this satisfies all parties. — JKVeganAbroad (talk) 15:33, 12 July 2021 (UTC)

@JKVeganAbroad:, as described above, there is no dedicated potential form for 来る -- there is the passive / potential form korareru, which is used to express both passive and potential, and there is the passive-derived "ra-nuki" version koreru, which is used in certain social contexts to express explicitly the potential (and not the passive). This "ra-nuki" conjugation form is a relatively recent innovation, arising around the late 1800s from what I can find (such as the Kokugo Dai Jiten (KDJ) entry here, in Japanese; scroll down to the こ・れる【来】 section for an initial citation of 1900).

I am not familiar with the specific Nakano and Miyake texts you mention -- I'm guessing that "Miyake 2016" is Toshihiro Miyake's 可能動詞の成立 (Kanō Dōshi no Seiritsu, "The Formation of Potential Verbs") paper, such as this PDF copy on J-Store? Section 2A there does mention the mizenkei, as Miyake lays out the three theories of how potential verbs arose:

• Theory A is the mizenkei stem of a verb (ending in -a for godan verbs) + the -reru ending, as previously written about by Yamada in 1936, Yuzawa in 1936, and Fukuda in 1996, among others
• Theory B is the ren'yōkei stem of a verb (ending in -i for godan verbs) + auxiliary verb 得る ("possible to", either uru or eru depending on grammatical context), as previously written about by Shibutani in 1993
• Theory C is the development of intransitive nidan conjugation forms from yodan transitive verbs, as previously written about by 坂梨 (Sakanashi? Sakari?) in 1969 and 2006, Yamada in 2001, Aoki in 1996 and 2010, among others

Theory C is where Miyake lands in his analysis.

I'm not sure what Nakano publication you might be pointing to. Do you have a link?

Regarding 来る (kuru), is it possible that the "potential form" at issue for this verb is indeed korareru or derivative koreru? These are the only conjugations that both are potential and include the ko- beginning. The passive for any verb is also a potential, albeit not the same thing as the "potential verb" described in the Miyake paper -- for example, dedicated potential form 知れる (shireru, "to be able to know") is composed via the transitive-to-intransitive shift and consequent change from yodan to nidan conjugation patterns (or modern godan to ichidan), while combined passive / potential form 知られる (shirareru, "to be known; to be able to know") is composed from the mizenkei base shira- + passive ending -reru.

@Excusememoi:, I realized I was incorrect above to state that the potential only exists for transitive godan verbs -- if Miyake's paper is correct, it arose first with transitive godan verbs, but it clearly exists for some intransitive godan verbs in modern usage, such as 行ける (ikeru, "to be able to go; to be able to be done; to go well"), first cited to 1753 here in the KDJ entry. Similarly, we can have valid forms like 歩ける (arukeru, "to be able to walk"), 座れる (suwareru, "to be able to sit"), 帰れる (kaereru, "to be able to go back, to be able to return"), from intransitive godan verbs. However, there are some intransitive godan verbs that cannot form similar potentials, such as 立つ (tatsu, "to stand"), or 付く (tsuku, "to stick to something"), due to the existence of transitive forms ending in -eru: 立てる (tateru, "to stand something up"), 付ける (tsukeru, "to stick something to something else").

Cheers, ‑‑ Eiríkr Útlendi │^{Tala við mig} 18:42, 12 July 2021 (UTC)

Re-reading the thread, I realize there's room for confusion in what we call things. Here's my take on this:

1. We have a specific kind of potential expressed by an ichidan verb ending in -eru, that is a counterpart to a godan verb.
   • Formed only for godan verbs. Examples: 行ける (ikeru, "able to go", from 行く iku, "to go"), 知れる (shireru, "able to know", from 知る shiru, "to know something"), 正せる (tadaseru, "able to correct", from 正す tadasu, "to correct something").
   • As an ichidan verb, the verb base is the same for all conjugation forms, ending in -e-.
   • These are the verb types described as "potential verbs" in Miyake's paper.
2. We have the multipurpose passive / potential / polite expressed using the -(r)areru ending.
   • Formed for any verb type. Examples: 食べられる (taberareru, "to be eaten, to be able to eat", from 食べる taberu, "to eat"), 行かれる (ikareru, "to go [polite], to be able to go", from 行く iku, "to go"), 来られる (korareru, "to come [polite], to be able to come", from 来る kuru, "to come").
   • This attaches to the mizenkei. For godan verbs, this attaches to the mizenkei verb base ending in -a-.
   • For irregular verb 来る (kuru), this attaches to the mizenkei verb base ending in -o-.
   • For irregular verb する (suru), this forces a shift to the mizenkei verb base ending in -a-.

