I Guess Korean Has More Mirative Markers: -Napo and -Nmoyang

Iksoo Kwon


I Guess Korean Has Some More Mirative Markers: -Napo- and -Nmoyang-
The aim of this paper is to shed light on the semantics of some newly discovered Korean evidential markers -napo- and -n moyang- and to discuss their underlying cognitive mechanism by investigating their functional similarities and differences.
1. PHENOMENA IN FOCUS. The markers, to begin with, are inferential evidential markers; as shown in (2)-(3), they indicate that the speaker infers that the event referred to occurs based on her own observation:
(1) cip-ey koyangi-ka iss-e [an utterance without the markers]
house-Loc cat-Nom be-Ending “There is a cat in the house.”
(2) [After finding a cat’s footprints around the door,]
a. cip-ey koyangi-ka iss-napo-a Lit. See whether there is a cat
house-Nom cat-Nom be-napo-Ending in the house.
b. cip-ey koyangi-ka iss-nu-n moyang-i-a Lit. It is a shape (=situation) that
house-Nom cat-Nom be-Pres-n moyang-cop-Ending there is a cat in the house.
“(I can infer that) It seems that there is a cat in the house.”
(3) [Seeing that the addressee’s pants is wet,]
a. pakk-ey pi-ka o-napo -a Lit. See whether it rains outside
outside-Loc rain-Nom come-napo-Ending
b. pakk-ey pi-ka o-nu-n moyang-i-a Lit. It is a shape (=situation) that
outside-Loc rain-Nom come-Pres-n moyang-cop-Ending it rains outside.”
“(I can infer that) It seems that it rains outside.”
The evidential markers, however, are intriguing in that they can function as a mirative evidential (DeLancey 1997): (2) and (3) sometimes indicate unexpected information, even when the speaker has direct visual access to the information. For instance, (2a-b) should be used in a context where the speaker does not have a cat in her house, whereas the cat in (1) can be specific; (3a-b) can be used in a context where the interlocutors have not known that it was raining and where the speaker who discovered it first let the addressee know.
Notice that the marker’s unique semantics is also employed as a politeness strategy to hedge the speaker’s assertiveness for politeness: In (4), even though she knows that the event referred to is a fact, the speaker pretends not to know it in front of her senior:
(4) [A student explaining what happened the previous night to her advisor, who was too drunk to remember anything,]
ecey sensayingnim-kkeyse manhi chwiha-si-ess-*(napo-/-te-n moyang-i)e-yo.
yesterday teacher-Nom [Hon] much be.drunk-Hon-Past-napo-/retro-n moyang-cop)Ending-Hon
“(I can infer that) It seems that you were drunk too much.”
2. BACKGROUNDS. Most Korean linguists apparently regard the category of evidentiality as a redundant category due to its functional overlapping with mood, since there are more than a few cases where each of evidentiality and epistemic modality induces inferences about the other and since Korean has an elaborate mood system that can cover some of the territory of evidentiality (hinted by Y.-K. Ko 2007; inter alia), on the one hand. On the other hand, there are a number of linguists, (Kwon 2009, J.-M. Song, 2002, K.-S. Chung 2005, inter alia) who assume that the category of evidentiality definitely exists in Korean and focus on some of the evidential markers case by case. This paper will not treat this controversy any further, but it seems clear that the category of evidentiality is conceptually salient, and cross-linguistically pervasive. Meanwhile, it is surprising that most of studies on evidentials in Korean only concentrate on an individual item and that there has not been a comprehensive work outlining the overall picture of the overall system of evidentiality in Korean (cf. K.-S. Chung 2005). My long-term goal is to figure out the overall system of evidentiality in Korean and to characterize the semantic/pragmatic properties of the markers thoroughly. This paper focuses on the two inferential evidential markers -napo and –nmoyang among them. These markers have not been thoroughly discussed in previous literature. To explore their semantic and pragmatic properties would be of particular interest, since their extended properties are inherited from their original constructions (to be discussed in section 3) where elaborate pragmatic inferences are involved.
3. ANALYSES. This paper argues that the semantics of the markers converges from two different origins (-na ‘whether’ and po- ‘see’ > -napo-; -n relativizer and -moyang- ‘shape’> -nmoyang-), because the speaker employs different intersubjective pragmatic tactics that converge into a single function: to draw the addressee’s attention to what the speaker inferred. The original constructions can be shown as follows:
(7)a. Chelswu-ka (o-na po-a /b. o-n moyang-i-a)
Chelswu-Nom come-whether see-Decl come-Relativizer shape-Cop-Decl.
“See whether Chelswu is coming.”
