Scholarly article on topic 'The “Dual Voice” of Free Indirect Discourse: A Reading Experiment with Croatian Speakers Reading a Literary Text in English and Croatian'

The “Dual Voice” of Free Indirect Discourse: A Reading Experiment with Croatian Speakers Reading a Literary Text in English and Croatian Academic research paper on "Languages and literature"

CC BY-NC-ND
0
0
Share paper
OECD Field of science
Keywords
{"free indirect speech" / "dual voice" / "possibility of a third voice" / "reading experiment" / "Pride and Prejudice"}

Abstract of research paper on Languages and literature, author of scientific article — Krunoslav Mikulan, Vladimir Legac

Abstract The possibility of “dual voice” in free indirect discourse has attracted considerable attention in the fields of stylistics, cognitive science and literary criticism. An area suitable for empirical research studies in this regard is point of view and how it is identified during the reading process by non-native speakers as well as the role of translators in influencing the reader to accept a particular point of view. This paper presents results of a research study designed to examine how readers respond to free indirect discourse. The research considered four research questions: RQ1: Do readers hear a ‘dual voice’ when reading passages of free indirect discourse? RQ2: What kind of ‘contexts’ influence the identification of point of view? RQ3: Are there any differences in the results obtained between non-native speakers reading the original literary text and native speakers of the language reading the translation of the same original? RQ4: Has the translator influenced the readers of the translation in any way (thus serving as a “third voice”)? The text for the experiment was from Jane Austen's Pride and Prejudice. The experiment was conducted with two groups of participants: 38 non-native university students studying English as one of their majors to become elementary school teachers of English and 40 university students and native speakers of Croatian studying to become elementary school teachers teaching all core subjects. The results show that a quarter of the study participants could hear the “dual voice” and that the translator's influence remained for the most part irrelevant. The difference between the two study groups was not substantial and the reasons for choosing their answers mentioned by the participants did not significantly differ.

Academic research paper on topic "The “Dual Voice” of Free Indirect Discourse: A Reading Experiment with Croatian Speakers Reading a Literary Text in English and Croatian"

CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Procedia - Social and Behavioral Sciences 158 (2014) 214 - 221

14th International Language, Literature and Stylistics Symposium

The "dual voice" of free indirect discourse: A reading experiment with Croatian speakers reading a literary text in English and

Croatian

Krunoslav Mikulan, Vladimir Legac*

University of Zagreb, Faculty of Teacher Education, Savska 77, Zagreb 10000, Croatia

Abstract

The possibility of "dual voice" in free indirect discourse has attracted considerable attention in the fields of stylistics, cognitive

science and literary criticism. An area suitable for empirical research studies in this regard is point of view and how it is

identified during the reading process by non-native speakers as well as the role of translators in influencing the reader to accept

a particular point of view. This paper presents results of a research study designed to examine how readers respond to free

indirect discourse. The research considered four research questions:

RQ1: Do readers hear a 'dual voice' when reading passages of free indirect discourse?

RQ2: What kind of 'contexts' influence the identification of point of view?

RQ3: Are there any differences in the results obtained between non-native speakers reading the original literary text and native speakers of the language reading the translation of the same original?

RQ4: Has the translator influenced the readers of the translation in any way (thus serving as a "third voice")? The text for the experiment was from Jane Austen's Pride and Prejudice. The experiment was conducted with two groups of participants: 38 non-native university students studying English as one of their majors to become elementary school teachers of English and 40 university students and native speakers of Croatian studying to become elementary school teachers teaching all core subjects. The results show that a quarter of the study participants could hear the "dual voice" and that the translator's influence remained for the most part irrelevant. The difference between the two study groups was not substantial and the reasons for choosing their answers mentioned by the participants did not significantly differ. © 2014PublishedbyElsevierLtd.Thisis anopen access article under the CC BY-NC-ND license (http://creativecommons.Org/licenses/by-nc-nd/3.0/).

Peer-review under responsibility of Dokuz Eylul University, Faculty of Education.

Keywords: free indirect speech; dual voice; possibility of a third voice; reading experiment; Pride and Prejudice

* Corresponding author. Tel.: +385 40 313 190 E-mail address: vladimir.legac@ufzg.hr

1877-0428 © 2014 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).

Peer-review under responsibility of Dokuz Eylul University, Faculty of Education. doi:10.1016/j.sbspro.2014.12.077

1. Introduction

The emergence of cognitive poetics has focused attention on how stylistic features are processed by readers. An area suitable for empirical research studies in the field is point of view and how it is identified during the reading process by non-native speakers as well as how the role of the translator influences the reader to accept a particular point of view.