     Nota bene: Historically, the mizenkei for する (suru) was consistently せ (se-), and the passive for this is also recorded earlier as that せ verb base + らる (predecessor to modern られる), producing せらる. Modern される may have developed through influence from regular godan verbs ending in す, which would have changed to the regular godan mizenkei ending sa-, to which the passive -ru (later -reru) then attached.

3. We have the potential expressed using the so-called "ra-nuki" -reru ending. Derived from the passive.
   • Formed only for ichidan verbs and 来る. Examples: 食べれる (tabereru, "able to eat"), 広めれる (hiromereru, "able to widen"), 来れる (koreru, "able to come"). JA Wikipedia content at ja:日本語の乱れ#ら抜き言葉.
   • Treated as "proscribed" or "slang" by some grammarians; however, this pattern has been around for centuries. Despite my earlier subjective take that this pattern is more common in the northeast, it is apparently evidenced even longer ago in the southwest (per the JA WP link just above).

For the table at Japanese_verb_conjugation#Verb_bases, I'm not sure how best to proceed. The Miyake-style "potential verbs" and the "ra-nuki" verbs appear similar, but actually have zero overlap -- I don't think there is any verb that can fit both categories, since the Miyake verbs develop from a godan root, while the "ra-nuki" verbs develop from ichidan verbs and 来る. And the above only applies to the kanōkei in the table.

I note too that the onbinkei shown now is missing details. There are two three kinds of 音便 (onbin, "euphony") that happen for the ren'yōkei -- mora vowel deletion plus gemination of the following /t/, or deletion of the preceding /k/, or vowel deletion and shift to a nasal with voicing of the following /t/.

1. T-doubling: 作る (tsukuru) → 作り (tsukuri-) → 作って (tsukutte) // 行う (okonau) → 行い (okonai-) → 行って (okonatte)
   (Likewise for past-tense ending -ta: tsukutta, etc.)
2. K-deletion: 聞く (kiku) → 聞き (kiki-) → 聞いて (kiite) // 乾く (kawaku) → 乾き (kawaki) → 乾いて (kawaite)
3. Nasal shift: 読む (yomu) → 読み (yomi-) → 読んで (yonde) // 転ぶ (korobu) → 転び (korobi) → 転んで (koronde)

There's also a few exceptions, such as 行く (iku) → 行き (iki-) → 行って (itte) // 乞う (kou) → 乞い (koi) → 乞うて (koute).

For the volitional, all of the ichidan verbs take よう (-yō) on the end. This evolved from the regular ichidan verb stem + suppositional / volitional suffix む, which caused sound shifts. 見る (miru) → 見 (mi-) + む (mu) = 見む (mimu), phonologically shifting to mĩu → mʲoː, with the palatal glide + vowel then reanalyzed as a suffix attached back onto the regular verb stem, as 見 (mi-) + よう (yō). This historical process is described in Japanese in this KDJ entry at Kotobank, among other places. Historical sound changes explain the otherwise-odd modern form しよう (shiyō): the mizenkei base for する (suru) was se-, giving us se- + -mu = semu. That said, the consonant in the せ kana was previously a fricative, so せむ was actually pronounced more like shemu. Given the same sound change as above for む (mu), shemu became sheu, shifting to shʲō, and then that palatal glide + vowel was again reanalyzed as a suffix, giving us shi- + yō.

Apologies for the length.  :) To sum up:

Suggestion: The tables at ja:五段活用 might be a useful comparison. The kanōkei (the Miyake-style "potential verbs" ending in -eru) are actually treated as entirely separate verbs by monolingual Japanese grammars, with their own independent ichidan verb bases, and thus these are omitted from listings of verb base forms.