The pragmatics underlying the original structure of the -napo construction is that he speaker urges the addressee to confirm what the speaker conjectures, on the one hand. On the other hand, -nmoyang is originally used in contexts where the speaker intends to describe a situation from a certain epistemic distance, which implies that the speaker has some level of knowledge of the focal event. The difference in their original constructions results in subtle functional difference of their grammaticalized forms.
SUBJECT CONSTRAINTS. Other than by showing their functions, there is another supporting evidence that the markers are evidentials: they show an intriguing asymmetry in subject usages in utterances that employ either of the markers, as Aikhenvald and Dixon (2003: 16) noted that “[E]videntiality systems often interact with the grammatical person of the subject or experience.” If the markers show limitations on first person subject usages, then their evidential function requires that the speaker’s inference be based on her objectified observation, which yields asymmetry between usages involving first person subjects and those with non-first person subjects:
(8) a. ku (?nay)-ka hakkyo-ey ka-ess-napo-a
He(?I)-Nom school-Loc go-Past-napo-Decl
b. ku (?nay)-ka hakkyo-ey ka-nmoyang-i-a
He(?I)-Nom school-Loc go-nmoyang-Cop-Decl
“(I can infer that) It seems that he (I?) went to school”
The markers presuppose that the event referred to in the utterance is based on the speaker’s inference. It is also implicated that the speaker is not certain about the information. But, people are normally aware of their own conscious, volitional actions such as going to school. In this context, if the speaker mentioned what she had done with either of the inferential markers, it would be contradictory: Since the mode of access that the indirect evidential encodes is not direct, the inferential evidential conflicts with the direct access that a first person subject usually has with respect to her own past actions. This is a typical characteristic of evidentials.
MIRATIVITY. The inferential evidential markers -napo and -nmoyang are also licensed in context where the speaker intends to express the unexpectedness of the focal event. As Delancey (1997, 1999) noted, there is a kind of evidential that encodes/implicates the ‘unprepared mind’ of the speaker, which is a mirative evidential, and this paper argues that the two inferential markers are extended to be used as such function:
(9) cip-ey koyangi-ka iss-e
house-Loc cat-Nom be-Decl
“There is a cat in the house.” [an unidentified cat or a pet cat]
(10)a. cip-ey koyangi-ka iss-napo-a
house-Loc cat-Nom be-napo-Decl
b. cip-ey koyangi-ka iss-nu-nmoyang-a
house-Loc cat-Nom be-Imperf-nmoyang-Decl
“(Oh) Wow, there is a cat in the house.” [an unidentified cat]
We can see that in (10), the cat should not be a pet cat that is already identifiable to the speaker, whereas (9) could refer to either a specific or a non-specific cat. In sum, considering the given examples above that are not prototypical examples that show inferential semantics, we can learn that markers -napo and -nmoyang can encode mirativity as well as inferential evidential semantics.
POLITENESS TACTIC. The markers can be used in a politeness strategy neutralizing the speaker’s assertiveness:
(11) [When a bartender politely tells a drunken customer, who is demanding more drinks, to stop drinking and to go home]
a. cip-ey ka-si-e-yacyo. manhi chwiha-si-ess-#(napo-)e-yo.
home-Loc go-Hon-Decl-Hon.end much be.drunk-Hon-Past-napo-Decl-Hon
b. cip-ey ka-si-e-yacyo. manhi chwiha-si-#(nmoyang-i)e-yo.
home-Loc go-Hon-Decl-Hon.end much be.drunk-Hon-nmoyang-Cop-Decl-Hon
“You should get yourself going home. (It seems that) You’re very drunk.” [Koreterm #3643]
The utterances would have been rude, without the markers -napo and -nmoyang. The utterances involve the speaker’s pragmatic tactic of flouting: although the bartender has perceived the customer’s obvious drunken state, he pretends not to have directly perceived what is referred to and also pretends to indirectly infer it from some other evidence. It is noted that again, this usage of the marker is not prototypical inferential marking: The speaker has observed it, she has certainty about the information, and thus, its inferential marking function is not needed, logically speaking. Part of the pragmatic tactics involved is related to the markers’ inferential function and the other part of it is related to the markers’ mirative function. That is, in case of (11), the speaker should not assert directly that her customer is drunk, since otherwise, the utterance will threaten her customer’s face. Thus, the speaker employs either of the inferential markers in order to distance herself from the focal event, i.e., in order to pretend not to know what she observed, as if she had not perceived the scene directly. Also, the speaker employs it as if she had just recognized that her customer is drunk. The shorter the period of time during which the customer has been “obviously” drunk, the less face threat is involved in telling the customer about it.
4. DISCUSSION. So far, it has been shown that inferential evidential markers -napo- and -nmoyang- are distributed in a similar way and that even their extended functions, such as implicating mirativity and giving a rise to politeness reading, are similar to each other.
