Acta Linguistica Asiatica, 14(2), 2024. ISSN: 2232-3317, http://revije.ff.uni-lj.si/ala/ DOI: 10.4312/ala.14.2.147-171 Direct Evidentials in Korean: From the Perspective of the Multi-Store Memory Model MOON Chang-Hak Department of Japanese Language, Literature and Culture, Korea moonchanghak@hufs.ac.kr Abstract This study clarifies the meaning of direct evidential markers in Korean and examines their semantic characteristics using a multi-store memory model. Korean direct evidential markers are categorized into those indicating either “present perception-based knowledge” or “past acquisition-based knowledge.” The former are subdivided into “unaccepted present perception-based knowledge” (-네 -ney) and “accepted present perception-based knowledge” (-군 -kwun), and the latter into “knowledge derived from past perception” (-더라 -tela) and “knowledge integrated into personal or general understanding” (-지 -ci). Within the multi-store memory model, -네 -ney aligns with the indicator of the maintenance rehearsal process, while -군 -kwun serves as an indicator of elaborative rehearsal, and -더라 -tela and -지 -ci correspond to retrieval indicators from long-term memory. Keywords: Korean, evidential markers, perception, acquisition, multi-store memory model Povzetek Študija razjasni pomen neposrednih evidenčnih označevalcev v korejščini in preuči njihove semantične značilnosti z uporabo modela večplastne shrambe spomina. Te razdelimo na tiste, ki nakazujejo "na sedanjosti temelječe znanje" in tiste, ki nakazujejo "na pretekli pridobitvi temelječe znanje." Prve nadalje razdelimo na "nepriznano na sedanjosti temelječe znanje" (-네 -ney) in "priznano na sedanjosti temelječe znanje" (-군 -kwun). Slednje razdelimo na "znanje pridobljeno iz preteklega zaznavanja" (-더라 -tela) in "znanje, integrirano v osebno ali splošno razumevanje" (-지 -ci). V okviru modela večplastne shrambe spomina -네 -ney ustreza kazalcu procesa vzdrževalnega ponavljanja. Nasprotno pa -군 -kwun služi kot kazalec elaborativnega ponavljanja. Medtem -더라 -tela in -지 -ci ustrezata kazalcema priklica iz dolgoročnega spomina. Ključne besede: korejščina, evidenčni označevalci, dojemanje, izkustvenost, model večplastne shrambe spomina 148 MOON Chang-Hak 1 Introduction The theoretical concept of “evidentials” or “evidentiality” has been described as the linguistic encoding of the source of information (see Aikhenvald, 2003, 2006; Aikhenvald & Dixon, 1998, 2003, 2016; Chafe & Nichols, 1986; Comrie, 2000; Lazard, 1999; Palmer, 1998; Plungian, 2001; Willett, 1988). While all languages provide ways to signal the source of information, not all include grammatical evidentiality; indeed, the existence of lexical resources that indicate the source of information is common. For example, English demonstrates this with phrases like ‘I think,’ ‘I feel,’ ‘I see,’ ‘I hear,’ ‘they say,’ and ‘reportedly’ (see Chafe, 1986). (1) a. I think it’s a spider. (Chafe, 1986) b. I feel something crawling up my leg. (Chafe, 1986) c. I see her coming down the hall. (Chafe, 1986) d. I hear her taking a shower. (Chafe, 1986) e. They say James is a fool. (Boye, 2018) f. She was reportedly suffering from depression. The grammatical method for indicating the source of information is indicated by closed systems, which may be demonstrated through bound morphemes, affixes, and clitics. As described by Aikhenvald (2006), in Tariana, an Arawak language spoken in the multilingual area of the Vaupés in northwest Amazonia, conveying the message “José plays football” involves the indication of how the information was obtained through grammatical forms, as follows (here and elsewhere evidential morphemes in bold type). (2) a. Jusé iɾida di -manika -ka José football 3sgnf play REC.P.VIS ‘José has played football (we saw it).’ (2) b. Jusé iɾida di -manika -mahka José football 3sgnf play REC.P.NONVIS ‘José has played football (we heard it).’ (2) c. Jusé iɾida di -manika -nihka José football 3sgnf play REC.P.INFR ‘José has played football (we infer it from visual evidence).’ Direct Evidentials in Korean: From the Perspective of … 149 (2) d. Jusé iɾida di -manika -sika José football 3sgnf play REC.P.ASSUM ‘José has played football (we assume this on the basis of what we already know).’ (Aikhenvald, 2006) The main emphasis of this study will be solely on the grammatical representation of evidentiality in Korean, akin to that in Tariana. Recently, several studies have commenced investigating evidentiality phenomena within the Korean language (see Chung, 2011; Chung, 2007, 2010; Kim, 2000; Kim, 2005, 2007; Kwon, 2011, 2013; Lee, 2011, 2012; Lee, 1991, 1993; Lim & Lee, 2015; Papafragou et al., 2007; Sohn, 2018; Strauss, 2005). It is generally agreed that the Korean evidential system comprises three grammatical subclasses within inflectional suffixes: direct (- -ney1, - -kwun, - -tela, etc.), inferential (- - -keyss-, - - -ul kes-i-, etc.), and quotative/reported (- -tay, - -tamyense, etc.). (3) Direct evidential markers: - -ney, - -kwun, - -tela a. - - - ! ………………… Pakk -ey pi -ka o -ney! oustside -LOC rain -NOM come -EVI ‘[I see] it’s raining outside!’ (Sohn, 2018) b. - 2! ……………. Ne moksoli cham coh -kuwna! you voice very good -EVI ‘‘[I hear] your voice is very good!’ (Park, 2011) c. - . ……………. Chelswu moksoli cham coh -tela. ‘[I heard] Chelswu’s voice is very good.’ (Park, 2011) 1 The Yale transcription system is used for Korean transcription. 