In this respect our attention will be focused on free indirect discourse which has lately been in the core of scholarly interest. It was first mentioned by German and French linguists at the beginning of the twentieth century. The term has since appeared in a variety of forms: free indirect speech, free indirect style, narrated monologue, erlebte Rede, style indirect libre (cf. Pascal, 1977: 8-32, Vetters, 1994: 179) and others. The term free indirect discourse, which if such a need should arise can be divided into free indirect speech and free indirect thought, has in the meantime been widely accepted and has received significant attention followed by research in the fields of narratology, cognitive linguistics and literary theory (Pascal, 1977; Fludernik, 1993; Yamaguchi, 1993; Noh, 2000; Pit, 2003; Sanford and Emmott 2012; Leech, 2013).

According to Leskiv (2009: 53) when free indirect speech (FIS) functions as stream-of-consciousness the speaker or "thinker" is identifiable and when FIS functions as the vehicle of irony or empathy we can differentiate between at least two sources - the character whose words are reported, and the author who intervenes in the report and is responsible for the irony or sympathy. According to Bray (2007: 39) in free indirect discourse it is possible to hear the so called "dual voice" in which both the narrator's and an experiencing character's points of view appear to be present. In Fludernik's (1993: 3) words "the overlap between characters' and narrator's language in free indirect discourse, where standard narratological accounts posit the site of a dual voice effect ... acquires supreme theoretical relevance as a transgression of an otherwise neat separation line between the words of the text in the narrating process and the plot level of the fictional world, which is linguistic in a merely derivative, incidental and partial manner." When Pascal (1977: 17) discusses the "style indirect libre" he states that the duality of voice is an underlying feature of the style and that it "may be heard as a tone of irony, or sympathy, of negation or approval, underlying the statement of the character". Volosinov (1973: 144, cited in Bray 2007: 40) insists that the form "does not contain an either/or dilemma; its specificum is precisely a matter of both author and character speaking at the same time" and Ginsburg (1982: 135, 140, cited in Bray, 2007: 40) describes free indirect discourse as a "completely bivocal utterance" which "contains two sets of contradictory signs, one pointing to the speech of the characters and the other toward the narration. These two "voices", as Volosinov has emphasized, coexist in a structure of undecidability (where one cannot opt for either one or the other being the "meaning" of the utterance)".

This study presents the results obtained from a reading experiment based on a similar reading experiment conducted by Bray (2007) and the results published in Language and Literature, Vol. 16(1). The same excerpt from the novel Pride and Prejudice by Jane Austin was used, but in this experiment the Croatian translation was used as well. The experiment was conducted with two groups of participants: 38 non-native university students studying English as one of their majors to become elementary school teachers of English and 40 university students and native speakers of Croatian studying to become elementary school teachers teaching all the core subjects.

2. Problem of the "third voice"

In the passage from Pride and Prejudice in Croatian that served as the foundation of the study, a third party appears along with the speaker and the author - that of translator. The translator decided to clarify the words for the reader and directly intervened into the text adding the name of the "speaker" (Elizabeth) in the fourth sentence. The following excerpt in English was used (Austen, 1996 [1813]: 227):

(1) She had never perceived, while the regiment was in Hertfordshire, that Lydia had any partiality for him; but she was convinced that Lydia had wanted only encouragement to attach herself to anybody. (2) Sometimes one officer, sometimes another, had been her favourite, as their attentions raised them in her opinion. (3) Her affections had been continually fluctuating, but never without an object. (4) The mischief of neglect and mistaken indulgence towards such a girl! - oh how acutely did she now feel it.

The passage in the Croatian variant was the text taken from the translation of the novel into Croatian (Austen, 1997 [1813]: 214; translator: Tomislav Odlesic):

(1) Nikad nije opazila, dok je pukovnija bila u Hertfordshireu, da Lydia osjeca bilo kakvu naklonost prema njemu; no vjerovala je da je Lydiji potreban samo malen poticaj da se veze uz bilo koga. (2) Katkad je ovaj, katkad onaj casnik bio njezin miljenik, a to je ovisilo o tom tko joj se vise udvarao. (3) Osjecaji joj bijahu nepostojani, no uvijekje imala nekog ljubimca. (4) Elizabeth je sada osjecala svu tezinu pogreske sto se nije vise utjecalo na takvu djevojku i sto joj se previse pustalo na volju.

