Keimyung University, Academia Koreana
  • Shifts in Characterization in Literary Translation:Representation of the "I"-Protagonist of Yi Sang's Wings

Characterization (i.e., the way a character is lexicogrammatically constructed in a text) introduces and develops the reader's mental representation of the character. Previous research shows that in fiction various means of characterization have been used to encode the character's views of the world, the relationship between characters, and the underlying theme of the story. Considering the importance of characterization in fiction, this study aims to find out what effects the translator's lexicogrammatical choices may have on characterization and the implied reader's construction of the character. For this purpose, Yi Sang's Nalgae (Wings) and its three English translations were compared in terms of how the "I"-protagonist is characterized in four key passages of the story. The theoretical framework for this study is transitivity, a Hallidayan system of how experiential meaning is represented in the clause.


Korean literature, literary translation, character construction, systemic functional grammar, transitivity

In fiction, character plays a crucial role in developing fictional incidents and themes and affects how the story is received.2 James (1972: 200) argues in his seminal essay on Anthony Trollope that "character […] is action, and action is [End Page 283] plot, and any plot […] plays upon our emotion." Culler (1975: 230) also points out that "for many readers character serves as the major totalizing force in fiction." Harvey (1968: 23) even asserts that "most great novels exist to reveal and explore character."

Character is shaped through characterization. The term characterization refers to "the depicting, in writing, of clear images of a person, his [his or her] actions and manners of thought [… that] go to make [the] people what they are" (Thrall and Hibbard 1936: 74–75; quoted in Chatman 1978: 107). In principle, characterization can be constructed by any element in the text, and thus skillful writers use various means of characterization to make their characters clear to the readers. Lothe (2000: 81) succinctly explains the relationship between character and characterization as follows:

[T]he distinction between character (at the level of story) and characterization (at the textual level) is not absolute […]. Discussions of fictional characters become more convincing if they refer to, and are based on, characterization, for it is through such characterization that the characters are introduced, shaped, and developed.

Characterization is constructed in a way appropriate to thematic/character development. For example, in the climactic scene of Joseph Conrad's The Secret Agent, where Mrs. Verloc kills her husband with a carving knife, she is portrayed as an insensate being whose actions rarely influence her surroundings, while her husband is described as an acting person who is (paradoxically) keenly aware of his environment (Simpson 2004: 76). In Sheila Watson's The Double Hook, the central character is encoded as undergoing a dramatic transformation from a matricide to a good upright man, contributing to the fictional theme of 'building an integrated community' (Ji and Shen 2004). Similar examples can be found in various studies, especially those in the fields of literary criticism and stylistics.

Characterization may be re-constructed through translation. It has been shown that translation concerns choosing words under linguistic and sociocultural constraints and thus entails various levels of "shifts" (see, e.g., Catford 1965; Vinay and Darbelnet 1995; Chesterman 1997). Importantly, these shifts may occur in the lexical and grammatical levels of the text, pertaining to characterization. Shifts in characterization may in turn cause the source text (ST) and target text (TT) readerships to develop different views and feelings of the same character.

This study aims to reveal potential effects of the translator's lexicogrammatical choices on characterization and the implied reader's construction of character. For this purpose, the researcher chose Yi Sang's short story Nalgae and its three English translations, and compared the "I"-protagonist in four key passages of [End Page 284] the story, using M. A. K. Halliday's system of transitivity (Halliday 1985, 1994; Halliday and Matthiessen 2004). A comparative analysis of a great literary work and its international versions could reveal an important aspect of how narrative fiction should and can be translated.

This article is divided into five sections. The next section presents Halliday's transitivity as the theoretical framework for this study and discusses previous research on characterization-in-transitivity and the character development of Nalgae. The third section of the article briefly introduces the research method. The fourth section then shows in transitivity terms how the "I"-protagonist is characterized in the ST and TTs. Finally, the fifth section concludes with a brief discussion of the influence of characterization-in-transitivity on translation quality.


Transitivity in systemic functional grammar

The theoretical framework for this study is transitivity, a system for describing the experiential aspect of the clause according to systemic functional grammar (SFG). SFG is a theory of lexicogrammar developed by M. A. K. Halliday who began to describe language systematically in the early 1960s. Halliday (1994) argues that through language all facets of human experience can be transformed into meanings and that the lexicogrammar of every language is dedicated to experiential metafunction. The experiential metafunction construes all types of human experience as transitivity, a systematic way of constructing clausal meaning in the frame who does what to whom under what circumstances.

Transitivity consists of three components: process, participant, and circumstance (Halliday 1994; Halliday and Matthiessen 2004). The process is, in traditional terms, the verbal phrase, and is the most central element of a transitivity configuration. The participant is an entity involved in the process and is realized typically by the noun phrase. The circumstance is time, place, means, reason, or other background information which can be expressed by the adverbial/prepositional phrase. The following is an example of the 'participant^process^circumstance' configuration.3

John         studied         in his room.

participant         process         circumstance [End Page 285]

Halliday (1994) identifies six types of process: material, mental, relational, behavioral, verbal, and existential. The material process, the most salient of the six types, is the process of 'doing' (action) or 'happening' (event). In the actionrelated material clause, the participant representing the doing is called the actor, while the affected participant is referred to as the goal. The mental process is the process of 'sensing,' which relates to something that goes on in one's inner world. Where a mental process appears, the conscious being and the thing sensed are called the senser and the phenomenon, respectively. The relational process posits a relationship of equivalence ('X is Y'), possession ('X has Y'), or circumstance ('X is at Y'). In the attributive-relational clause (e.g., the weather is nasty), the participant is the carrier (the weather), while the quality ascribed to the participant is the attribute (nasty). The behavioral process is the process that embodies physiological or psychological behavior such as breathing, staring, and dreaming. It sits halfway between the material and mental process. The key participant in the behavioral clause is the behaver. The verbal process is the process of 'saying' realized by the verbal group. The participant roles associated with verbalization are the sayer (the producer of saying), the receiver (the one to which the saying is directed), and the verbiage (that which gets said). Finally, the existential process asserts that something exists. It typically includes the word there as a dummy subject, as in There is a piano. The only participant role in the existential clause is called the existent (a piano).

The transitivity system provides various ways of capturing one and the same situation. For example, in the clause The street runs between two religious sites, the material process runs actually functions as a relational process (e.g., 'is'), while giving "the stative description a different tone [action coloring], especially if there are a number of similar choices in that area of the text" (Thompson 2004: 117). Examples like this suggest that transitivity configurations may be chosen to impact how the text is interpreted and received.