2 - -kwuna, along with “(noun+ )- ; (noun+i)-lokwuna”is a free variant form of - -kwun (see Han, 2004; Yoon, 2000). 150 MOON Chang-Hak (4) Inferential evidential markers; - - -keyss-, - - -ul kes-i- a. (Seeing the sky full of dark clouds) - - - . Kot pi -ka o -keyss -ta. soon rain -NOM come -INFR -DECL ‘[I guess] it is going to rain soon.’ (Park, 2011) b. (Based on meteorological observation data) - - - . Nayil pi -ka o -lkesi -ta. tomorrow rain -NOM come -INFR -DECL ‘[I guess] it will rain tomorrow.’ (Park, 2011) (5) Quotative/reported evidential markers; - -tay, - -tamyense a. - - - - - - - . Kyoswu -nim -un phathi -ey o -si -keyss -tay -yo. professor -HON -TOP party -to come -HON -INTEN -QUO -POL ‘The professor said he would come to the party.’ (Sohn, 2018) (5) b. - - - . …………. John -un aphu -tamyense -yo. John -TOP sick -REPO -POL ‘I heard that John is sick, is it true?’ (Sohn, 2018) This study examines direct evidential markers in Korean referring to directly acquired information from the perspective of a multi-store model of human memory. While little attention has been paid to analyzing Korean direct evidential markers from the standpoint of a multi-store memory model, this would enable a clearer elucidation of their characteristics. 2 Semantic characteristics of the direct evidential markers in Korean 2.1 The paradigmatic set of forms: - -ney, - -kwun, - -ci, - -tela This paper focuses on - -ney, - -kwun, - -ci, and - -tela, which function as direct evidential markers in Korean. Previous research has analyzed - -ney, - -kwun, and - -tela as evidential markers, whereas - -ci has not been considered as such (Sohn, 2018; Strauss, 2005). However, this study analyzes - -ci as an evidential because - -ci basically appears alongside Direct Evidentials in Korean: From the Perspective of … 151 - -ney, - -kwun, and - -tela within one paradigmatic set of forms. The morphemes - -ney, - -kwun, - -ci, and - -tela are mutually exclusive of the other morphemes that occupy the same slot in the predicational structure (Moon, 2015a, 2015b). For example, - -ney cannot co-occur with - -kwun, - -ney with - -ci, or - ( ) -te(la) with - -ney. However, it is worth noting that - ( ) -te(la) and - -kwun can exceptionally combine into forms such as - -te-kwun. Further discussion of this matter will be provided later. (6) a. ?? - - - . Pi -ka o -ney -kwun. rain -NOM come -EVI -EVI ‘It is raining!’ b. ?? - - - . Pi -ka o -ney -ci. rain -NOM come -EVI -EVI ‘It is raining!’ c. ?? - - - . Pi -ka o -ney -tela. rain -NOM come -EVI -EVI ‘It is raining!’ d. ?? - - - . Pi -ka o -kwun -ci. rain -NOM come -EVI -EVI ‘It is raining!’ e. ?? - - - . Pi -ka o -te -ney. rain -NOM come -EVI -EVI ‘It was raining!’ f. ?? - - - . Pi -ka o -te -ci. rain -NOM come -EVI -EVI ‘It was raining!’ 152 MOON Chang-Hak (7) - - - . Pi -ka o -te -kwun. rain -NOM come -EVI -EVI ‘It was raining!’ Moreover, considering the semantic interrelations and contrasts observed between - -ci, - -ney, - -kwun, and - -tela, it is reasonable to regard - -ci as an evidential marker. 2.2 “Present perception-based knowledge” versus “past acquisition- based knowledge” A contrasting linguistic relationship holds between - -ney, - -kwun, - -ci, and - -tela. (8) (Seeing that Swumi is sleeping) a. - - - - ! Icey po -ni Swumi -ka ca -ko iss -ney! now see -CONN Swumi -NOM sleep -PROG -EVI ‘Now I see, Swumi is sleeping.’ b. - - - - ! Icey po -ni Swumi -ka ca -ko iss -kwuna! now see -CONN Swumi -NOM sleep -PROG -EVI ‘Now I see, Swumi is sleeping.’ c. ?? - - - - ! Icey po -ni Swumi -ka ca -ko iss -ci! now see -CONN Swumi -NOM sleep -PROG -EVI ‘Now I see, Swumi is sleeping.’ d. ?? - - - - 3! Icey po -ni Swumi -ka ca -ko iss -tela! now see -CONN Swumi -NOM sleep -PROG -EVI ‘Now I see, Swumi is sleeping.’ 3 The use of 지금 cikum, a synonym for 이제 icey, in (8d) and (9d) can also be accepted as natural. While 지금 cikum and 이제 icey are temporal adverbs that indicate the ‘present moment’, 이제 icey, unlike 지금 cikum, implies a ‘break from the past’, (see Park, 2019; Kwuklipkwukewen). Additionally, since the time we perceive as the ‘present’ Direct Evidentials in Korean: From the Perspective of … 153 (9) (Hearing that Swumi is singing in the next room) a. - - ! Icey tul -uni Swumi moksoli -ney! now listen -CONN Swumi voice -EVI ‘Now that I listen, it sounds like Sumi’s voice.’ b. - - ! Icey tul -uni Swumi moksoli -kwun! now listen -CONN Swumi voice -EVI ‘Now that I listen, it sounds like Sumi’s voice.’ c. ?? - - ! Icey tul -uni Swumi moksoli -ci! now listen -CONN Swumi voice -EVI ‘Now that I listen, it sounds like Sumi’s voice.’ d. ?? - - ! Icey tul -uni Swumi moksoli -tela! now listen -CONN Swumi voice -EVI ‘Now that I listen, it sounds like Sumi’s voice.’ In examples (8a, b) and (9a, b), the phrases Icey po-ni ‘Now I see’ and Icey tul-uni ‘Now that I listen’ align smoothly with - -ney and - -kwun, respectively, as noted (Chang, 1985; Moon, 2014; Park, 2006; Shin, 2001). However, in examples (8c, d) and (9c, d), this alignment is not natural with - -ci and - -tela (Moon, 2010, 2015b). The grammatically appropriate alignment of - -ney and - -kwun with the expression Icey po-ni ‘Now I see’ and Icey tul-uni ‘Now that I listen’, which signify the meaning of knowledge acquired through sensory perception at speech time, suggests that - -ney and - -kwun represent the statement of knowledge acquired through present perception (Chang, 1985; Moon, 2013, 2015a, 2015c; Park, 2006). Below, we classify this knowledge as “present perception-based knowledge.” quickly becomes the past, adverbs denoting the ‘present’ can sometimes co-occur with past tense forms (see Son, 1995). As explained below, -더라 -tela in (8d) and (9d) represents ‘past acquisition-based knowledge’, and it is naturally incompatible with 이제 icey, which implies a ‘break from the past’. Conversely, 지금 cikum, which can convey a sense of ‘the past’, can naturally co-occur with -더라 -tela, representing ‘past acquisition- based knowledge’. 