It is necessary to point out the following differences between the two texts. In sentence (1), the translator uses the active form of the verb "believe" in the same tense ("she believed') instead of the passive of the verb "convince" ("she was convinced"). In sentence (2), the translator uses the demonstratives "this" and "that' instead of "one" and "another". In sentence (3), instead of the verb ""fluctuate" in the past perfect continuous and the adverb "continually" in the combination ("had been continually fluctuating') the translator uses the adjective that could be back-translated into English with "wavering", "changeable" "unstable"' or "volatile"', i.e. "the affections were wavering...."" The biggest differences between the original text in English and the translated Croatian text appear in sentence (4) which we shall repeat to facilitate easier comparison, and we shall add the back-translation of the sentence from Croatian into English:

Original sentence:

The mischief of neglect and mistaken indulgence towards such a girl! - oh how acutely did she now feel it.

Croatian translation:

Elizabeth je sada osjecala svu tezinu pogreske sto se nije vise utjecalo na takvu djevojku i sto joj se previse pustalo na volju.

Back-translation into English:

Elizabeth now felt the full weight of the mistake of not having exercised more influence on such a girl and of allowing her too much freedom.

According to Bartolussi (2003: 223) the "central theoretical problem in the scholarship on speech and thought has been identifying the voice associated with the text of the narrative." In our case the translator manipulated the point of view in order to enable the reader to better understand the passage, but in the process he has obliterated several semantic/linguistic markers, namely the exclamation mark, the dash, the interjection "oh", and the inversion of indirect question/exclamation. It is as if the translator followed Kelepsky's opinion that "such exclamations ... are unguarded outbursts of the writer that betray the true nature of SIL (style indirect libre) as a masked expression of authorial opinion, smuggled in in this form in order to ingratiate itself with the innocent reader" (Pascal 1977: 13) and they therefore need to be eliminated from the text or modified in order to remove any uncertainty. The translator becomes, in a sense, the "third voice" of the discourse, albeit in the background, invisible to the reader who does not have the original text in front of him/her.

The translator modified the shift of perspective from Elizabeth as the narrator towards the author as narrator and thus potentially altered the reader's perception of the analysed passage, which is a fact that this study attempted to prove or disprove.

3. Description of study

The study aimed to answer the following four research questions:

RQ1: Do readers hear a 'dual voice' when reading passages of free indirect discourse?

RQ2: What kind of 'contexts' influence the identification of point of view?

RQ3: Are there any differences in the results obtained between native speakers of Croatian reading the original litarary text in English and native speakers of Croatian reading the Croatian translation of the same original?

RQ4: Has the translator influenced the readers of the translation in any way?

3.2 Methodology

3.2.1 Instrument

Only one instrument was used for this experiment. It was a self-constructed questionnaire. It had two variants: an English and a Croatian one. Both of them consisted of instructions, a passage from Jane Austen's Pride and Prejudice, and two questions. The instructions in both variants were given in Croatian. It was stated that the experiment that they were going to take part in had been designed to test how point of view affected the reading process and that they would be asked to read the above mentioned passage only once and to answer the two questions. The first was a multiple-choice question with four possible answers (a), (b), (c) and (d). The respondents had to answer whose voice was represented in sentence 4, i.e. whether it was Elizabeth's alone, the narrator's alone, both Elizabeth's and the narrator's or whether it was impossible for them to decide whose voice was represented. The second question was an open end one. It asked the respondents to elaborate on their reasons for their answer to the first question. They were invited to identify the features of language which had led them to their selection. They had to write down on the page the crucial features on which they had based their choice. Additionally, they were told that they did not need to write in full sentences or be exhaustive and that they should not spend more than five minutes on the whole experiment.

The authors believed that the word "Elisabeth' in the Croatian text that was not mentioned in the English original could lead the readers to circle answers (a) or (c) and thus result in a higher percentage of these answers among the "Croatian group" than among the "English" group" (see 3.2.2. Participants). Both questions in the English variant of the questionnaire were in English and both questions in the Croatian variant of the questionnaire were in Croatian. Students answered the questions in the respective languages: the English group in English and the Croatian one in Croatian.