Thanks for the appended clarification!

"…I am not familiar with the specific Nakano and Miyake texts you mention…" — I assumed you might have remembered them, since you made a mention to the Miyake cited in the article in the comments of this edit. Here's Nakano 2008 and here's Miyake 2016 for reference. As for the missing details you mentioned, since they apply to the "sub-bases", perhaps this information can be included in the deeper explanations within the Japanese verb conjugation § Derivative verb bases section. Excessive footnotes can be overwhelming, and I think more than 4 footnotes is borderline too many so we shouldn't add a 6th one. I think the table is good right now, and doesn't need further amendments, but the extra information should be added in the "Derivative verb bases" sub-section below. What does everyone think? — JKVeganAbroad (talk) 01:44, 13 July 2021 (UTC)

I think the kanōkei base should be abolished as per Eiríkr's suggestion. Even Japanese Wikipedia has an article on the verbs known as 可能動詞 in the Miyake paper (jp:可能動詞), which are to be treated as separate verbs not derived via 活用形＋助動詞 (base + suffix) morphology as with other verb forms like the negative, causative, volitional, conditional, etc. (To clarify, 可能動詞 are the -eru verbs found in "行ける", "知れる", "正せる", etc. to mark the potential in godan verbs.)

Eiríkr also made a point that the passive and potential (-reru/-rareru) are really one and the same; this sparked improvement needs for the sections on the passive and potential forms. Perhaps the two sections should be combined into one, and have the 可能動詞 and ら抜き verbs be included in subsections. Any other suggestions are welcome!

@Eirikr: You also mentioned how the onbinkei is missing details. That's because the current information in this article treats the 音便-resulting kana of the ren'yōkei (e.g. the っ in 作って) as part of the suffix rather than part of the onbinkei of the verb (see: Japanese verb conjugation#te form). I do anticipate your discontentment of this logic, as the tables at ja:五段活用 treats that kana as part of the 音便-affected ren'yōkei (=onbinkei). If we reflect this in the verb base table, we would need to add at least seven more godan verbs (with different ren'yōkei endings) in order to illustrate the onbinkei of Japanese verbs, and I reckon that it would take a lot of space. Do you have any ideas on how to approach this? Excusememoi (talk) 04:19, 13 July 2021 (UTC)

@JKVeganAbroad and Excusememoi:

(For the next few days, I'll be very busy with other responsibilities IRL, so apologies for not being as active here in the meantime.)

Re: the Miyake text, I was pretty sure you meant the same one, but I'm more familiar with document titles than I am with Name + Year notation.  :) The Nakano paper looks vaguely familiar, but if I have read that before, it would have been years ago, and I am happy to have the link. Thank you!

Re: onbinkei, one compact option might be to copy what the editors did in the JA WP article. It's not elegant, but it seems clear. This also keeps the conjugational suffixes simpler -- て (te, conjunctive) is always て, た (ta, perfective) is always た, etc. Gemination and voicing can be explained as processes resulting from the preceding phonology, rather than having to describe different suffixes. I suspect they got included in the suffixes here in the EN WP article perhaps in part due to the English propensity to think in syllables rather than morae, whereby any geminate consonant might wind up belonging to the following syllable, rather than being its own (albeit silent) mora. Alternatively, and perhaps as a better option, explain the onbin mechanics separately, outside of the table -- the ren'yōkei for these verbs does still end in -i when used as a standalone, or when combined with other suffixes -- it's just the -te and -ta endings that cause this, if memory serves at the (slightly bleary-eyed, coffee-still-kicking-in) moment.

Re: kanōkei, agreed that it's probably best to not include that as a "verb base" per se. We don't include the passive as a "verb base", nor the causative, etc. Removing these from the table reduces the complexity considerably. That said, the article as a whole should mention these, how they are formed and how they function.

That's all I have time for today, I'm afraid. Thank you both for your effort on this page! ‑‑ Eiríkr Útlendi │^{Tala við mig} 17:16, 13 July 2021 (UTC)

"…it's probably best to not include that as a "verb base" per se. We don't include the passive as a "verb base", nor the causative…" — That example is a little precarious, because the passive and causative forms begin with the mizenkei base. Furthermore, the kateikei base is an accepted example that serves only a single verb formation (the conditional form). If we arbitrarily remove the kanōkei base because it only serves a single formation, then we should also remove the kateikei base (conditional) and the meireikei base (imperative) — but that would be totally unacceptable because those bases are fundamental to the whole base-theory, and removing them would redefine godan verbs as yodan verbs yet again.