154 MOON Chang-Hak (10) A: - - - - ? Cinancwuey Swumi -ka mwues ha -yess -nunci al -a -yo? last week Swumi -NOM waht do -PAST -NMLZ know -INTER -POL ‘Are you aware of what Swumi did last week?’ a. B: ?? , - - . Un, Swumi -nun cinancwuey yakwu ha -yess -ney. yes, Swumi -TOP last week baseball do -PAST -EVI ‘Yes, (I knew) Swumi played baseball last week.’ b. B: ?? , - - . Un, Swumi -nun cinancwuey yakwu ha -yess -kwuna. yes, Swumi -TOP last week baseball do -PAST -EVI ‘Yes, (I knew) Swumi played baseball last week.’ c. B: , - - . Un, Swumi -nun cinancwuey yakwu ha -yess -ci. yes, Swumi -TOP last week baseball do -PAST -EVI ‘Yes, (I knew) Swumi played baseball last week.’ d. B: , - . Un, Swumi -nun cinancwuey yakwu ha -tela. yes, Swumi -TOP last week baseball do -EVI ‘Yes, (I knew) Swumi played baseball last week.’ On the contrary, in examples (10a, b), the question ? al-a-yo? ‘Do you know?’ did not align smoothly with - -ney and - -kwun, respectively. However, in (10c, d) this alignment occurred naturally with - -ci and - -tela. The grammatically appropriate correspondence of - -ci and - -tela with ? al-a-yo? ‘Do you know?’ depends upon whether one already knew the fact, implying that - -ci and - -tela signify statements of knowledge already acquired in the past. Below, we classify this knowledge as “past acquisition-based knowledge.” The observations above reveal that in Korean, - -ney, - -kwun, - -ci, and - -tela function as direct evidential markers, differentiating between whether the source of information is derived from “present perception- based knowledge” or “past acquisition-based knowledge.” Direct Evidentials in Korean: From the Perspective of … 155 2.3 “Unaccepted present perception-based knowledge” versus “accepted present perception-based knowledge” The observations above reveal that in Korean, - -ney, - -kwun, - -ci, and - -tela function as direct evidential markers, differentiating between whether the source of information is derived from “present perception- based knowledge” or “past acquisition-based knowledge.” Although both - -ney and - -kwun commonly signify “present perception-based knowledge,” they demonstrate divergent linguistic phenomena, as outlined below. (11) (Looking at the watch at work) a. - - - . Eme pelsse hansi -ka nem -ess -ney. oh dear already one o'cock -NOM pass -PAST -EVI ‘Oh dear! (I see) it’s already past one.’ b. ? - - - 4. Eme pelsse hansi -ka nem -ess -kwuna. oh dear already one o'cock -NOM pass -PAST -EVI ‘Oh dear! (I see) it’s already past one.’ (12) a. - - - . As pelsse yenghwa -ka sicaktoy -ess -ney. wow already movie -NOM start -PAST -EVI ‘Wow! (I see) the movie has already started.’ b. ? - - - . As pelsse yenghwa -ka sicaktoy -ess -kwuna wow already movie -NOM start -PAST -EVI ‘Wow! (I see) the movie has already started.’ In examples (11a) and (12a), the astonishment-expressing discourse markers eme ‘oh dear!’ and as ‘wow!’ align smoothly with - -ney, as 4 어머 eme ‘Oh dear!’ or 앗 as ‘wow!’ could possibly co-occur with -군 -kwun in (11b) and (12b), however, the occurrence is most likely when a pause follows 어머 eme ‘Oh dear!’ or 앗 as ‘wow!’. In this way, 어머 eme ‘Oh dear!’ or 앗 as ‘wow!’ are separated by a pause, and -군 -kwun co-occurs with the pause. The pause functions similarly to the understanding-expressing discourse marker 아아 aa- ‘Ah-’. Thus, the grammaticality judgments for (11b) and (12b) assume no pause following 어머 eme ‘Oh dear!’ or 앗 as ‘wow!’, consistent with the assumptions for (11a) and (12a). 156 MOON Chang-Hak previously observed. However, in (11b) and (12b), this alignment is not natural with - -kwun (Moon, 2014). The grammatically appropriate alignment of - -ney with the discourse markers eme ‘oh dear!’ and as ‘wow!’, which indicate that “present perception-based knowledge” is not understood yet, suggests that - -ney indicates that the statement is not accepted as reliable knowledge. Below, we classify this knowledge as “unaccepted present perception-based knowledge.” (13) (When someone is approaching without knowing who they are) a. ?? - - - - . Aa tangsin -i Swumi -ssi -i -ney -yo. ahaa you -NOM Swumi -HON -be -EVI -POL ‘Ahaa, (I realize) You are Swumi!’ b. - - - . Aa tangsin -i Swumi -ssi -i -kwun -yo ahaa you -NOM Swumi -HON -be -EVI -POL ‘Ahaa, (I realize) You are Swumi!’ (14) (While viewing the artwork of the acclaimed artist at the art gallery) a. ?? - - - - Cakphwum -ul Po -ni somwun -ey tutten -taylo artwork -ACC See -CONN rumor -INST hear -CONN a. - - - 5. i salam -un kwayen hwullyunghan yeyswulka -i -ney this person -TOP indeed good artist -be -EVI ‘After seeing his work, I must say, he indeed is as good an artist as they say!’ 5 It is possible that 과연 kwayen ‘indeed’ and -네 -ney co-occur in (14a). If so, it would be due to the accompanying evaluative expression 훌륭하다 hwulyunghata ‘good’. In certain evaluative predicates, the evaluative meaning seems to neutralize the “unaccepted” meaning of -네 -ney and the “accepted” meaning of -군 -kwun. For example, 영화가 너무나 감동적이{네/군}. Yenghwa-ka cengmal kamtongcek-i-{ney/kwun} ‘The movie is truly touching.’ However, as in (14b), the meaning of 과연 kwayen ‘indeed’ makes -군 -kwun more natural than -네 -ney. Direct Evidentials in Korean: From the Perspective of … 157 (14) b. - - - - Cakphwum -ul po -ni somwun -ey tutten -taylo artwork -ACC see -CONN rumor -INST hear -CONN a. - - . i salam -un kwayen hwullyunghan yeyswulka -lokwuna this person -TOP indeed good artist -EVI ‘After seeing his work, I must say, he indeed is as good an artist as they say!’ (Kwuklipkwukewen) On the contrary, examples (13b) and (14b) show that the understanding- expressing discourse markers aa- ‘Ah-’ and kwayen ’indeed’ align smoothly with - -kwun. However, in (13a) and (14a), this alignment does not occur naturally with - -ney (Moon, 2014). The grammatically-fitting correspondence of - -kwun with the discourse markers aa- ‘Ah-’ and kwayen ’indeed’ indicates that what is currently perceived matches with what was previously thought (Choi, 2000; Kwuklipkwukewen), implying that - -kwun denotes a statement that is accepted as reliable knowledge. Below, we categorize this knowledge as “accepted present perception-based knowledge.” The widely used evidential markers - -ney and - -kwun, which generally represent present perception-based knowledge, can be divided based on whether the conveyed information is “unaccepted present perception-based knowledge” or “accepted present perception-based knowledge.” 2.4 “Knowledge derived from past perception” versus “knowledge integrated into personal or general understanding” As confirmed in (10c, d), both - -ci and - -tela generally indicate “past acquisition-based knowledge.” However, - -ci and - -tela exhibit contrasting linguistic phenomena, as discussed below. 158 MOON Chang-Hak (15) (Seeing that Swumi was sleeping yesterday) a. ?? - - - - - . Ecey po -ni Swumi -ka ca -ko iss -yess -ci yesterday see -CONN Swumi -NOM sleep -PROG -PAST -EVI ‘Yesterday I saw that Swumi was sleeping.’ b. - - - - . Ecey po -ni Swumi -ka ca -ko iss -tela yesterday see -CONN Swumi -NOM sleep -PROG -EVI ‘Yesterday I saw that Swumi was sleeping.’ (16) (Hearing that Swumi was singing in the next room yesterday) a. ?? - - - - - . Ecey tul -ui Swumi -ka nolay pwulu -ko iss -yess -ci yesterday listen -CONN Swumi -NOM song sing -PROG -PAST -EVI ‘Yesterday I heard that Swumi was singing a song.’ b. - - - - . Ecey tul -uni Swumi -ka nolay pwulu -ko iss -tela yesterday listen -CONN Swumi -NOM song sing -PROG -EVI ‘Yesterday I heard that Swumi was singing a song.’ In examples (15b) and (16b), the phrases ecey po-ni ‘Yesterday I saw’ and ecey tul-uni ‘Yesterday I heard’ align smoothly with - -tela, as noted. However, in (15a) and (16a), this alignment is not natural for - -ci. In examples (15a, b) and (16a, b), only - -tela, which aligns grammatically with the past perception and awareness expressed in ecey po-ni ‘Yesterday I saw’ and ecey tul-uni ‘Yesterday I heard’, indicates knowledge acquired through sensory perception before speech time within the domain of “past acquisition-based knowledge.” However, in example (10c), where expressions like ecey po-ni ‘Yesterday I saw’ and ecey tul-uni ‘Yesterday I heard’ are absent, - -ci categorizes the information as past acquisition-based knowledge, regardless of whether it was perceptually acquired in the past, obtained from someone else, or inferred by oneself. Thus, - -ci does not merely signify knowledge perceived at a specific past moment but encompasses a broad range of knowledge acquired in the past, including personal comprehension and general understanding. Below, we categorize the knowledge indicated by - Direct Evidentials in Korean: From the Perspective of … 159 -tela as “knowledge derived from past perception” and the one indicated by - -ci as “knowledge integrated into personal or general understanding6.” The details discussed above are summarized in Table 1. Table 1: Semantic characteristics of the direct evidential markers in Korean Meaning Present perception based knowledge (PPBK) Past acquisition based knowledge (PABK) Affixes Unaccepted PPBK Accepted PPBK Knowledge derived from past perception Knowledge integrated into personal or general understanding - -ney ○ – – – - -kwun – ○ – – - -tela – – ○ – - -ci – – – ○ 3 The multi-store memory model In this section, we present the theory proposed by Atkinson and Shiffrin (1968) to examine the meaning of evidential markers in Korean within the framework of the multi-store memory model, as discussed in Section 2. We opted for this approach because the theory outlined by Atkinson and Shiffrin (1968), as a representative multi-store memory model, would appear to offer a comprehensive account of the features associated with evidential markers in Korean. Human memory involves the organized gathering of information and experiences and the intention to retrieve them in the future. The multi-store memory model posited by Atkinson and Shiffrin (1968) elucidates human memory, postulating three distinct memory stores through which 6 Due to the characteristics of -지 -ci and -더라 -tela, -더라 -tela clearly conveys a past meaning, while -지 -ci tends to obscure the past meaning. Therefore, to explicitly indicate a past meaning, -더라 -tela does not require the past tense morpheme -었 -ess-, whereas -지 -ci needs to be accompanied by -었 -ess-. For this reason, -었 -ess- is used in (10c), (15a), (16a), and (18a), but not in (10d), (15b), (16b), and (18b). 