3.2.2 Participants

Research was conducted among 78 students of the Faculty of Teacher Education in Zagreb, which is a constituent of the University of Zagreb, Croatia. At the time of data collection they were all about twenty years old and in their second year of the 5-year-integrated master's degree study programme studying to become elementary school teachers. Over ninety percent of them were female. This is not unusual as for several decades now the overwhelming majority of the primary school teachers in Croatia teaching in grades 1-4 have been women. They were divided in two groups: an English group and a Croatian group. The English group consisted of 38 students studying English as one of their majors to become qualified teachers of English in grades 1-8. The Croatian group consisted of 40 students studying to become qualified teachers of all core subjects taught in grades 1-4. Those core subjects are Croatian (mother tongue), Mathematics, Natural Science, Physical Education, Music and Arts. (N.B. Croatian elementary school lasts eight years and is divided into so-called "lower" and "upper classes". In the lower classes all the subjects are taught by the same class teacher. The only exceptions are foreign language classes and religion which are taught by specialists qualified for those two subjects. In the upper classes each subject is taught by a specialist for the relevant subject). The Faculty of Teacher Education of the University of Zagreb has three branches (Central Unit in Zagreb and two additional extensions - Departments in Cakovec and Petrinja). The participants in the English group were studying at the central location in Zagreb and those in the Croatian group in Cakovec. Students in both groups were from all parts of the Republic of Croatia. A vast majority of the English group students had been learning English for at least nine years before they enrolled at the Faculty of Teacher Education and had had approximately 945 school periods of English. At the time of data collection they had been

studying English for two years and had had several English language, grammar and literature courses and were considered to be at the advanced level, so the authors thought that from the language point of view they were a suitable sample group for their experiment. Additionally, the authors would also like to emphasize that it was highly unlikely that any of the students in the sample had had any specialist knowledge of free indirect discourse as this item it is not mentioned in any of the syllabi in the first two years of their study programme.

3.2.3 Procedure

At both locations the data were collected during the last week of the summer term in mid-June 2014. The students were openly informed about the purpose of the study, and they were all willing to take part in the experiment. The researchers also orally mentioned all the information that the respondents would later read in the instructions from the questionnaire. They were then asked to read the instructions and to give sincere answers, as the researchers guaranteed them anonymity. The whole procedure (explanation of the aim of the experiment, expression of students' willingness to take part in it, explanation of the instruction, distribution of the questionnaires and their collection from the respondents did not last more than quarter of an hour at both locations of the experiment.

4 Analysis of results and discussion

All the data were analyzed by means of SPSS 17.0 for Windows.

4.1 Research question 1

The first aim of the researchers in this study was to find out whether or not readers hear a 'dual voice' when reading passages of free indirect discourse. As it can be clearly seen from Table 1 'dual voice' is heard by more than one quarter of the participants in both of the studied groups. The alternative (c) - which was Elisabeth's and the narrator's voice - i.e. the alternative that includes two constituents: voice of one of the characters as well as the narrator's voice was chosen by 10 students of 38 in total or 26.32% in the English group and 13 students of 40 in total or 32.5% in the Croatian group. This percentage is much less lower than in case of the alternative (b). Exactly half of the students in the English group (19 students of 38 in the sample) and exactly half of the students in the Croatian group (20 of 40 in the sample) opted for the narrator's voice. The alternative (a), which stood for Elizabeth's voice alone, ranked third in the English group (7 students or 18.42%) and fourth in the Croatian group (3 students or 7.5%). One tenth of the students in the Croatian group (4 students) thought it was impossible to decide whose voice was represented in sentence 4. The percentage of those who agreed with this opinion was twice as low in the English group (5.26%, i.e. only two students). The authors compared the results of this survey with the results obtained with the native speakers of English in a survey by a distinguished researcher of free indirect discourse (Bray, 2007, p. 43). The participants in Bray's study were 32 students from the Scottish University of Sterling. They read the same passage and were asked the same questions. A considerably higher percentage of students in his study opted for 'dual voice' (43.75%). It still meant that fewer than half of the respondents claimed to hear a 'dual voice' in sentence 4. The same percentage of participants in Bray's survey heard Elizabeth's 'voice' alone (43.75%), four subjects (12.5%) said they heard only the narrator's voice, while there was not a single student who thought it was impossible to decide whose voice was represented. The authors' conclusion from this comparison was that native speakers of Croatian are less prone to construct a 'dual voice' both when reading a text in their mother tongue and in English as a foreign language than native speakers of English reading a text in their mother tongue.

Table 1: Students' Answers to Question No. 1

Voice represented in sentence 4 English group Croatian group

Number of circled answers Percentage of circled answers Number of circled answers Percentage of circled answers

(a) Elisabeth's alone 7 18.42 3 7.5

(b) The narrator's 19 50 20 50

(c) Both Elisabeth's 10 26.32 13 32.5

and the narrator's

(d) It is impossible to decide 2 5.26 4 10

Total 38 100 40 100

4.2 Research question 2

The second research question, RQ2, was concerned with identifying the kind of 'contexts' that influence the identification of point of view. In order to get some insight into the process of students' identification of those contexts, they were asked to write down the features of language that had led them to select the alternatives in the first question.