If the kanōkei base is removed from the table, then none of the bases are a consistent fit for the patterns pertaining to the potential form. Such an anomaly must be explained (though, not in the "Verb bases" section, just in the "Potential form" section). But given the citations and resources that exist to indicate that the kanōkei base is derivative, I think it would almost be irresponsible to exclude the kanōkei base from the table.

I also don't think the table is too complex; if a reader is interested enough, this table will make sense to them. However, looking at the tables in Japanese Wikipedia for months was a nightmare. They're convoluted, and I don't support migrating versions of them into this article. — JKVeganAbroad (talk) 03:05, 18 July 2021 (UTC)

• Counterargument: Does the kanōkei really constitute a verb base?

I'd argue that it doesn't.

The other bases are all just that -- 1) conjugation forms of the core verb morphological unit, 2) onto which additional suffixes attach, and 3) which exist for all verbs. The kanōkei meets criteria 1 and 2, but not 3. The other various -kei meet all three: mizenkei (ending in -a for godan verbs), ren'yōkei (ending in -i), shūshikei + rentaikei (ending in -u), kateikei + meireikei (ending in -e), plus the derived pseudo-base that is the ishikei (ending in -o). Granted, for the ichidan verbs, all of these -kei end in the same vowel (either -i or -e, depending on the verb stem), but the -kei are recognized as existing for these verbs, and we can see this functionally by the fact that ichidan verbs can be conjugated to take the various suffixes that we know historically only attach to certain -kei -- such as the negative -nai that only attaches to the mizenkei, or the polite -masu that only attaches to the ren'yōkei, etc.

But the kanōkei doesn't even exist for ichidan verbs, not as defined by Miyake, nor as indicated in monolingual Japanese dictionaries -- Daijirin, for instance, includes kanōkei forms for those verbs that have them, as does Daijisen in its entries:

• Godan verb 行く (iku, "to go") -- points to the potential form ゆける (yukeru)
• Godan verb 帰る (kaeru, "to return, to go back") -- points to potential form かえれる (kaereru)
• Ichidan verb 食べる (taberu, "to eat") -- no potential given
• Ichidan verb 起きる (okiru, "to wake up; to occur") -- no potential given
• Irregular verb 来る (kuru, "to come") -- no potential given

Japanese dictionary compilers appear to treat the kanōkei as irregular enough to merit special handling, probably out of the consideration that the existence of the kanōkei is limited to godan verbs, and that there are even some godan verbs that don't have kanōkei (due to collisions with older ichidan variants with different meanings). I posit that this serves as circumstantial evidence in favor of the view that the kanōkei is not sufficiently "basic" enough to treat as a verb base -- at least, not for purposes of this table, which is trying to summarize the basic verb-stem forms common to all verbs. ‑‑ Eiríkr Útlendi │^{Tala við mig} 23:35, 19 July 2021 (UTC)

Your argument is very convincing. Although I did miss something: the dictionary compiler doesn't know to replace 「る」 with 「られる」(or even the colloquially unambiguous 「れる」?) to form the potential of any ichidan verb? That's a bit odd, isn't it? — JKVeganAbroad (talk) 01:32, 20 July 2021 (UTC)

English learning materials of Japanese grammar (example) often present this concept of a "potential form", where the ichidan potential form is homophonous to the passive form while the godan potential form all end in -eru. That's how it is presented in this article. However, as far as I'm aware, this concept doesn't exist in traditional Japanese grammar. As Eiríkr mentioned, the ichidan potential form is the the same as the passive form. It's not two forms that are spelled and pronounced the same, but it's a single form that carries multiple functions (passive/potential/polite; yet we refer to this form simply as "passive"). The passive form of godan verbs is also multifunctional in that way. That's the nature of れる・られる. It just so happens that, for godan verbs, there is a more common way to express the potential besides れる: the -eru form that English learning materials recognize as the "godan potential form". This form is treated as its own type of verb, called 可能動詞 kanō dōshi "potential verb", and it has no ichidan counterpart. Although, one might consider the ら抜き forms as an ichidan counterpart, but it's mostly recognized simply as a shortening of the passive られる form. Hopefully this helped to explain why Japanese dictionaries don't treat the ichidan (ら)れる as equivalent to -eru verbs when it comes to listing the [可能(動詞)] of a given verb. Excusememoi (talk) 04:02, 20 July 2021 (UTC)