160 MOON Chang-Hak information is sequentially transferred in a linear sequence. These three components of human memory are detailed as follows (Alsaeed, 2017; Atkinson & Shiffrin, 1968, 1971; Braisby & Gellatly, 2012; Eysenck & Keane, 2020; Groome, 1999; Hitch, 2005). Figure 1: The multi-store memory model (Adopted from Atkinson & Shiffrin, 1968; Alsaeed, 2017) The three primary stores are sensory memory, short-term memory (STM), and long-term memory (LTM), which differ in terms of information processing (encoding), storage capacity, and the duration for which information can be retained. Information flows sequentially from one store to another in a linear manner, resembling an information processing model consisting of the input, process, and output stages. Initially detected by sensory organs, information enters the sensory memory, where it is temporarily retained as sensory stimuli. Upon selective attention, information progresses to short-term memory (STM); through elaborative rehearsal, if endowed with meaning, it is transferred to long-term memory (LTM). Sensory stores continuously receive a stream of information, yet the majority passes unnoticed and resides briefly in the sensory register. Within the sensory memory store, information is inputted from all five senses: sight (visual information), sound (auditory information), taste (gustatory information), smell (olfactory information), and touch (tactile information). Despite the expansive capacity of sensory memory stores, their durations are remarkably brief. It can encode information from any sensory modality, although much of this information dissipates through decay. Attention marks the initial stage of the remembering process. When an individual’s Direct Evidentials in Korean: From the Perspective of … 161 focus is directed toward one of the sensory stores, information is subsequently transferred to short-term memory (STM). Short-term memory7 (STM) refers to the memories held in conscious awareness that are currently receiving attention. In short-term memory (STM), maintenance rehearsal refers to the verbal or mental repetition of information. This type of rehearsal typically involves the repetition of information without considering its meaning or linking it to other information. Through continual rehearsal, the information in the memory trace is “regenerated” or “renewed,” its strength enhanced as it is transferred to the long-term memory store. Without maintenance rehearsal, however, information is susceptible to being forgotten and lost from short- term memory through displacement or decay. Long-term memory (LTM) denotes the memories not presently held in conscious awareness but stored and able to be recalled, specifically encompassing declarative knowledge associated with “knowing that” and procedural knowledge associated with “knowing how.” When information is endowed with meaning through elaborative rehearsal, it proceeds to long- term memory (LTM). Elaborative rehearsal involves associating new information with existing knowledge stored in long-term memory in a meaningful manner. Compared to maintenance rehearsal, elaborative rehearsal is more efficacious in the retention of new information by facilitating robust encoding. This process signifies a higher level of information processing. Retrieval failure occurs when information resides in the long-term memory but cannot be accessed. Such information is considered available (i.e., it remains stored) but inaccessible (i.e., it cannot be retrieved). The inability to access this information stems from the absence of retrieval cues. This is clearly depicted in the scheme in Figure 2 proposed by Atkinson and Shiffrin (1971). 7 A concept closely associated with short-term memory is Baddeley’s (1986, 2003) working memory model. Although working memory predominantly emphasizes cognitive task performance and control functions in the context of short-term information retention, it would not be unreasonable to regard it primarily as short-term memory, especially in the analysis of evidential markers in Korean. 162 MOON Chang-Hak Figure 2: The multi-store memory model (Adopted from Atkinson & Shiffrin, 1971; Hitch, 2005) 4 Korean direct evidential markers in the multi-store memory model In this Section, we analyze the Korean direct evidential markers discussed in Section 2 through the lens of the multi-store memory model explored in Section 3. As observed in (8a), (9a), (11a), and (12a), - -ney signifies the unaccepted present perception-based knowledge accompanied by surprise, which is not understood yet. The presence of the - -ney suggests surprise because although the sensory organs receive environmental stimuli and attention is allocated, the individual is verbally rehearsing the information for retention without contemplating its meaning or linking it with other information. Therefore, it can be inferred that the information conveyed by - -ney represents the transition from sensory memory to short-term memory, indicating the initiation of the “maintenance rehearsal” process. As noted in (8b), (9b), (13b), and (14b), - -kwun signifies “accepted present perception-based knowledge” accompanied by understanding, indicating