Characterization-in-transitivity in translation

It has been shown that transitivity configurations are closely related to characterization. Bernard MacLaverty, for example, portrays the eponymous character of his novel Cal as a passive and ineffectual observer by constructing Cal as the senser of perception-related mental processes and the actor of 'reflexive' material processes (Simpson and Montgomery 1995).4 James Joyce also [End Page 286] constructs interesting characterization in his short story Two Gallants by encoding the two central characters, Lenehan and Corley, into contrasting transitivity configurations (Kennedy 1982; Nørgaard 2003). According to Kennedy (ibid., 91–96), Lenehan is a passive observer featured in the roles of the senser and goal-less actor, while Corley is an active initiator inscribed into material clauses. These and similar examples (e.g., Kies 1992; Hubbard 1999; Nørgaard 2003; Ji and Shen 2004; Shen 2007) suggest that "the transitivity profile [a regular pattern of transitivity choices] embodied by a text is a generally useful indicator of character in prose fiction," and that "the relationship between transitivity and characterisation is a close one" (Simpson 2004: 119–120).

The relationship between transitivity and characterization is also important in translation. Translation involves choosing from possible transitivity configurations what may be considered 'appropriate' or 'sufficient.' Importantly, transitivity configurations chosen by the translator may contribute to shaping particular forms of characterization, which may possibly differ from those in the ST. Indeed, a growing body of research has demonstrated that characterization in transitivity is different between the ST and TT.

Among the pioneers in research on characterization-in-transitivity in translation is Vasconcellos (1998). She compares the "I"-narrator's conceptualizations of reality in James Joyce's Araby and its two Portuguese translations, using Halliday's (1971) analytic perspective of William Golding's The Inheritors. More specifically, she analyzes lexicogrammatical features linked to the narrator's progression to self-search in three sections of the story, namely "Textualization A," "Textualization B," and "Textualization C." For example, she argues that in Textualization A, material clauses with inanimate agents (e.g., the cold air stung us, her name sprung to my lips) are used to highlight the protagonist's limited view of the world, while in Textualization B, 'volitional' mental/material clauses are exploited to represent the protagonist's growing sense of reality (e.g., I recognized a silence, I held a florin tightly in my hand).

Vasconcellos (1998) is particularly relevant to the present study, because it deals with more than one "retextualization" (i.e., translation) and explores the relationship between alterations in transitivity and the protagonist's mental transformations. Nonetheless, it seems to have two limitations. First, Vasconcellos bases her analysis on a small number of short individual clauses. Her findings would thus have been richer if longer stretches of the text had been analyzed in context. Second, it is uncertain why she introduces the notion of translation quality assessment (TQA). Although her paper is entitled "[…] A systemic functional view of translation quality assessment," it provides only brief hints at the relationship between choices of transitivity and TQA. [End Page 287]

While Vasconcellos (1998) examines shifts in characterization in inter-lingual translation, Martínez (2002) addresses similar issues in 'intra-lingual' translation by comparing Thomas Pynchon's Under the Rose with his rewriting of the story as Chapter 3 in the novel V ("V3"). Martínez argues that in Under the Rose the characterization of Porpentine, the 'active' and 'aware' protagonist, contradicts the plot because the character is caught in a trap and shot dead at the end. In V3, however, Porpentine is endowed with the "features of passiveness and low agentivity not present in the first version of the story" through detransitivizing devices such as ergativity, metonymy, and generic plurals (p. 644).5 As a result, transitivity configurations in V3 "diminish the import of human agency in the reader's mind, and […] contribute to develop[ing] one of the themes of the novel V" (p. 653).

Martínez (2002) reveals not only the relationship between the translator's lexicogrammatical choices and character/thematic development, but also provides details on what kinds of linguistic devices can be exploited for re-characterization. In particular, she discusses the qualitative and quantitative aspects of transitivity shifts in a mutually complementary way (see also Hasan 1989; Hubbard 1999; Lee 2008). However, it is not surprising that Under the Rose and V3 differ in characterization, because in the original version fictional events are presented from the perspective of Porpentine, while in the revised version seven secondary characters replace Porpentine as 'focalizers' (i.e., the primary consciousness of a story). It can be argued that in the case of rewriting, transitivity-for-characterization is chosen by the same, omniscient writer; in translation proper, however, it is chosen (not always consciously) by a different 'writer' in a different context. In inter-lingual translation, alterations in characterization are likely to occur in a less systematic and less perceptible way.

Finally, Lee (2008) explores the possibly differing effects of two translation strategies on character construction by analyzing how the President of South Korea is described in a Newsweek feature article and its two Korean translations. One of the translations is a Newsweek Korea article published by a private publishing company where translated texts undergo an editing process (J. Kang 2004: 67), whereas the other is a 'literal' translation released by a government institution. Lee thus hypothesizes that in the former case there are more shifts in transitivity related to the president, leading to greater differences in character construction between the ST and TT. To test this hypothesis, Lee divides transformations from and into material processes into six categories, using [End Page 288] Calzada-Pérez's (2007: 152–175) typology of transitivity shifts. His findings show that in either translation there is no general trend in the use of material processes, and that the two translations are similar in the degree of the character's involvement in active roles.

However, Lee (2008) focuses on the number of shifts concerning the material process without giving enough consideration to other transitivity components. Although the material process is probably the most important component of transitivity, the circumstance and non-material processes are also influential in shaping characterization. It is argued that characterization will be revealed more convincingly if all transitivity components are analyzed in a comprehensive way.

Yi Sang's Nalgae and character construction

Yi Sang (1910–1937) was a writer ahead of his time. Under the influence of Dada and surrealism, he experimented with highly idiosyncratic narrative, while addressing complex themes such as self-consciousness and soul-searching. It is commonly acknowledged that Yi Sang has been more influential than any other writer and that his work made a great contribution to developing modern Korean literature.

Nalgae (1936) is one of Yi Sang's best-known stories. It has appeared in high school textbooks for decades as a fine example of early-modern literature in Korea. Its storyline and thematic development have inspired post-war psychological novels and various art forms including films, paintings, and even cartoons. It has also been the subject of a great deal of scholarly discussion. As of May 2017, there were more than 3,730 journal articles and books about Nalgae, according to RISS (http://www.riss/kr), a government-financed information service for researchers.