• @JKVeganAbroad: Much as Excusememoi explained: in Japanese grammars, on the one hand, the kanōkei is not the same thing as the passive / potential / polite form (hereafter, "passive") ending in -(r)areru, while on the other, the passive is also recognized as a kind of potential, in addition to the kanōkei (for those verbs that have a kanōkei).

Moreover, it's not that dictionary compilers don't know how to form the passive, but rather that the passive exists for all verbs, so there is no value in including that in a dictionary entry: it is lexically superfluous information.

PS: I will be mostly offline for the better part of the coming week. ‑‑ Eiríkr Útlendi │^{Tala við mig} 06:16, 20 July 2021 (UTC)

So, if I'm not mistaken, I believe there is a consensus to remove the kanōkei form from the table of bases, since it's not a base. We should do that.

I stated somewhere in the discussion that I think it's a better idea to migrate information regarding the kanōkei form into the "Potential form" section, rather than leaving it in the "Verb bases" section. For example, migrating the kanōkei footnotes from the table. Each section should only contain relevant information. Somebody should also write an explanation in the "Potential form" section for why this form doesn't have a base, and is exceptional in that regard. — JKVeganAbroad (talk) 16:31, 21 July 2021 (UTC)

I retract my previous statement. I do believe the kanōkei base belongs where it is, in the verb bases section. We do not have consensus. — JKVeganAbroad (talk) 07:19, 28 December 2021 (UTC)

References

Verb abbreviation removal

So I haven't been on this page in a long time and noticed it was completely overhauled. However, all reverences to verb abbreviations have been removed. Is there a reason for this? It doesn't add any confusion and would be helpful for anyone referencing with a dictionary or trying to quickly find something on the page (vs typing out kana). Even if there is a reason for not adding back the column to the tables, why was the table on abbreviation -> verb type not kept? lukini (talk | contribs) 21:06, 4 November 2021 (UTC)

Hi. Firstly, we should define precisely what "references to verb abbreviations" actually means on this page, so that others know what we're talking about. It's this chunk of code that was removed from the page:

Table key

The conjugation tables below will include the EDICT word class abbreviations to disambiguate classes with similar word endings. See Japanese godan and ichidan verbs for more information about verb groups and their conjugations.

Abbreviation	Explanation
adj-i	adjective (keiyoushi)
adj-na	adjectival nouns or quasi-adjectives (keiyou-doushi)
adj-t	'taru' adjective
adv-to	adverb taking the 'to' particle
aux	auxiliary
aux-v	auxiliary verb
aux-adj	auxiliary adjective
v1	Ichidan verb
v5	Godan verb (not completely classified)
v5aru	Godan verb - -aru special class
v5b	Godan verb with 'bu' ending
v5g	Godan verb with 'gu' ending
v5k	Godan verb with 'ku' ending
v5k-s	Godan verb - Iku/Yuku special class
v5m	Godan verb with 'mu' ending
v5n	Godan verb with 'nu' ending
v5r	Godan verb with 'ru' ending
v5r-i	Godan verb with 'ru' ending (irregular verb)
v5s	Godan verb with 'su' ending
v5t	Godan verb with 'tsu' ending
v5u	Godan verb with 'u' ending
v5u-s	Godan verb with 'u' ending (special class)
vk	Kuru verb - special class
vs	noun or participle which takes the aux. verb suru

I removed this because I felt it was verbose and unhelpful: "Will the reader fail to understand how the verbs are conjugated without having the abbreviations?" — I believe not.