that what is currently perceived matches what was previously thought. The presence of - -kwun indicates understanding; since the sensory organs receive environmental stimuli, attention is directed and individuals attribute meaning by linking it with existing knowledge. Thus, it can be deduced that the information signified by - -kwun signifies the transfer from short-term to long-term memory, marking the onset of the “elaborative rehearsal” process. Direct Evidentials in Korean: From the Perspective of … 163 As seen in (10c) and (10d), it was noted that - -ci and - -tela signify “past acquisition based-knowledge.” A linguistic phenomenon demonstrates that the information represented by - -ci and - -tela is closely related to the information stored in long-term memory (LTM). (17) (While searching for Suwmi’s glasses, I cannot remember where Suwmi placed them.) a. - - - - - ? Swumi -ka ankyeng -ul eti -ey twu -yess -ci? Swumi -NOM glasses -ACC where -LOC put -PAST -EVI ‘Where did Suwmi put her glasses?’ b. - - - - - ? Swumi -ka ankyeng -ul eti -ey twu -yess -tela? Swumi -NOM glasses -ACC where -LOC put -PAST -EVI ‘Where did Suwmi put her glasses?’ c. ?? - - - - - ? Swumi -ka ankyeng -ul eti -ey twu -yess -ney? Swumi -NOM glasses -ACC where -LOC put -PAST -EVI ‘Where did Suwmi put her glasses?’ d. ?? - - - - - ? Swumi -ka ankyeng -ul eti -ey twu -yess -kwun? Swumi -NOM glasses -ACC where -LOC put -PAST -EVI ‘Where did Suwmi put her glasses?’ (18) (While searching for Suwmi’s glasses, I suddenly remember where she placed them.) a. - ! - - - . kulay mac -a! chaksang wi -ey twu -yess -ci. Yeah! right -DECL desk on -LOC put -PAST -EVI ‘Yeah! That’s right. Suwmi left her glasses on the desk.’ b. - ! - - . kulay mac -a! chaksang wi -ey twu -tela. Yeah! right -DECL desk on -LOC put -EVI ‘Yeah! That’s right. Suwmi left her glasses on the desk.’ 164 MOON Chang-Hak c. ?? - ! - - - . kulay mac -a! chaksang wi -ey twu -yess -ney. Yeah! right -DECL desk on -LOC put -PAST -EVI ‘Yeah! That’s right. Suwmi left her glasses on the desk.’ d. ?? - ! - - - . kulay mac -a! chaksang wi -ey twu -yess -kwun. Yeah! right -DECL desk on -LOC put -PAST -EVI ‘Yeah! That’s right. Suwmi left her glasses on the desk.’ The suffixes - -ci and - -tela in (17a) and (17b), where the syntactically matched wh-questions denote a partial lack of information, can suggest the partial forgetting of information. In contrast, - -ney and - -kwun in (17c) and (17d) cannot convey the partial forgetting of information as they are not compatible with wh-questions. This linguistic phenomenon can be regarded as directly reflecting the manifestation of retrieval failure, where information is retained in the long-term memory but remains inaccessible. Furthermore, the suffixes - -ci and - -tela in (18a) and (18b), when grammatically aligned with ! kulay maca! ‘Yeah! That’s right’, signifying the recollection of something that was not well remembered, can indicate success in locating the information. In contrast, - -ney and - -kwun in (18c) and (18d) cannot convey success in locating information as they do not grammatically align with ! kulay maca! ‘Yeah! That’s right’. This linguistic phenomenon directly mirrors the occurrence of successful retrieval, in which the information stored in long-term memory remains accessible. Therefore, it can be inferred that the information conveyed by - -tela and - -ci represents the transition from long-term memory to short-term memory, indicating the initiation of the “retrieval process.” Regarding (15a, b) and (16a, b), it - -tela specifically represents “knowledge derived from past perception,” in contrast to - -ci. Therefore, one might interpret the meaning of - -tela as being exclusive to that of - -ci. However, a more accurate understanding suggests that the relationship between - -tela and - -ci is not exclusive, but inclusive. In other words, while - -ci signifies comprehensive “past acquisition-based knowledge,” including personal comprehension or general understanding, - -tela specifically denotes “knowledge derived from past perception.” Distinguishing knowledge obtained through past perception among other types of “past acquisition-based knowledge” likely indicates the more reliable information. Direct Evidentials in Korean: From the Perspective of … 165 Here, we briefly examine the exceptional phenomenon in which, as seen in Example (7) in Section 2.1, - -tela and - -kwun co-occur. As observed above, - -tela denotes “knowledge derived from past perception,” representing the transition from long-term memory to short- term memory, thereby indicating the initiation of the retrieval process. Conversely, - -kwun signifies “accepted present perception-based knowledge,” indicating a transfer from short-to long-term memory, thus marking the onset of the elaborative rehearsal process. Based on this understanding, when considering the co-occurrence phenomenon of - ( ) -te(la) and - -kwun, it can be interpreted that after “knowledge derived from past perception” is retrieved from long-term memory, it is brought into short-term memory for elaborative rehearsal. Thus, the co-occurrence of - -tela and - -kwun more clearly demonstrates the elaborative rehearsal associated with existing knowledge in short-term memory. The content provided above can be concisely illustrated in the subsequent diagram. Figure 3: The correspondence between direct evidential