Manner Adverbs

There have been a few questions about how adverbs work in Dothraki, but the topic is actually larger than one might expect. For that reason, I decided to break it down by category.

Those who grew up with English may be surprised (or, at least, I was at first) to learn that there are actually three types of adverbs, and they behave differently from language to language. The three types are: manner adverbs; temporal adverbs; and spatial adverbs (I’d like to link the other two, but I can’t find a nice description for either online). The type of adverb that most of us think of when we hear the word “adverb” is the manner adverb, so I thought that would be a good place to start.

Manner adverbs modify the verb by specifying the manner or way in which the action is performed. Most common adverbs that end in “-ly” in English are manner adverbs. In fact, most derivable adverbs in English are manner adverbs (“crazily”, “jerkily”, “monetarily”, “mockingly”, etc.), though we do have a couple ways of creating others (e.g. “-ward(s)” to make spatial adverbs, such as “homeward”, “store-wards”, “computer-wards”, etc.).

In Dothraki, I liked the idea of manner adverbs being a closed class, for the most part. There are a few manner adverbs that are derived in a regular way (or were derived in a regular way some time in the past), but that derivational process is frozen. Here are a few examples:

  • alle (adv.) farther, further (cf. ale, “more”)
  • atte (adv.) first (cf. at, “one”)
  • disse (adv.) only, just (cf. dis, “simple, plain”)
  • yomme (adv.) across (cf. yom, “crossing”)

Most old adverbs work this way, though there are one or two that don’t (the prime example being chek). If you want to make a conventional manner adverb out of a modern adjective, then, you have to do something a little different.

If you want to say, “That colt is running with a limp” or “That colt is running lamely”. First thing you do is take the word for “lame” or “limp”, darin, and turn it into a noun: athdarinar, “lameness”. Next you prepose it with the preposition ki. Since ki assigns the genitive, you put athdarinar in the genitive, and since the word begins with a vowel, ki gets shortened to k’-. Then you put it at the end of the sentence and you get:

  • Rek manin lana k’athdarinari.

Literally, that would be something like, “That colt runs by lameness”. The idea is that ki preposes the cause of the action. Since the action, in this case, is running that looks painful or unnatural, the suggested cause is lameness. And that’s the story behind the construction.

(Note: One can imagine that, a century or so later, this construction might give birth to a new circumfix that could be used to form adverbs productively.)

Regarding placement, adverbs of manner usually occur sentence-finally. They can be fronted for emphasis, but the construction would be marked (i.e. the adverbs would be noticeably out of place, and so could only be there for some pragmatic purpose). If there’s more than one post-verbal phrase, the adverb could occur in a non-final position, but usually it’d only be on account of what’s known as heavy shift. Here’s an example:

  • Me dothrae chek rek hrazef fin azh anha yeraan oskikh hajinaan m’anha vo zigerok mae.

That is, “She rides well that horse I gave you yesterday because I didn’t need it.” Ordinarily you’d put chek at the end, but since there’s so much material in the object clause, you can shift chek closer to the verb so that the hearer (and the speaker!) don’t forget what it’s referring to.

I think that covers manner adverbs. If I’ve missed anything (or mistyped anything), I’m sure Qvaak will let me know (heh, heh!). If anything’s unclear, feel free to ask about it!

12 comments

  1. No error-like oddities I can see :)

    The way you discuss adverbs seems to center on real “adjectives for verbs” type words/word use, and I trust that was what ingsve foremost tried to ask about.. but there’s also this wider interpretation, as wikipedia says: “Adverbs can modify verbs, adjectives (including numbers), clauses, sentences, and other adverbs.” How do you, or Dothraki, treat these other “adverb” thingies?

    Hajinaan is a new interesting word. I think we know only haji, and even that only as a preposition. Suordinating conjustion? Inflected? Does it always come with me-, or do I even interpret the m’anha correctly?

    1. The way you discuss adverbs seems to center on real “adjectives for verbs” type words/word use…

      Another way of saying that is “manner adverbs”. And, yes, that was the subject of today’s post. ;)

      Does it always come with me-, or do I even interpret the m’anha correctly?

      You are interpreting m’anha correctly. Indeed, the phrase is hajinaan me…, and you’d use it the way you’d use “because” in English—crucially not “because of”: just “because”.

  2. When I think of adverbs modifying other words than verbs, the fist word that springs to mind is very. Sekke seems to work rather similiraly, and since it isn’t locative or temporal adverb, and fits only so-and-so to manner category, I was wondering, if you consider it adverb at all.

    Also, right after very, my mind jumps to sentences like “We watched astoundingly beautiful sunset.” and “He isn’t attending, probably because he’s sick.” These seem precisely manner type adverbs, so the way you only speak of manner adverbs modifying verbs makes me think maybe Dothraki does not use these kind of structures.

    1. These seem precisely manner type adverbs, so the way you only speak of manner adverbs modifying verbs makes me think maybe Dothraki does not use these kind of structures.