If we look to other languages that conjugate verbs, and look at their Wikipedia pages on the subject, tables with abbreviated categorisations of the words usually aren't present. Wikipedia is not a dictionary. Such examples include: Korean verbs; Arabic verbs; French conjugation; Spanish conjugation. Is the Japanese language exceptional in a way that requires abbreviations beside every (or any) example word? I suspect not.

Additionally, these abbreviations are in no way global or standardised across dictionaries.

Furthermore, when reading the page on mobile devices, adding an additional column's worth of information makes reading more difficult for the reader. I've tried to make the tables as simple as possible whilst still providing clues of the transformation from [verb]→[conjugated variant]. I think the outcome does this efficiently in a way that the reader can compare conjugation patterns of the same verb with little confusion.

Finally, is the information really useful? The 8 verbs used in the conjugation tables are identical, and they are amongst the most common verbs in the Japanese language. Certainly the most commonly used on Japanese Wikipedia (I did a lot of research before deciding on which verbs to use). If any reader is savvy enough to want to the the specific abbreviated category of those specific 8 verbs (13 verbs, if you include the additional 5 verbs in the た and て form sections only) — well, the chances are that they are students of the Japanese language and already knew the classification or how to find the classification. Meanwhile, readers who are unfamiliar with the abbreviation scheme are likely to be intimidated by its complexity. If those reader's are students of the Japanese language, they might be discouraged from pursuing their studies further. The system is certainly confusing.

So to summarise why I removed the table: I thought it was verbose and added confusion to the article, and articles on Wikipedia should be concise. — JKVeganAbroad (talk) 13:38, 12 November 2021 (UTC)

References

First section

Re this, I'm not especially knowledgeable on the subject, but the sections on perfectives and -te forms give several examples of verbs where the stem is altered in those forms, e.g. oku (stem presumably ok-) -> oita/oite.

And generally, in contrast to the rest of the article, which has lots of concrete information and excellent detailed explanation, this first section ("Verb groups") threw me a bit and seems rather vague and misleading (in particular because it talks about these "verb forms" as being the basis for conjugation, while the next section suddenly switches to talking about the bases, which is what are actually used throughout the article). I think the first section could do with some pruning and clearer explanation of what's going on and what its significance is (or isn't). W. P. Uzer (talk) 21:43, 22 December 2021 (UTC)

Thanks starting this conversation! So as for the perfective and -te forms, the stem is consistent with the other conjugations. In your example of oku (おく, to put, 置く), this is classified as a godan verb, and the verb stem is お〜 (o-). There is no standalone "k" in the hiragana syllabary, so "ok-" is not a valid verb stem.

This is an example of a problem arising from using romaji to analyse Japanese. With alphabetical representation, it appears that the stem for "oku" includes the letter 'k', with exceptions in the perfective and -te forms. The reality, however, is that when analysing the language with only hiragana in mind, the only written (and pronouncable) constant is "o-". Only the suffix is variant across all conjugations (even if the suffix appears to abandon a 'consonant' preserving rule, suffixes are inherently variant so this isn't truly an anomaly in the scope of things). Furthermore, "verb stem" is a linguistic term which does not equal "verb base" (which is a jargon term with exclusive use cases in the Japanese language). You can think of the verb stem as a superset of the verb base, and any conjugation is therefore a subset of verb bases which are subsets of verb stems.

I hope that clears up some confusion.

As for the verb bases, originally the article didnt include this information and was more straightforward. It can be entirely ignored. Most of the Japanese people I've talked to dont even know about them, and it doesnt seem to be a part of their education. However, Japanese conjugations are entirely derivative from these verb bases, so anybody looking into the historical explanations of modern Japanese patterns can start their research from that section of the article.

I agree that its a little jarring how the verb groups are promptly followed by the verb bases. I kind of feel like disclosing at the start of the verb bases section "This section is not necessary to read in order to understand the rest of this article. Feel free to skip this section." But I'm not sure breaking the 4th wall is appropriate for Wikipedia's expectations.