Nalgae revolves around a jobless young man, who depends for survival on a prostitute whom he calls "my wife." He ekes out a miserable life without any human contact, and spends much of his time contemplating or sleeping in his sunless room, the inner part of his wife's room. Normally, he is characterized as an inactive, lazy and powerless person who rarely engages in meaningful physical activities. However, during one of his wife's outings, he sneaks into her outer room and enjoys himself playing with fire. This change in character construction reoccurs in a similar form when he goes out to the streets. In the whole story of Nalgae, the protagonist goes outside five times. His last outing, however, is most relevant to the theme of the story because it leads him to desire 'wings' as a means of escape from self-imprisonment. The story of Nalgae can be divided into a few substories, as explained in the following: [End Page 289]

Literature shows that Nalgae can be divided into three sections. The first section is the moment when the protagonist resorts to despondent isolation in his room, content with his state of inaction […]. The second is the moment when he goes out to the streets […]. The third is the moment when he reflects on himself on the top floor of a department store and desires to fly on the streets (Y. Kang 2006: 98).

In Nalgae the plot develops in two major steps. Firstly, the protagonist goes into his wife's room. Secondly, he ventures out to the outer world (Kwŏn Yŏng-min 2009: 312).

Considering previous research on the plot of Nalgae, the researcher divided the story into four sections, in each of which there occurs a major shift in the character construction of the "I"-protagonist. The key passages of the four substories, hereinafter called "Extract 1," "Extract 2," "Extract 3," and "Extract 4," can be summarized as follows:

- Extract 1: The protagonist stays inert in his sunless room.

- Extract 2: The protagonist enjoys himself in his wife's room.

- Extract 3: The protagonist first goes out to the streets.

- Extract 4: The protagonist feels a desire to 'fly' on the streets.

These extracts are individually analyzed in transitivity terms in the analysis section of the article.


Translations chosen for analysis

The three translations chosen for this study come from one booklet and two anthologies readily available in They are (1) Wings (1990) by Peter H. Lee, (2) The Wings (1983/2001) by Ahn Jung-hyo, and (3) Wings (2005) by Walter K. Lew and Youngju Ryu.6 [End Page 290]

For comparison of the ST and TTs, the four key passages of the story, namely Extracts 1, 2, 3, and 4, were chosen and analyzed in transitivity terms. In each Extract, however, only two TTs were chosen for comparison for two reasons. First, due to space limitations, all the translations cannot be analyzed in each Extract. Second, in some Extracts, two of the three translations resemble each other in terms of transitivity. Two TTs chosen for analysis in each Extract are a pair of translations that best reveal differences in transitivity. A set of texts analyzed in each Extract are as follows:

- Extract 1: ST, TT1, and TT3

- Extract 2: ST, TT2, and TT3

- Extract 3: ST, TT1, and TT2

- Extract 4: ST, TT2, and TT3

Focus of analysis

For the purpose of this study, the clauses into which the "I"-protagonist of Nalgae is inscribed were analyzed in detail, with an emphasis on the process and its associated participant roles. The circumstance was also considered when it pertains to the protagonist's agency. The following shows how the text can be analyzed in transitivity terms.

[Material Process:] Pocketing [Goal:] them, [Actor:] I [Material Process:] walked and walked, [(Relational Process) Attribute:] heedless [Circumstance:] of where [Actor:] I [Material Process:] was going, in order only to [Mental Process:] forget [Phenomenon:] the reason [Circumstance:] for [(Actor/Material Process) Nominalization:] my walking.

Each Extract is discussed in a way that provides different flavors of how characterization can be analyzed in transitivity terms. Extract 1 reveals unusual transitivity configurations concerning the protagonist's perceptual power and the relationship between the protagonist and his environment. Extract 2 concerns the relation of character construction to alterations in the proportions of three major process types. Extract 3 focuses on effects of 'pseudo-material' processes on character construction. Finally, Extract 4 shows material processes representing the protagonist's desire and the role of 'present-tense' processes in determining how the character is received. [End Page 291]


Extract 1: Protagonist in his dark room


In Extract 1, the central figure of Nalgae lurks in his sunless room, feeling estranged from the outer world. His sense of inertia and alienation manifests itself in the opening part of the story.


(1) 나는 어데까지든지 내방이—집이 아니다. 집은 없다.—마음에 들었다. (2) 방안의 기온은 내 체온을 위하여 쾌적하였고 방안의 침침한 정도가 또한 내 안력을 위하여 쾌적하였다. (3) 나는 내 방 이상의 서늘한 방도 또 따뜻한 방도 희망하지는 않았다. (4) 이 이상으로 밝거나 이 이상으로 아늑한 방을 원하지 않았다. (5) 내 방은 나 하나를 위하여 요만한 정도를 꾸준히 지키는 것 같아 늘 내 방에 감사하였고 나는 또 이런 방을 위하여 이 세상에 태어난 것만 같아서 즐거웠다.

In Extract 1, the protagonist bluntly says that he does not have a 'home,' suggesting that he does not have a stable ego identity (Hwang 1993: 23). This is clearly shown in (1), a grammatically twisted sentence, which literally translates as 'Every part of my room—[This is] not a home. There is no home.—pleases me.' In (1), two "parenthetical sentences," i.e., 집이 아니다 (chibi anida, 'this is not a home') and 집은 없다 (chibŭn ŏpta, 'there is no home'), grammaticalize the protagonist's mental digression (Leech 2008: 141).7 It is important to note that in several of Yi Sang's stories, the main characters are antic, self-deprecating persons who do not have a home.

In Extract 1, the protagonist is encoded as a passive, objectified being who perceives his environment in an idiosyncratic way. Specifically, in the two clauses constituting (2), the protagonist's narration sounds unnatural, because the two circumstances 내 체온을 위하여 (nae ch'eon ŭl wihayŏ, 'for my body's temperature') and 내 안력을 위하여 (nae allyŏk ŭl wihayŏ, 'for my eyesight') are combined with the attribute^relational process 쾌적하였다 (k'waejŏkhayŏtta, 'was pleasant').8 In [End Page 292] this transitivity configuration, the sensing is realized by the abstract, inanimate circumstantial elements 체온 (ch'eon, 'body's temperature') and 안력 (allyŏk, 'eyesight'), not by the holonymic agency "I" (see Simpson 2004: 76–77).9 The senser role is inscribed in the circumstance, "the most passive in the cline of dynamism" (Hasan 1988, Hubbard 1999), while the thing sensed 내 방 (nae pang, 'my room') is foregrounded into the theme position.10

It appears that in (5) the grammatical subject and inanimate entity 내 방 (nae pang, 'my room') takes a life of its own and acts upon the protagonist. The room is encoded as a human-like participant to whom the protagonist has always been grateful (늘 내방에 감사하였고, nŭl nae pang e kamsahayŏkko, 'I always thanked the room'). The material process ascribed to the room (지키다, chik'ida, 'maintain') also makes the room look like an animate actor with conscious volition ('my room constantly maintains its conditions for me alone'). Furthermore, the protagonist says, '이런 방을 위하여 이 세상에 태어난' (irŏn pang ŭl wihayŏ i sesang e t'aeŏnan, 'I was born into this world for the sake of the room'), indicating that he is a powerless character inherently subordinate to the room.