      What a bizarre conclusion to come to! Anyway, “probably” certainly modifies the verb in your example (at least in the way I was thinking). I guess I was thinking of it syntactically (i.e. where it attaches in the tree). Saying it modifies the verb is a shorthand. What I mean is you have a situation like this:

      [NP [ [V [NP] ](VP) ](TP)

      So that’s a basic transitive clause (the parts in parentheses name the constituent to the left). Manner adverbs attach at the top level level, and so can occur at the beginning or end of the sentence (though the end is most common), but not comfortably in between. In this sense, they modify the clause, the primary part of which is the verbal idea—so they modify the verb. In the clause “Probably because he’s sick”, “probably” modifies the entire clause “because he’s sick”; it doesn’t modify “because”. So “probably” there is basically the same type of adverb—and, indeed, we’ve seen other adverbs just like the manner adverbs described above already that modify, essentially, the truth value of the sentence (e.g. Anha tih mae k’athjilari. “Indeed I saw him.”).

      Sekke seems to work rather similiraly, and since it isn’t locative or temporal adverb, and fits only so-and-so to manner category, I was wondering, if you consider it adverb at all.

      That really depends what you think of the word “adverb”. In English, the definition of “adverb” is vast and includes modifiers that aren’t adjectives. Notice that you can’t drop other manner adverbs into that same slot:

      (1) That’s really good ice cream!
      (2) That’s truly good ice cream!
      (3) * That’s quickly good ice cream!
      (4) * That’s happily good ice cream!
      (5) */? That’s chivalrously good ice cream!

      Let me go through why I chose all those examples. In (1) and (2), there’s no question that the adverb modifies the adjective, and the sentences are grammatical. In (3) and (4), we’re forced to accept an interpretation where the adverb modifies the adjective, and the result is infelicitous. The same is true of (5), but I put a question mark there because it might be acceptable. The reason is that the adverb is so bizarre in that context that we can almost interpret it as a simple intensifier—or, rather, an intensifier with flare (like Homer’s adverb “groin-grabbingly”). “Happily” and “quickly” are both bizarre, but aren’t extraordinary enough to get that interpretation.

      Anyway, intensifiers behave differently from all other types of these clause-level adverbs. In English we call them adverbs, but I don’t think they properly should be called adverbs. They can’t be used in any other way (e.g. “I walked to the store very.”), and perform a function that’s more similar to articles than anything else. They’ll need to be treated separately.

      Though, of course, the form of sekke is similar to the old adverbial form (and that’s no accident. Consider “really” which no longer means “in a way that’s real” [e.g. “He ate food really.” as a counterpart to “He ate food virtually.” (i.e. in some virtual reality setting) doesn’t work]).

  3. What a bizarre conclusion to come to!
    Thanks. That’s what I do best.

    Anyway, “probably” certainly modifies the verb in your example (at least in the way I was thinking).
    Ach. I’m so out of my depth here, not a linguist and not even that much of an English speker, but anyway.. The example wasn’t that great. “She isn’t attending, probably because she’s pregnant.” might be a bit better. The sentence seems to me somewhat ambivalent, but still different in meaning from “She isn’t attending, because she’s probably pregnant.” In the latter the reason for not attending is certain: probable pregnancy. In the former I think the pregnancy would be in most contexts taken as a solid fact and the only uncertainity would be if or if not it were the reason for not attending.

    Anyway, intensifiers behave differently from all other types of these clause-level adverbs. In English we call them adverbs, but I don’t think they properly should be called adverbs. They can’t be used in any other way (e.g. “I walked to the store very.”), and perform a function that’s more similar to articles than anything else. They’ll need to be treated separately.

    Thanks. That clarifies that issue quite [qualifier :)] well, I think.

    1. Regarding your pregnant example, what you’re illustrating isn’t showing us a different type of adverb, but different syntactic arrangements that give rise to different pragmatic interpretations. There is a possible reading (in your second example) where “probably” would work like a modifier (similar to “very”), but there’d have to be a reason you get that reading (e.g. “Is she just probably pregnant, or possibly definitely pregnant…?”). The more natural reading (where “probably” modifies the clause) is just pragmatics. I’d wager (though I won’t say for sure) that it might be some type of quantifier float like you see in examples such as those below:

      (1) All the children left.
      (2) The children all left.

      I’m not 100% sure about how a standard tree analysis of that example would work, though, so I’m not sure what’s moving where or why. I’m pretty sure, though, there are two readings: one where “probably” is acting like a modifier of “pregnant”, and the simpler reading where it’s modifying the clause as a whole.

  4. Athdavrazar, zhey David!, every time I read your posts, not only do I learn more about Dothraki, I learn more about language in general!

    1. Feel free to let me know when something needs more explanation. I tend to think I’m pretty good at explaining stuff, but I seem to keep on using terminology that isn’t generally known, so I need to make sure to define everything that needs defining.

      1. I’m sorry. My mistake was another one and only now I get it. I thought “rek” to be Inanimate Accusative, but that would be true if it were a pronoun instead of a modifier.
        I hope I’m not wrong again, this is getting rather confusing for me! Anyway, thank you for taking the time to reply

Leave a comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.