In Japanese there are two morphemes which are used before certain nouns as part of the honorific system:

  • (o)
  • (go)

Which terms can be used to refer to these out of "prefix" and "particle"?

I haven't asked which is "correct" because I don't know if that's meaningful. But if it is that's fine, otherwise whichever is used by linguists.

I suspect that other than these Japanese might not actually have prefixes, which might be reason to not analyse these as such either.


3 Answers 3


"Particles" in Japanese are actually a fairly diverse class of words. Some things which are traditionally called particles are suffixes. For instance, the representative plural marker, -tachi is very closely bound to the noun it attaches to--not even core argument case markers like =ga or =o can interrupt it.

Other things traditionally called particles, especially the non-core argument-marking case particles are clitics. This is most clearly seen based on the fact that some of them, like the genitive case marker =no, which, while not being very closely attached to the noun it modifies, it still has some phonological influence. For example, words where accent falls on the last mora, like hashí LH(L) 'bridge' become accentless when =no is added: hashi=no LHH. Compare this with what happens when you add the case marker =ga: hashí=ga LHL--we see the expected pattern for a final-mora accented, two mora word.

Yet other things are postpositions. Here I'm thinking specifically of sentence-final particles like yo. These clearly don't attach to individual words, but instead to phrases or entire sentences (depending on their scope). They are prosodically independent, have some semantic independence, but are still somewhat "stuck" onto the phrase they modify.

Still other things are likely entirely separate words which happen to be one or two morae long, and thus just happened to be analyzed traditionally as particles. Here I'm specifically thinking of things like the na in so-called na adjectives (or adjectival nominals, if you prefer), which is actually one of the attributive forms of the copula no, itself a defective verb.

So what about these honorifics you can stick at the beginning of a word? I'd just call them prefixes. They are very closely bound. When an accented noun (of any kind) takes the prefix o-, it becomes accentless. So, for instance, intially-accented íkutsu HLLL 'how many?' becomes o-ikutsu LHHH. Similarly, iwái LHL 'congratulations', which has its second mora accented, becomes o-iwai LHHH. And so on. Even outside of a more rigorous linguistic analysis, such as under the traditional Japanese system of traditional grammatical analysis, to the best of my knowledge, these are not called particles (助詞 joshi), but rather prefixes (接頭語 settoogo).

  • 1
    "When an accented noun (of any kind) takes the prefix o-, it becomes accentless". Interesting, do you have any references on this rule? There seems to be some some exceptions, e.g. nábe/onábe, pot.
    – dainichi
    Mar 12, 2014 at 11:05
  • 1
    Sam Martin's reference grammar lists the rule (*o*+NOUN -> atonic), as well as the large number of exceptions (Martin 1975: 331-6). Prototonic nouns seem especially likely to have exceptional forms.
    – limetom
    Mar 12, 2014 at 22:50
  • Thanks for the reference. Unfortunately those pages are not in the Google Books preview, and the book is not exactly cheap, but I'll keep it in mind. I might cross paths with it sometime.
    – dainichi
    Mar 13, 2014 at 0:20

A couple of things to note: Japanese has other prefixes, mi- and on- are others I think, but even if o- and go- were the only ones, you could still call them prefixes and this is related to the second thing I was talking about: the number of them is not a criterion for establishing whether you can call them prefixes or not, correct me if I'm misinterpreting here, but I think that's what you intended in your last sentence.

If -san as in Daniel-san is considered a honorific suffix, then you could call o- a honorific prefix.1 In that case it does act as a prefix, because it's attached to the word. In Japanese there's no use of hyphens in this sense, so there is no visual separation. You could argue that no distinction is made for words in Japanese in general, and no space is present between them, however the honorific is undoubtedly attached to that word and not to others. See for instance:


-san here is a suffix.

o genki desu ka

where we have 元気 (genki - health), です (desu - the copula) and か (particle indicating that this is a question), all preceded by お, the honorific o-. You could say 元気ですか (genki desu ka) and it would still be correct, however not appropriate for situations that require some formality.

So, these are prefixes, however, addressing your other point, I wouldn't call honorifics — particles. Particles, when referring to the Japanese language, are a different category and I think you might create confusion in your listener if you called お- a particle.

1: To be honest, you don't even need to call them "honorific prefixes/suffixes", since saying "honorifics" is also a noun and not just an adjective.

  • I wanted a less ambiguous term that "honorifics" because on its own that term seems to refer to the whole system of grammaticalized politeness/respect/etc and sometimes there's a need to be more specific. Both "honorific particle" and "honorific prefix" get Google hits but I'm not sure how many are lay/informal uses that professionals might from upon. Compare with a lay person talking about , , as "suffixes". You can see why people might call them that but I imagine they would mostly be corrected if they were talking to an expert. Mar 7, 2014 at 13:31
  • In your last paragraph you say "Particles, when referring to the Japanese language, are a different category" but actually the particles are not one but several quite different categories. は, が, を are in one category, the postpositions are in another, sentence final particles are in yet another. So I'm unsure whether the honorifics might be in one more again. Also, following your analysis and reasoning, on a similar question about the postposition particles, would you also come to the orthogonal conclusion "So, these are suffixes"? Mar 7, 2014 at 13:34
  • There are subcategories? Probably, but they are still called all particles. I would never call (and never heard/seen someone calling) は, が, を, ね, よ, etc., simply suffixes. When you say Japanese particle you mean those. And I don't think there are "pre-positioning" particles... They are always positioned later as far as I can remember.
    – Alenanno
    Mar 7, 2014 at 13:45
  • I don't think I'd come to that conclusion. When you say "Compare with a lay person talking about は, が, を as "suffixes".", well, they would be at least inaccurate, because all students, all teachers, and all resources usually refer to those as particles. Now, there might be a better term for them, but suffix would be confusing here, if not simply because everybody calls them particles.
    – Alenanno
    Mar 7, 2014 at 13:51
  • 1
    If the particles are a category in Japanese, the subcategories apparently are 格助詞, 並立助詞, 終助詞, 間投助詞, 副助詞, 係助詞, 接続助詞, and 準体助詞. This analysis doesn't include honorifics, but I don't know if it's the only analysis. Wikipedia / Japanese particles / Types of particles Mar 7, 2014 at 14:11

The expressions o- and go- are prefixes. They are not derivational, but also not inflectional affixes. I believe that's what's causing the confusion between "prefix" and "particles". Derivational prefixes are the negation prefixes hu-, bu-, mu-, mi-, and hi-. Attachment of these prefixes causes derivation from noun to nominal adjective (な形容詞・形容名詞). Here's an example:

 a. [[mu- [kankei]]  -na]
      neg- relation  -attr
 b.*  kankei-na

The structure in (a) must be correct, because only mu-, but not kankei (b), allows the appearance of attributive -na. By the way, be careful with the data there, that's not correct. Rather take a look at "Rickmeyer, Jens. 1995. Japanische Morphosyntax. Groos, Heidelberg".

Returning to the meaning of o- and go-: naming them as honorifics is not entirely correct. They can appear in constructions meaning 1. respect, 2. modesty, 3. eulogy (美化語 'bikago' from 美化 'bika' = "euphemism"):

 c. o-   musume   -san      respect
    rsp- daughter -prs
    'your daughter'
 d. o-   mise -si -mas -u   modesty
    mds- show -do -pol -npst
    'I show [you'
 e. o-   mizu               eulogy (refined register)
    eul- water

I have used here examples with o-, but go- means basically the same, but is normally used with Sino-Japanese vocabulary. There are, of course, exceptions.

To summarize: o- and go- are prefixes, not particles.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.