ST vs. TTs

In Extract 1, the TT1 and TT3 were chosen for comparison.

TT1 (p. 51)

(1) I like my room very much. (2) I find its temperature comfortable, and the semidarkness suits my eyes. (3) I wouldn't want a room any cooler or warmer, or any brighter or darker. (4) I keep on thanking the room for maintaining such an even temperature and brightness. (5) I'm pleased to think that I was born to enjoy this kind of room.

TT3 (p. 68)

(6) Every part of my room—Not a house. There is no house—pleases me.

(7) The room's temperature suits my body's and its degree of dimness is agreeable to my eyesight. (8) I never hoped for a room cooler or warmer than mine. (9) Nor have I wished for one that is brighter or cozier. (10) I have always been grateful to my room, which seems to constantly maintain these conditions for me alone, and I am delighted that it may be for the [End Page 293] sake of this room alone that I was born into the world.

In the TT1, the protagonist is described as an ordinary person who can properly perceive his environment. In (1), he takes the theme position as the full-fledged "I"-senser, while making no mention of a 'home.' In (2), he is featured in the role of the holonymic "I"-senser, with no circumstantial element used to encode his sensing. These transitivity patterns indicate that the protagonist narrates like 'us.'

In addition, in the TT1 there are additions that could impact the character construction of the protagonist. In (5), for example, the cognitive-mental process "think," a verb not present in the ST, makes his perceptual power become stronger. This feeling is further strengthened as the protagonist is inscribed in the emotive-mental process "enjoy," another new verb that highlights one's intention and emotion. After all, in the TT1 the protagonist appears as a person who can consciously enjoy his room.

In contrast, in the TT3 the protagonist is characterized in the same way as in the ST. In (6), there are two parenthetical sentences denoting his mental digression, which are incorrect in English grammar. In (7) the protagonist as the senser is inscribed in a circumstance ("to my eyesight"), while the room as the sensed takes the theme position as a major participant. In (10), there are no mental processes with conscious volition that could have strengthened the protagonist's perceptual power. Although the TT3 appears unnatural to native English speakers, it is very similar to the ST in terms of characterization.

Extract 2: Protagonist in his wife's room


The story of Extract 2 is set in the outer, bright part of the divided room. It has been argued that this outer room contributes to thematic development in two respects (e.g., Son 1981; Hwang 1993; Ch'oe 1983: 262–263). First, it is only through this room that the protagonist can escape from his enclosure to the real world. Second, the outer room is the space in which the protagonist begins to fight back against his tyrannical wife and to search for the meaning of sunlight (his ego).

In Extract 2, the protagonist is characterized in a different way. He feels a degree of liberation, toying with his wife's possessions and playing with fire. The persistent sense of powerlessness can thus hardly be felt, as seen in the following:


아내가 외출만 하면 나는 얼른 아랫방으로 와서 그 동쪽으로 난 들창을 열어 놓고 열어 놓으면 들여비치는 볕살이 아내의 화장대를 비쳐 가지각색 병들이 [End Page 294] 아롱이 지면서 찬란하게 빛나고 이렇게 빛나는 것을 보는 것은 다시 없는 내 오락이다. 나는 쪼꼬만 「돋보기」를 꺼내 가지고 아내만이 사용하는 지리가미를 끄실려가면서 불장난을 하고 논다. 평행광선을 굴절시켜서 한 초점에 모아 가지고 고 초점이 따끈따끈해 지다가 마지막에는 종이를 끄실르기 시작하고 가느다란 연기를 내이면서 드디어 구녕을 뚫어놓는데까지에 이르는 고 얼마 안 되는 동안의 초조한 맛이 죽고 싶을 만치 내게는 재미있었다.

In the above passage, processes ascribed to the protagonist are mostly actionrelated material processes (e.g., 열어 yŏrŏ 'open,' 꺼내 kkŏnae 'take out,' 논다 nonda 'play'). This is a tremendous shift in transitivity, because in several preceding paragraphs the protagonist is encoded largely by mental and relational processes. In other words, in Extract 2, the dominant transitivity pattern of 'thinking-andbeing' gives way to the 'action' pattern, as visualized in Figure 1.

Figure 1. Process patterns in the unfolding story (from Extract 1 to Extract 2)
Click for larger view
View full resolution
Figure 1.

Process patterns in the unfolding story (from Extract 1 to Extract 2)

In Figure 1, the number of material processes skyrockets in [6], the paragraph corresponding to Extract 2, while the mental and relational processes show only marginal changes between Paragraph [1] (i.e., the paragraph corresponding to Extract 1) and Paragraph [6]. This means that in his wife's room the protagonist becomes a person with actional properties. In particular, in Extract 2 the protagonist is often featured in the role of the 'actor^inanimate goal,' the second highest level in Hasan's (1989) cline of dynamism. [End Page 295]

ST vs. TTs

In Extract 2, the TT2 and TT3 were chosen for comparison.

TT2 (p. 12)

(1) It has become a major recreation of mine that I promptly go to the front room in the morning when my wife goes out and watch various bottles on her make-up chest brilliantly glimmer with the sunbeam trickling in through the eastern window I opened. (2) I have fun scorching tissue paper, which only my wife is entitled to use, with a small magnifying glass. (3) The straight rays of the sun are refracted to gather at a focus, and that focus gets warmer and hotter until it starts scorching the soft paper, a feeble thread of smoke wriggling up, leaving a tiny dark hole; (4) I am so thrilled by this short spell of anxious waiting that it almost kills me.

TT3 (p. 69)

(5) The moment my wife goes out, I steal into the outer room and lift up the east-facing window. (6) When I do, the sun streams in and shines upon her makeup stand, speckling the array of colored vials and making them glow; (7) to view this is my favorite amusement. (8) I take out a magnifying glass and play with fire, scorching the sheets of chirigami that only my wife uses. (9) I refract the sun's parallel rays and gather them at a focal point until it heats up, singeing the paper and giving off wisps of smoke. (10) It's only a matter of seconds before a hole appears in the burning paper, but the suspense is so pleasurable that it almost kills me.