SUBTLE FUNCTIONAL DIFFERENCE. Nevertheless, the markers carry subtly different degrees of the speaker’s attitude towards novelty of the information that is talked about. For instance, -napo can encode the speaker’s urgency towards the focal event, whereas -nmoyang encodes the speaker’s controlled and calm emotional state towards the focal event:
(12) [While sleeping, hearing that someone has come in, the speaker is scared and says to her husband]
a. totwuk-i tul-ess-napo-a
thief-Nom come.in-Past-napo-Decl
b.?? totwuk-i tul-unmoyang-i-a
thief-Nom come.in-nmoyang-Cop-Decl
“(I can infer that) It seems that a thief broke into our place.”
In brief, in order for -nmoyang to be licensed, the speaker will keep a relatively farther cognitive distance toward the focal event, as if the speaker had observed the event very objectively as a third party. In contrast, -napo is likely to encode the speaker’s urgent attitude toward the target event.
DISTANCING STRATEGY. In a number of cases, it is not easy to choose exactly one of these functions as the relevant one in context. This might implicate that the three functions in context are compatible with each other and thus, entangled conceptually. Here is an example show the entangled functions of the markers:
(13) [The speaker is scheduled to meet her friend, but she feels sick. She feels sorry about not being able to keep the promise and tries to apologize to her friend, saying,]
nay-ka aphu-{n-kapo/nmoyang-i}-a
I-Nom be.sick-{Conn-napo/nmoyang-Cop}-Decl
“(I can infer that) It seems that I’m sick.”
Literally, the speaker expresses a guess about her own internal state. Notice that first person subject is used with either of the markers. This is possible, first, because she does not want to take responsibility for the consequence of her illness—canceling the appointment. If no evidential marker is used, the transfer of the information is direct and the fact that the speaker is sick will be directly vouched for by the speaker, which might yield an assertive reading. In (13), the speaker is less assertive and thus, avoids direct responsibility. Second, the markers allow the speaker to express the unexpectedness of the information. And obviously an unexpected problem is a better excuse for inability to keep a promise. In brief, (13) includes all the functions mentioned above: The speaker employs an inferential evidential construction in a context where it is not only a prototypical inferential statement, but also a mirative statement, which is further used as a politeness strategy. The cognitive mechanisms underlying the multiple usages of the markers stem from the speaker’s effort to distance herself from the event in question, exploiting the indirective semantics of the markers.
5. CONCLUSION. This paper argues that the markers -napo and -nmoyang are to be regarded as evidentials, by looking into their basic properties and extended properties. Specifically, they are inferential markers, which encode the speaker’s inference based on observed entities, properties, or relations. It was also shown that the markers can encode a mirative function that marks unexpectedness or unpreparedness to the speaker’s mind. Moreover, the markers are intentionally employed in a politeness strategy. This strategic indirectness derives from their inferential and mirative characteristics. Based on these observations, this paper provides a unified account for the semantic functions of these two markers, following DeLancey’s assumption that inferentials, evidentials, and miratives are semantically related, and argues that the relationship is based on pragmatic tactics of exploiting inferential and mirative senses for indirectness and politeness. It is also discussed that even though the two markers are distributed very similarly, they are subtly different in that -napo encodes immediacy, whereas -nmoyang encodes the speaker’s non-immediacy and even the speaker’s controlled and calm state of mind toward an event. Based on the similarities and differences between the two markers’ (extended) functions, this paper argued that the cognitive mechanism underlying the two markers’ functions is a distancing strategy for interactive meaning negotiation.
Aikhenvald, Alexandra Y. and R.M.W. Dixon (eds.). 2003. Studies in Evidentiality. Oxford: Oxford University Press.
Chung, Kyung-Sook. 2005. Spaces in Tense: The Interactionof Tense, Aspect, Evidentiality and Speech Act in Korean. Ph. D. Dissertation. Simon Fraser University.
DeLancey, Scott. 1997. Mirativity: The Grammatical Marking of Unexpected Information. Linguistic Typology 1. 33-52.
DeLancey, Scott. 1999. The Mirative and Evidentiality. Journal of Pragmatics 33, 369-382.
Ko, Yeong-Keun. 2007. Hankwukeuy Sicey, Sepep, Tongcaksang [Tense, Mood, and Aspect in Korean]. Kyeonggi-do, Korea: Taehaksa.
Kwon, Iksoo. 2009. The Korean Evidential Marker -te- Revisited: Its Semantic Constraints and Distancing Effects in Mental Spaces Theory. Paper presented at Fillmore Fest 2009: Frames and Constructions, University of California, Berkeley, July 31-August 2, 2009.
Song, Jae-Mog. 2002. A Typoloical Analysis of the Korean Evidential Marker ‘-Te-.’ Enehak [The Linguisitic Society of Korea] 32, 147-164.

Full Text:


DOI: https://doi.org/10.3765/exabs.v0i0.499