markers in Korean and multi-store memory 166 MOON Chang-Hak 5 Conclusion This study examined the semantic characteristics of direct evidential markers in Korean, namely - -ney, - -kwun, - -tela, and - -ci, using a multi-store memory model framework. In summary, - -ney signifies “unaccepted present perception-based knowledge,” aligning with the “maintenance rehearsal” process during the transition from sensory memory to short-term memory; - -kwun indicates “accepted present perception-based knowledge,” corresponding to the “elaborative rehearsal” process during the transfer from short-term memory to long-term memory; - -ci represents “knowledge integrated into personal or general understanding,” reflecting stable settlement in long- term memory for retrieval and recall; and finally, - -tela specifically denotes “knowledge derived from past perception” within long-term memory. In this study, we elucidated the characteristics of evidential markers in Korean more clearly by explaining their meaning within the framework of Atkinson and Shiffrin’s (1968) multi-store memory model. In the future, we aim to empirically validate these analytical findings through cognitive psychological experiments. Acknowledgments This work was supported by Hankuk University of Foreign Studies Research Fund of 2024. Abbreviations 3 Third person ACC Accusative ASSUM Assumed CONN Connectives DECL Declarative EVI Evidentials HON Honorifics INFR Inferential INTER Interrogative LOC Locative nf Non-feminine NMLZ Nominalizer Direct Evidentials in Korean: From the Perspective of … 167 NONVIS Non-visual P Past PAST Past tense PROG Progressive QUO Quotative REC Reciprocal REPO Reported sg Singular TOP Topic VIS visual References Aikhenvald, A. I., & Dixon, R. M. W. (2016). The grammar of knowledge: A cross- linguistic typology. Oxford: Oxford University Press. Aikhenvald, A. Y. (2003). Evidentiality in Tariana. Amsterdam: John Benjamins. Aikhenvald, A. Y. (2006). Evidentiality. Oxford: Oxford Univ. Press. Aikhenvald, A. Y., & Dixon, R. M. W. (1998). Evidentials and areal typology: A case study from Amazonia. Language Sciences, 20, 241-257. https://doi.org/10.1016/S0388-0001%2898%2900002-3 Aikhenvald, A. Y., & Dixon, R. M. W. (2003). Studies in evidentiality (1st ed.). Amsterdam: John Benjamins. Alsaeed, N. H. (2017). Wish you were here: A psychological analysis using Atkinson-Shiffrin memory mode. Journal of Literature and Art Studies, 7(5), 521- 527. https://doi.org/10.17265/2159-5836%2F2017.05.004 Atkinson, R. C., & Shiffrin, R. M. (1968). Chapter: Human memory: A proposed system and its control processes. In K. W. Spence & J. T. Spence (Eds.), The psychology of learning and motivation (Volume 2) (pp. 89-195). New York: Academic Press. Atkinson, R. C., & Shiffrin, R. M. (1971). The control of short-term memory. San Francisco: Freeman. Baddeley, A. D. (1986). Working memory. Oxford: Oxford University Press. Baddeley, A. D. (2003). Working memory and language: An overview. Journal of Communication Disorders, 36(3), 189-208. https://doi.org/10.1016/S0021- 9924(03)00019-4 168 MOON Chang-Hak Boye, K. (2018). Evidentiality: The Notion and the Term. In A. Y. Aikhenvald (Ed.), The Oxford Handbook of Evidentiality (pp. 261-272). Oxford: Oxford University Press Braisby, N., & Gellatly, A. (2012). Cognitive Psychology. Oxford: Oxford University Press. Chafe, W. (1986). Evidentiality in English conversation and academic writing. In W. L. Chafe & J. Nichols (Eds.), Evidentiality: The linguistic coding of epistemology (pp. 261-272). Norwood, NJ: Ablex. Chafe, W., & Nichols, J. (Eds.). (1986). Evidentiality: The linguistic coding of epistemology. Norwood, NJ: Ablex. Chang, K. (1985). Hyuetay kwukeyu yangtaypemcu yenkyu [The study of modality category in Korean]. Seoul: Thopchwulphansa. Choi, H. (2000). Hyentay kwuke kamthansayu pwuncel kwuco yenkwu [A study of the disjunctive structure of Modern Korean discourse markers]. Hankwuke nayonglon [The Contents Theory in Korean], 7, 361-408. Chung, J. Y. (2011). Uncommon common grounds and the Korean reportative evidential -tay. LSA Annual Meeting Extended Abstracts, 2(35), 1-5. Chung, K. (2007). Spatial deictic tense and evidentials in Korean. Natural Language Semantics, 15(3), 187-219. https://doi.org/10.1007/s11050-007-9017-8 Chung, K. (2010). Korean evidentials and assertion. Lingua, 120(4), 932-952. https://doi.org/10.1016/j.lingua.2009.06.006 Comrie, B. (2000). Evidentials: Semantics and history. In L. Johanson & B. Utas (Eds.), Evidentials: Turkic, Iranian and Neighbouring Languages (pp. 1-12). Dordrecht: De Gruyter. Eysenck, M. W., & Keane, M. T. (2020). Cognitive psychology: A student’s handbook. Hove, UK: Psychology Press. Groome, D. (1999). An introduction to cognitive psychology: Processes and disorders. Hove, UK: Psychology Press. Han, K. (2004). Hyentay wulimaluy machimssikkuth yenkyu [A study on Final Endings in Koeran]. Seoul: Yeklak. Hitch, G. J. (2005). Working memory. In N. Braisby & A. Gellatly (Eds.), Cognitive Psychology (pp.307-341). Oxford: Oxford University Press. Kim, M. S. (2005). Evidentiality in achieving entitlement, objectivity, and detachment in korean conversation. Discourse Studies, 7(1), 87-108. https://doi.org/10.1177/1461445605048768 Direct Evidentials in Korean: From the Perspective of … 169 Kim, M. S. (2007). Evidential strategies in Korean conversation: An analysis of interactional and conversational