The TT2 and TT3 differ in the proportions of three major process types (see Figure 2). In the TT2 the material, mental and relational processes are almost evenly divided, indicating that the protagonist's actions may be expressed less explicitly. This is because some of the material processes in the ST were translated into mental or relational processes or circumstances. For example, in (2) the material process 놀다 (nolta, 'play') was replaced by the relational process "have [fun]," with the likely result that the protagonist's mental situation, not his action, is emphasized. In addition, the goal^material process 돋보기를 꺼내 가지고 (toppogirŭl kkŏnae kajigo, 'take out a magnifying glass [and play with fire]') was changed into a static circumstance ("with a small magnifying glass"). It should be noted that the protagonist's act of taking out a magnifying glass symbolizes the beginning of his search for sunlight or self-search (see Ch'oe 1983: 264; Kim Chong-kŏn 2004: 230). [End Page 296]

Figure 2. Proportions of three major process types in Extract 2
Click for larger view
View full resolution
Figure 2.

Proportions of three major process types in Extract 2

Furthermore, the first clause in (3) is a receptive one, with the protagonist taking part as an implicit/suppressed actor (i.e., depersonalization).11 Without any "I"-referent in (3), the action-related material process 모아 가지고 (moa kajigo, 'I gather [… the sun's rays]') was changed into the event-related material process "[The straight rays of the sun …] are refracted to gather." It follows that the TT2 construes 'happening' rather than 'doing,' and that the protagonist is characterized as less active.

In contrast, the TT3 is almost the same as the ST in that it encodes the protagonist as an external agent engendering all the processes. In (8), for example, a material process is used to feature the protagonist's act of taking out a magnifying glass. In (9), the clause remains operative, with the explicit "I"-actor functioning as the grammatical subject. The two material processes "refract" and "gather" are of the action type, indicating that 'playing with fire' is described as the protagonist's action, not as a self-caused event.

Extract 3: Protagonist on his first outing


Extract 3 features the protagonist sneaking out of his wife's room. It has been argued that by going out to the street the protagonist can achieve the 'enlightenment' of the real world and become an independent being (Son 1981: 235; Hwang 1993: 21; Pak 1997: 166). [End Page 297]


(1) 나는 아내의 밤 외출 틈을 타서 밖으로 나왔다. (2) 나는 거리에서 잊어버리지 않고 가지고 나온 은화를 지폐로 바꾼다. (3) 五원이나 된다. (4) 그것을 주머니에 넣고 나는 목적을 잃어 버리기 위하여 얼마든지 거리를 쏘다녔다. (5) 오랜간만에 보는 거리는 거의 경이에 가까울 만치 내 신경을 흥분시키지 않고는 마지 않았다. (6) 나는 금시에 피곤하여 버렸다. (7) 그러나 나는 참았다. (8) 그리고 밤이 이슥하도록 까닭을 잊어 버린 채 이 거리 저 거리로 지향없이 헤매었다.

With the protagonist's outing comes a significant change in his characterization (Y. Kang 2006: 117). As he sneaks outside, he turns into an active, free person whose action appears intentional. This can be noticed in the following transitivity patterns.

(1) [Senser/Actor:] Nanŭn [Phenomenon:] anaeŭi pam oech'ul t'ŭm ŭl [Mental:] t'asŏ [Circumstance:] pak ŭro [Material:] nawatta.

(2) [Actor:] Na nŭn [Circumstance:] kŏri esŏ [Material:] idŏbŏriji anko kajigo naon [Goal:] ŭnhwa rŭl [Circumstance:] chip'yero [Material:] pakkunda […]

(4) [Goal:] Kŭgŏt ŭl [Circumstance:] chumŏni e [Material:] nŏk'o [Senser:] na nŭn [Phenomenon:] mokchŏg ŭl [Mental:] irhŏ pŏrigi wihayŏ […] [Scope:] kŏri rŭl [Material:] ssodanyŏtta.

(5) [Circumstance:] Oraen'ganmane [Mental:] ponŭn […] [Goal:] nae sin'gyŏng ŭl [(Metaphorical wording) Material:] hŭngbunsik'iji […]

(6) [Carrier:] Na nŭn [Circumstance:] kŭmsi e [Attribute/Relational:] p'igonhayŏ pŏryŏtta.

(7) [Senser:] na nŭn [Mental:] ch'amatta.

(8) [Phenomenon:] kkadalg ŭl [Mental:] idŏ pŏrin ch'ae […] [Material:] hemaeŏtta.

The above analysis shows that in Extract 3 the protagonist is an acting character. He is grammaticalized with seven material processes, five mental processes, and one relational process. In the four paragraphs leading to Extract 3, the protagonist generally assumes the senser roles, with nine out of sixteen verbal groups belonging in the category of mental processes. The remaining seven roles are three carriers, one circumstance, one receiver, one goal, and one actor. These roles (except for one actor role) are classified as "very passive" in Hasan's (1989) cline of dynamism.

ST vs. TTs

In Extract 3, the TT1 and TT2 were chosen for comparison. [End Page 298]

TT1 (p. 56)

(1) Taking advantage of my wife's absence, I went out to the streets. (2) I brought along all my coins and changed them into paper notes—five won. (3) Pocketing them, I walked and walked, heedless of where I was going, in order only to forget the reason for my walking. (4) The streets that came into view after so long stimulated my nerves. (5) I began to feel tired in no time, but I persisted. (6) I kept walking till late at night, heedless of where I was going and oblivious to my original purpose.

TT2 (p. 21)

I sneaked out of my room while my wife was out. Outside on the street, I did not forget to change my coins for bills. They amounted to five won. With that money in my pocket, I loitered around as I pleased so that I might get lost. The wonderful outside world, which I had not seen for a long time, did not fail to arouse my nerves. I was tired out immediately, but I endured. Until dark, I sauntered aimlessly here and there not knowing why I was doing this.

The TT1 and TT2 differ in the material process. In the TT2 there are five material processes, five mental processes, and two relational processes, while in the TT1 there are thirteen material processes, two mental processes, and four relational processes. In wording terms, the TT1 has more than twice as many material processes as the ST.

It should be noted that in the TT1 the protagonist is often inscribed in 'pseudo-material' processes. In (3) and (6), for example, three material processes embedded in the circumstances are not real descriptors of the acting protagonist (e.g., "I kept walking heedless of where I was going"). Also the verbal group "walked and walked" in (3) may be counted as a single process because it only highlights the continuity of the protagonist's walking. In (4), the material processes "came" and "stimulated" are of the 'blending' type, indicating that the form brings in material coloring but its dominant meaning relates to a mental process.12 After all, from a semantic perspective, the TT1 contains seven material processes. [End Page 299]

Figure 3. Potential influence of pseudo-material processes on character construction
Click for larger view
View full resolution
Figure 3.

Potential influence of pseudo-material processes on character construction

Although half of the material processes in the TT1 are pseudo-material, it does not necessarily mean that the pseudo-material processes will have no influence on character construction. It is rather suggested that the way a message is constructed may constitute meaning (see Thompson 2004: Ch. 1), and that as visualized in Figure 3, a great number of pseudo-material processes may reinforce the image of the main character as an active person in the mind of readers.