Your suggestions for improvement, if you have any, are more than welcome. — JKVeganAbroad (talk) 15:25, 24 December 2021 (UTC)

What you say about the stem always ending in a vowel makes more sense, but it seems inconsistent with some of what's written both in the section in question ("The unchanging portion of a godan verb stem can be analyzed as ending in a consonant"; "The invariant portion of the romanized godan verb yomu is yom-") and elsewhere (following the link to " Consonant and vowel nomenclature"). It all seems very confusing to me, as an uninitiated reader. Also, I don't know how standard the "verb form" terminology is, but it seems to be used inconsistently between this section and the next one, where "form" is made to mean something entirely different in different places within this section. W. P. Uzer (talk) 20:59, 24 December 2021 (UTC)

Oh! Yes! You're right! I actually removed all "consonant/vowel" terminology from the article early this year, because is outdated and flawed confusing terminology. Somebody added it back in recently but I think it should be removed again. It's problematic and—as you said—confusing to uninitiated readers.

As for inconsistencies with "verb form" (which has only 6 occurences in the article). This was just a stylistic choice, instead of writing "formation". Verb form/verb formation is intended to mean "fully conjugated verb". Do you think this should be reworded to be clearer? Please feel free to make a suggestion. — JKVeganAbroad (talk) 02:25, 27 December 2021 (UTC)

So maybe we could rearrange this section so that everything based on the "consonant/vowel stem" philosophy is clearly separated as an alternative approach? However, there still seem to be other problems. For example (in the explanation of why godan is called godan): "...the five different vowels that appear on the end of the stem in the different verb forms." If the stem is the unchanging part as you say, then these "five different vowels" are not on the end of, or even adjacent to, the stem. Maybe it means that the stem can be followed by five different kana - but then in ichidan verbs, the stem can also be followed by some number of different kana. W. P. Uzer (talk) 07:46, 27 December 2021 (UTC)

There's literally a whole article that discusses the categorisation and naming conventions of verb groups, so its not good to duplicate what isn't necessary from that page to explain Japanese verb conjugations on this page. i.e. We shouldnt include a consonant/vowel philosophy on this page at all, as its distracting from the point of the article and is problematic for more reasons than we've already discussed. I'll make an effort to remove it.

You're right though about the wording being confusing. It should say "conjugational stem in the different verb form(ation)s." since the verb base/conjugational stem ≠ verb stem.

Thank you for your fresh eyes on this article, I had a feeling it may have been unclear in some areas, but I was blind to it. — JKVeganAbroad (talk) 02:02, 28 December 2021 (UTC)

Also, you haven't answered my question about rephrasing "verb forms". I gave a couple of suggestions... never mind. — JKVeganAbroad (talk) 02:09, 28 December 2021 (UTC)

It's me again. It turns out that all the confusing language (verb forms, verb stems) you were confused by was added in by the same edit that re-introduced the consonant/vowel terminology. I rewrote the verb groups section with the context of the article in mind, hopefully it makes more sense. I also noticed some other small errors in the article, which I addressed as well.

Please tell me if this works better for you! — JKVeganAbroad (talk) 07:13, 28 December 2021 (UTC)

Yes, thanks! I think it's a lot more logically consistent now. One question - in the first table, the various forms of 見る (to see) are written all in kana, while the forms of the other verbs start with a kanji. Is there a reason for this, or is it an accident? W. P. Uzer (talk) 08:11, 28 December 2021 (UTC)

Awesome! As for 見る being written in hiragana for the first table (Verb groups) but writin in kanji everywhere else, it was intentional. The "learning moment" of the first table is to see how the kana in the verb (visually) matches a kana in the gojūon table of hiragana. Since 見る isa two-kana word, changing only the る might disguise that "learning moment". The challenge for the table was to find 3 basic (i.e. common) example words that share the same column of the gojūon table (in this case, the マ column).

Using kanji everywhere else is intended to show Japanese as it would be written. — JKVeganAbroad (talk) 09:01, 28 December 2021 (UTC)

I see - maybe then a footnote in small font under the table, briefly explaining this? (Otherwise a non-knowledgeable reader might think the 見 kanji can't be used in the derived forms.) W. P. Uzer (talk) 11:34, 28 December 2021 (UTC)

It wouldn't hurt to have a footnote. — JKVeganAbroad (talk) 13:06, 28 December 2021 (UTC)

OK, I've added one - feel free to correct or amend. W. P. Uzer (talk) 09:45, 29 December 2021 (UTC)