narrative functions. Los-Angeles: University of California. Kim, N. (2000). Reportative evidential in Korean. Korean Linguistics, 10, 105-124. https://doi.org/10.1075/KL.10.04NKK Kwon, I. (2011). Mental spaces in the Korean reportive /quotative evidentiality marker -ay. Tamhwa Wa Inji, 18(2), 23-50. Kwon, I. (2013). Evidentiality in Korean conditional constructions. Linguistics, 51(6), 1249-1270. https://doi.org/10.1515/ling-2013-0049 Kwuklipkwukewen. . Phyocwunkwuketaysacen [Standard Korean Dictionary], https://stdict.korean.go.kr/main/main.do Lazard, G. (1999). Mirativity, evidentiality, mediativity, or other? Linguistic Typology, 3(1), 91-110. https://doi.org/10.1515/lity.1999.3.1.91 Lee, C. (2011). Evidentials and epistemic modal in Korean: Evidence from their interactions. Proceedings of the 24th Pacific Asia conference on language, information and computation. Waseda University, 193-202. Lee, C. (2012). Evidentials and modals: What makes them unique. Sprache und Datenverarbeitung, International Journal for Language Data Processing, 71-98. Lee, H. S. (1991). Tense, aspect, and modality: A discourse-pragmatic analysis of verbal affixes in 169orean from a typological perspective. University of California, Los Angeles. Lee, H. S. (1993). Cognitive constraints on expressing newly perceived information, with reference to epistemic modal suffixes in Korean. Cognitive Linguistics, 4(2), 135-168. https://doi.org/10.1515/cogl.1993.4.2.135 Lim, D., & Lee, C. (2015). Perspective shifts of Korean evidentials and the effect of contexts. Proceedings from Semantics and Linguistic Theory, 22, 26. http://dx.doi.org/10.3765/salt.v0i0.2648 Moon, C. (2015c). A contrastive study of Korean and Japanese declarative sentence-final forms: From the perspectives of semantics and pragmatics. Information, 18(5(A)), 1609. Moon, C. (2014). Hyentay hankwuke congkyelemi -ney(yo) wa -kwun(yo) - ( ) - ( ) [The sentence-ending suffixes Ney and Kwun in Korean]. Enewa enehak [Language and Linguistics], 64, 83-110. Moon, C. (2010). Gendai kankogono syuketsugobi -ci(yo) no tagisei -ci(yo) [The polysemy of sentence final ending -ci(yo) in Korean]. Kanagawadaigakugengokenkyu [Kanagawa University Center for Language Studies], 33, 45-63. 170 MOON Chang-Hak Moon, C. (2013). Nihongo to kankokugo no tikakuhyomeibun ni kansuru taisyokenkyu : hinoda, noda to ney and kwun 対 [The contrastive study on the statement of perception in Japanese and Korean: Non-noda, noda and ney and kwun]. Ilponehakyenkwu [The Japanese Language Association of Korea], 36, 117-130. Moon, C. (2015a). Kaiwatai no noda to hay(yo)tai no syuketugobi ni kansuru nikkantaisyokenkyu-evidentiality no kanten kara- ( ) 対 (evidentiality) ‐ [The contrastive study on noda of colloquial style in Japanese and the final suffixes of hay(yo) style in Korean: From the perspective of evidentiality]. Ilponenemwunhwa [Journal of Japanese Language and Culture], 32, 81-102. https://doi.org/10.17314/jjlc.2015..32.005 Moon, C. (2015b). Ninsikiteki modariti no tyukakukeisiki ni kansuru nikkantaisyokenkyu-syokoteki modariti(evidential modality) no kanten wo (evidential modality)kuwaete- 対 拠 (evidential modality) ‐ The contrastive study on core forms of epistemic modality in Japanese and Koean: Including the prespective of evidential modality. Ilponenemwunhwa [Journal of Japanese Language and Culture], 30, 107-127. https://doi.org/10.17314/JJLC.2015..30.006 Palmer, F. R. (1998). Mood and modality. Cambridge: Cambridge University Press. Papafragou, A., Li, P., Choi, Y., & Han, C. (2007). Evidentiality in language and cognition. Cognition, 103(2), 253-299. https://doi.org/10.1016/j.cognition.2006.04.001 Park, J.H. (2011). Hankwukeyese cungkesengina uyoysenguy uimisengpunul phohamhanun mwunpepyoso [Grammatical elements containing evidential or mirative components in Korean]. Enewa cengposahoy [Language & Information Society], 15, 1-25. Park, J.Y. (2006). Hankwuke yangtay emi yenkwu [A Study of suffixes of modality in Korean]. Seoul: Tayhaksa. Park, Y.J. (2019). ‘Cikum’kwa ‘ice’uy uymi pyenpyelkwa kyoyuk pangan yenkwu ‘ ’ ‘ ’ [A Study on Semantic Differentation and Educational Methodology between ‘jigeum’ and ‘ije’], Wulimalkul [Korean], 83, 65-98. Plungian, V. A. (2001). The place of evidentiality within the universal grammatical space. Journal of Pragmatics, 33(3), 349-357. https://doi.org/10.1016/S0378- 2166(00)00006-0 Direct Evidentials in Korean: From the Perspective of … 171 Shin, S. (2001). -Kwun(yo) and -ney(yo)ui ssuimye kwanhan yenkywu ‘ ( ) ‘- ( )’ [A study of the usage of -kwun(yo) and -ney(yo)]. Morphology, 3(1), 69-84. Sohn, H. (2018). Evidentiality in Korean. In A. Y. Aikhenvald (Ed.), The Oxford handbook of evidentiality (pp. 693-708) Oxford: Oxford University Press. https://doi.org/10.1093/oxfordhb/9780198759515.013.33 Son, N.I. (1995). Kwuke pwusa yenkwu [A Study of Adverbs in Korean]. Seoul: Pakiceng. Strauss, S. (2005). Cognitive realization markers in Korean: A discourse-pragmatic study of the sentence-ending particles -kwun, -ney and -tela. Language Sciences (Oxford), 27(4), 437-480. https://doi.org/10.1016/j.langsci.2004.09.014 Willett, T. (1988). A cross-linguistic survey of the grammaticization of evidentiality. Studies in Language, 12(1), 51-97. https://doi.org/10.1075/sl.12.1.04wil Yoon, S.M. (2000). Hyentaykwukeuy mwuncangcongkyelpep yenkwu [A Study on Sentence Endings in Contemporay Korean]. Seoul: Cipmwuntang.