Extract 4: Protagonist on his last outing


Extract 4 is the climatic ending of Nalgae.


(1) 이때 뚜우하고 정오 사이렌이 울었다. (2) 사람들은 모두 네 활개를 펴고 닭처럼 푸드덕거리는 것 같고 온갖 유리와 강철과 대리석과 지폐와 잉크가 부글부글 끊고 수선을 떨고 하는 것 같은 찰나, (3) 그야말로 현란을 극한 정오다.

(4) 나는 불현듯이 겨드랑이 가렵다. (5) 아하, 그것은 내 인공의 날개가 돋았던 자국이다. (6) 오늘은 없는 이 날개, (7) 머릿속에서는 희망과 야심의 말소된 페이지가 딕셔내리 넘어가듯 번뜩였다.

(8) 나는 걷던 걸음을 멈추고 그리고 어디 한 번 이렇게 외쳐보고 싶었다.

(9) 날개야 다시 돋아라.

날자. 날자. 날자. 한번만 더 날자꾸나. 한번만 더 날아 보잤꾸나.

In Extract 4, 정오 (chŏngo, 'noon') is the moment when the protagonist reestablishes himself (Son 1981; Kim Chong-ku 1981; Hwang 1993; Pak 1997). The meaning of 정오 is highlighted in (3), where the nominal group is modified by the rank-shifted clause 현란을 극한 (hyŏllan ŭl kŭkhan, 'the splendor that has [End Page 300] peaked').13 The clause (3) ('the noon reached the peak of its splendor') thus indicates that the protagonist's self-search ends at noon and in the brightest light (Son 1981).

In (2) there is a concentration of material processes stimulating the protagonist's desire to 'fly' (Ch'oe 1983; Hwang 1993; Kim Chu-ri 2006). The four material processes, namely 펴고 (p'yŏgo, 'extend'), 떨고 (ttŏlgo, 'rumble'), 푸드덕거리는 (p'udŭdŏkkŏrinŭn, 'flap'), and 끓고 (kkŭlk'o, 'boil up'), create a strong sense of upward movement. In particular, the processes 펴고 and 푸드덕거리는 are closely associated with the image of wings (Hwang 1993: 31).

In (4), (5), and (6), the protagonist's desire to fly is expressed more vividly through present-tense processes. The narration switches in (4) from the past to the present, and returns to the past in (7). This shift in tense provides a strong contrast with the general temporal orientation of the story, and can be further explained as follows:

As the plot unfolds, […] the protagonist's mental state progresses from 'ignorance' to 'recognition.' […] In Nalgae, there is a moment when the reader senses 'tension.' […] It is the moment when the protagonist recognizes that he does not have wings. The moment is called the topic time. […] The topic time that the protagonist emphasizes is "here-andnow."

The shift to the topic time makes the readers step outside the past frame and move toward the protagonist. The present-tense processes, along with the temporal circumstance 오늘은 (onŭl ŭn, 'today'), offer a close-up and vivid account of the protagonist who now realizes 오늘은 없는 이 날개 (onŭl ŭn ŏpnŭn i nalgae, 'these wings are missing today').

ST vs. TTs

In Extract 4, the TT2 and TT3 were chosen for comparison.

TT2 (pp. 39–40)

(1) A siren wailed, announcing noon. (2) It was a glorious noon, (3) people vigorously whirling around amid the commotion of glass, steel, marble, money and ink.

(4) My armpits suddenly itched. (5) Ah, it was where my imitation wings [End Page 301] had split out. (6) The wings that I had no longer; (7) the deleted phantasms of hope and ambition flashed in my mind like the flipping pages of a pocket dictionary.

(8) I stopped my pace and wanted to shout.

(9) Wings, spread out again!

Fly. Fly. Fly. Let me fly once more. Let me fly just once more.

TT3 (pp. 83–84)

(10) At that moment, the noon siren wailed: Tuu-u--! (11) People extended their four limbs and flapped around like chickens, while all sorts of glass, steel, marble, money, and ink seemed to rumble and boil up (12) – right then, the noon reached the zenith of its dazzling splendor.

(13) All of a sudden, I feel an itch under my arms. (14) Aha! The itching is a trace of where my artificial wings had once sprouted. (15) Wings that are missing today: (16) pages from which my hopes and ambition were erased flashed in my mind like a flipped-through dictionary.

(17) I want to halt my steps and shout out for once:

"Wings! Grow again!

"Let's fly! Let's fly! Let's fly! Let's fly just one more time.

"Let's fly once again!"

There are three major mismatches between the ST and TT2. First, lexicogrammatical features encoding the temporal indicator 'noon' are greatly different. In (2), the word "noon" appears as a head in the non-dynamic, circumstantial attribute. The epithet "glorious" alone cannot convey the meaning and dynamism of the material clause 현란을 극한 (hyŏllan ŭl kŭkhan, 'the splendor that has peaked').

Second, in (3) the protagonist's mental construction of wings is suppressed. The two material processes that remind the protagonist of wings 펴고 푸드덕거리는 (p'yŏgo p'udŭdŏkkŏrinŭn, 'extend and flap') are reduced to "whirling," a process that has little to do with wings. The material processes 떨고 (ttŏlgo, 'rumble') and 끓고 (kkŭlk'o, 'boil up') are also reduced to the circumstance "amid the commotion."

Third, all the processes (except for the processes in the protagonist's interior monologue [9]) are tagged firmly on to the past, with no circumstance encoding the topic time 오늘은 (onŭl ŭn, 'today'). This means that the readers are not mentally present at the occurrence of the processes and they cannot feel a sense of immediacy (for details on English tense in SFG, see Bache 2008). That is, the readers may feel detached from the most critical moment of the story, or the moment when the protagonist comes to realize that 'wings are missing today.' [End Page 302]

In contrast to the TT2, the TT3 is very similar to the ST. In (11) the thematically important material processes "extend," "flap," "rumble," and "boil up" appear with their associated participants, emphasizing the protagonist's mental construction of 'wings.' In (12) the meaning of 'noon' is grammatically encoded in the material process^scope configuration ("reached the zenith of its dazzling splendor").14 The dash^circumstance "– right then" brings more attention to the timing of the protagonist's self-realization. The clauses (13), (14), and (15) are basically in the present tense, with the circumstance "today" present in (15). In (14) the use of past-perfect tense ("had once sprouted") in the middle of the topic time maximizes the temporal contrast.


This study revealed that transitivity configurations chosen by the translator may possibly alter characterization and the translation reader's construction of the character. It compared a Korean literary work with its English translations (TT1, TT2, and TT3), in terms of how the "I"-protagonist is characterized in four thematically important sections.

The analysis undertaken in this study may lead some readers to make evaluative judgments about the three translations each. However, the findings should not be interpreted as an announcement that one translation is good and the other translations are bad. A translated literary work cannot be assessed solely by the way its characters are described, even though characterization may influence the reader's interpretation of the work. Indeed, when asked to read Extract 1 in both TT1 and TT3, many native speakers of English said that the TT3 does not sound like a natural English narrative due to its lexico-syntactic patterns.15

Nonetheless, it is argued that characterization-in-transitivity, often linked to thematic development, is one of the most important considerations in literary translation. As shown in Extract 1, the TT1, a very natural translation, encodes the protagonist as an ordinary person whose perceptual ability is similar to ours. This characterization may not be desirable, however, since the original plot revolves around a powerless person's mental transformation into an active, independent being. [End Page 303]

This study has three limitations. First, it analyzed only four sections of the story. It can thus be suggested that to make the author's argument more viable, it is necessary to look into a longer stretch of Nalgae and to examine similar works of fiction in the context of translation. Second, this study focused almost entirely on lexicogrammatical features that are explicit in the texts. In other words, it hardly considered what may be implicit in the texts. Future researchers therefore need to consider the network of conceptual relations that underlie the surface text. Third, this study did not show whether shifts in characterization actually influence the reader's mental construction of the character. It would thus be meaningful to examine whether the "real reader" feels toward the protagonist in much the same way as the author argues the "implied reader" does (Chatman 1978). Future research (e.g., a large-scale reader-response study on character construction) should provide valid empirical evidence of character re-construction in translation. [End Page 304]

Lee Sang Bin

Lee Sang-Bin ( is an associate professor in the Department of English for International Conferences and Communication, Hankuk University of Foreign Studies, Seoul, Korea.

Submitted: December 1, 2017
Sent for revision: April 9, 2018
Accepted: April 30, 2018


Primary Sources

Yi Sang. (1936) 1991. Nalgae (Wings). In Yi Sang munhak chŏnjip 2 (A collection of Yi Sang's stories, Vol. 2), edited by Kim Yun-sik, 318–347. Seoul: Munhak sasang.
———. 1990. Wings. Translated by Peter H. Lee. In Modern Korean Literature: An Anthology, edited by Peter H. Lee, 49–65. Honolulu: University of Hawai'i Press.
———. 2001. The Wings. Translated by Ahn Jung-hyo. In The Wings, edited by Ahn Jung-hyo and James Lee, 7–40. Seoul: Jimoondang.
———. 2005. Wings. Translated by Walter Lew and Youngju Ryu. In Modern Korean Fiction: An Anthology, edited by Bruce Fulton and Youngmin Kwon, 66–84. New York: Columbia University Press.

Secondary Sources

Abrams, Meyer, and Geoffrey Harpham. 2009. A Glossary of Literary Terms. 9th ed. Boston: Wadsworth Cengage Learning.
Bache, Carl. 2008. English Tense and Aspect in Halliday's Systemic Functional Grammar: A Critical Appraisal and an Alternative. London: Equinox.
Calzada-Pérez, María. 2007. Transitivity in Translating: The Interdependence of Texture and Context. Bern: Peter Lang.
Catford, John. 1965. A Linguistic Theory of Translation. London: Oxford University Press.
Chatman, Seymour. 1978. Story and Discourse: Narrative Structure in Fiction and Film. London: Cornell University Press.
Chesterman, Andrew. 1997. Memes of Translation: The Spread of Ideas in Translation Theory. Amsterdam: John Benjamins.
Ch'oe Suk-in. 1983. "Yi Sang ŭi Nalgae e nat'anan hoehwasŏng" (Pictorial representations in Yi Sang's Nalgae). Ihwa ŏmun nonjip 6: 255–271.
Culler, Jonathan. 1975. Structuralist Poetics. London: Routledge.
Halliday, Michael. 1971. "Linguistic Function and Literary Style: An Inquiry into the Language of William Golding's The Inheritors." In Literary Style: A Symposium, edited by Seymour Chatman, 330–368. London: Oxford University Press.
Halliday, Michael. 1985. An Introduction to Functional Grammar. London: Edward Arnold.
Halliday, Michael. 1994. An Introduction to Functional Grammar. 2nd ed. London: Edward Arnold.
Halliday, Michael, and Christian Matthiessen. 2004. An Introduction to Functional Grammar. 3rd ed. London: Hodder Arnold.
Harvey, William. 1968. Character and the Novel. Ithaca, NY: Cornell University Press.
Hasan, Ruqaiya. 1988. "The Analysis of One Poem: Theoretical Issues in Practice." In Functions of Style, edited by David Birch and Michael O'Toole, 45–73. London: Pinter.
Hasan, Ruqaiya. 1989. Linguistics, Language and Verbal Art. Oxford: Oxford University Press.
Hubbard, Hilton. 1999. "Love, War and Lexicogrammar: Transitivity and Characterisation in The Moor's Last Sigh." JLS/TLW 15, no. 3/4: 355–376.
Hwang To-gyŏng. 1993. "Yi Sang ŭi sosŏl konggan yŏn'gu" (Spatiality in Yi Sang's fiction). PhD diss., Ehwa Womans University.
James, Henry. 1972. "Characters." In Theory of Fiction: Henry James, edited by James Miller, 195–226. London: University of Nebraska Press.
Ji, Yinglin, and Dan Shen. 2004. "Transitivity and Mental Transformation: Sheila Watson's The Double Hook." Language and Literature 13, no. 4: 335–348.
Kang, Ji-hae. 2004. "Evaluation of South Korea in News Media Translation: A Critical Perspective." Discourse and Cognition 11, no. 1: 1–30.
Kang Yong-un. 2006. Yi Sang sosŏl ŭi sŏsa wa ŭimi saengsŏng ŭi nolli (A study on how narration and meaning is generated in Yi Sang's fiction). Paju: Taehaksa.
Kennedy, Chris. 1982. "Systemic Grammar and Its Use in Literary Analysis." In Language and Literature: An Introductory Reader in Stylistics, edited by Ronald Carter, 82–99. London: George Allen & Unwin.
Kies, Daniel. 1992. "The Uses of Passivity: Suppressing Agency in Nineteen Eighty-four." In Advances in Systemic Linguistics: Recent Theory and Practice, edited by Martin Davies and Louise Ravelli, 229–250. London: Pinter.
Kim Chong-gu. 1981. "Yi Sang Nalgae ŭi sigan konggan kujo: kŭ sangjingjŏk ŭimi punsŏk ŭl chungsim ŭro" (Temporal and spatial structures of Yi Sang's Nalgae: An investigation into their symbolic meanings). Sŏgang ŏmun 1: 65–86.
Kim Chong-gŏn. 2004. "Yi Sang sosŏl ŭi konggan sŏlchŏng kwa chakka ŭisik: Nalgae rŭl chungsim ŭro" (Yi Sang's spatiality and consciousness: A case study of Nalgae). Ŏmunhak 84: 221–249.
Kim Chu-ri. 2006. "Kŭndaejŏk sinch'e hunyuk ŭi kwanjŏm esŏ pon Nalgae ŭi ŭimi" (Wings seen from the perspective of modern physical discipline). In Yi Sang munhak yŏn'gu ŭi saeroun chip'yŏng (A new horizon of research into Yi Sang's fiction), edited by Sin Pŏm-sun, 211–234. Seoul: Yŏngnak.
Kwŏn Yŏng-min. 2009. Yi Sang t'eksŭt'ŭ ŭi yŏn'gu (Research on Yi Sang's texts). Seoul: Ppul.
Kwŏn Yun-ok. 1986. "Yi Sang sosŏl ŭi sigan punsŏk: Nalgae rŭl chungsim ŭro" (Temporal factors in Yi Sang's fiction: A case study of Nalgae). The Research on Korean Language and Literature 21: 365–398.
Lee, Sang-Bin. 2008. "The Impacts of Translation on the Perceived Degree of an Actor's Kinesis." Korean Journal of Applied Linguistics 24, no. 2: 139–159.
Leech, Geoffrey. 2008. Language in Literature: Style and Foregrounding. Harlow: Longman.
Lothe, Jakob. 2000. Narrative in Fiction and Film: An Introduction. New York: Oxford University Press.
Martínez, Maria Angeles. 2002. "From 'Under the Rose' to V: A Linguistic Approach to Human Agency in Pynchon's Fiction." Poetics Today 23, no. 4: 633–656.
Matthiessen, Christian, Kazuhiro Teruya, and Marvin Lam. 2010. Key Terms in Systemic Functional Linguistics. London: Continuum.
Nørgaard, Nina. 2003. Systemic Functional Linguistics and Literary Analysis: A Hallidayan Approach to Joyce, A Joycean Approach to Halliday. Odense: University Press of Southern Denmark.
Pak Sin-hŏn. 1997. "Yi Sang ŭi Nalgae: Chuje kuhyŏn ŭl wihan chakp'um chŏn'gae pangsik yŏn'gu" (Developing a storyline for fictional themes: A case study of Yi Sang's Nalgae). Ŏmunhak 61: 155–174.
Shen, Dan. 2007. "Internal Contrast and Double Decoding: Transitivity in Hughes's On the Road." Journal of Literary Semantics 36, no. 1: 53–70.
Simpson, Paul. 2004. Stylistics: A Resource Book for Students. London: Routledge.
Simpson, Paul, and Martin Montgomery. 1995. "Language, Literature and Film: The Stylistics of Bernard MacLaverty's Cal." In Twentieth-century Fiction: From Text to Context, edited by Peter Verdonk and Jean Jacques Weber, 138–159. London: Routledge.
Son Pyŏng-hŭi. 1981. "Panbokchŏk haengwi wa sigan ŭi hoebok: Nalgae ŭi kyŏng'u" (Repetitive action and temporal changes: A case study of Nalgae). Munhak kwa ŏnŏ 2: 231–246.
Thompson, Geoff. 2004. Introducing Functional Grammar. 2nd ed. London: Arnold.
Vasconcellos, Maria Lúcia. 1998. "'Araby' and Meaning Production in the Source and Translated Texts: A Systemic Functional View of Translation Quality Assessment." Cadernos de Tradução 3: 215–254.
Vinay, Jean-Paul, and Jean Darbelnet. 1995. Comparative Stylistics of French and English: A Methodology for Translation. Translated by Juan Sager and Marie-Jo Hamel. Amsterdam: John Benjamins.


. This study was supported by the Hankuk University of Foreign Studies Research Fund.

1. The term character often means the reader's conception of character. According to Abrams and Harpham (2009: 42), characters are "the persons represented in a dramatic or narrative work, who are interpreted by the reader as possessing particular moral, intellectual, and emotional qualities by inferences from […] the dialogue and […] the action."

3. In SFG, the symbol ^ indicates the combination of transitivity elements in order.

4. The word reflexive is used in the sense that Cal's action affects himself or part of his body (e.g., he closed his eyes.)

5. The term transitivity here is used rather confusingly because it is taken to mean both Halliday's system of experiential meaning and the traditional (non-SFG) view of a property of verbs.

6. Ahn's translation is a reprint of The Wings published in Korea Journal (January 1983). It is part of The Portable Library of Korean Literature (Vol. 1) dedicated to Yi Sang. Lee's translation (an abridged version) is part of an anthology (titled Modern Korean Literature: An Anthology) consisting of four literary genres (poetry, fiction, drama, and essays). The translation by Walter K. Lew and Youngju Ryu is part of an anthology of Korean fiction (titled Modern Korean Fiction: An Anthology).

7. Parenthetical sentences are inserted into a sentence of which they are only loosely a part, breaking up the linearity of the text as the speaker/narrator spontaneously digresses.

8. The rules of English and Korean grammar are quite different. In Korean, for example, the order of the goal^material process is grammatically correct (e.g., 그는 [he] 나무 아래에서 [under the tree] 유리를 [the glass] 깼다 [smashed].)

9. In holonymic agency, the participant role is occupied by a complete being (e.g., "I", "he").

10. Hasan (1988) postulates a "cline of dynamism" in which thirteen participant roles can be ordered according to the degree of dynamism. The role configuration of the "actor + animate goal" is the most dynamic, while the "circumstance" is the least dynamic. Hubbard (1999) uses Hasan's cline to divide fourteen transitivity roles into "active" and "passive."

11. In traditional grammar, receptive is passive voice, while operative is active voice.

12. These processes represent abstract doings and happenings (see Halliday and Matthiessen 2004: 196)

13. Rank-shift is an "expansion of the systemic potential of semantics and grammar by the shift of a unit down the rank-scale to serve as a unit of a lower rank. For example, "a prepositional phrase may be down-ranked to serve as a qualifier in a nominal group" (Matthiesssen et al. 2010: 170).

14. The scope specifies the meaning of the material process itself or the domain over which the process occurs. It is different from the goal in that it is not affected by the performance of the process, as in I climbed the mountain.

15. The author conducted a small-scale questionnaire survey about this issue (N = 55). However, the findings of the survey are not in the scope of this study.