Of those who claim to have heard a 'dual voice' in the English group more than a half mention the use of pronouns and punctuation (dash and exclamation mark). Some of them concluded that the use of dash must indicate the change of subject. Others emphasized that the use of the determiner "such" and the use of the interjection "oh" in the same sentence pointed to the change of subject from the first to the third person. On the other hand, some participants thought that the first part of the fourth sentence represented Elizabeth's voice, while the second part after the dash was the narrator's voice. One participant wrote that the use of the exclamation mark in the last sentence "had told her that the narrator might have realized something". Some participants in the study did not mention any linguistic reasons for their view. One participant thought that the whole paragraph was written in Elizabeth's voice but she saw psychological reasons for the inclusion of the narrator. She thought that the narrator was placing herself into the Elizabeth's character in the second part of the sentence. Another participant thought that the first part of the sentence was an expression of Elizabeth's social status by the narrator whereas the second part was an expression of her direct awareness of it. This interpretation might indicate that this student had some problems in understanding certain vocabulary items in this sentence or that he or she was reading too much into it because of his or her vivid imagination.

Of those who thought they heard a 'dual voice' in the Croatian group more than a half mentioned again the use of pronouns and subjects. They pointed to the fact that the change of the subject between Elisabeth (the third person singular) and the subject of the passive form (the third person plural) must indicate the change of the point of view from the character to the narrator. Another person pointed to the change in number in the Croatian translation between the two parts of the sentence, i.e. singular in the first part and plural in the second (N.B. this cannot be seen from the back-translation into English). One participant noticed the same change of the subject through the use of the word "Elisabeth' and "such a girl" in the two different parts of the sentence. In the Croatian group there was a student who concluded that the change of the point of view was the result of the author's characterization of Elisabeth's feelings and thoughts. Only one participant mentioned the use of punctuation but gave no proof for that from the text.

Reasons mentioned for choosing alternative (b) in the English group were primarily the use of pronouns and punctuation. Participants emphasized that they lacked first person references and quotation marks. They claimed that if it had been Elisabeth's voice, the words would have been put into the inverted commas or there would have been some other 'sign' indicating the change of the point of view. Several participants underlined the pronoun "she" to emphasize their point. One participant mentioned "the previous context" as a reason for choosing alternative (b). Despite having circled alternative (b), one student still had some doubts. She mentioned the use of pronouns and "maybe punctuation" (put in brackets) as her reason for choosing this alternative, but she expressed

her uncertainty by writing that it "is a bit confusing because when you continue to read you have a feeling that it's Elizabeth speaking".

The third person personal pronouns as subjects were the major reason for the participants in the Croatian group to circle the narrator alone as the voice represented in sentence 4 of the given paragraph. One participant said it was "pure structure of the sentence that suggested narration" and no quotation.

The explanation for majority of the participants in the English group for choosing Elisabeth's voice as the only voice represented in sentence 4 was in the previous text. They thought that the last sentence was some kind of a natural continuation of the earlier text, because it was Elizabeth who knew her sister well. One of them even said that it was not natural to expect "from the narrator to be talking about someone else's feelings". Two of the participants pointed to the interjection "oh " as an indication of Elizabeth's voice speaking in that sentence.

The third person singular of the verb phrase and the third person of the subject were given as reasons for choosing Elisabeth's voice as the only one represented in sentence 4 by participants in the Croatian group.

The only two participants in the English group who chose alternative (d) as their answer said that it could easily be the narrator as well as Elizabeth's voice alone speaking in sentence 4, but there was no clear evidence that only one voice was represented.

The respondents from the Croatian group claimed that there was no language feature present in the text that would make it possible for them to conclude whose voice was represented in the last sentence of the given paragraph.

As we can see from this analysis the same reasons were sometimes given by some participants in the study as a proof for the representation of one voice and by others as a proof for the representation of a different voice or even for a 'dual voice'. This all suggests that the authors of this paper have to conclude that some caution is needed for the proposition that the dual voice hypothesis has a cognitive appropriateness to the reading experience.

4.3. Research question 3

The aim of the third research question was to find if there were any differences in the results obtained between native speakers of Croatian reading the original literary text in English and native speakers of Croatian reading the translation of the same original. A quick glance at the percentages in Table 1 reveals that the difference in the percentages of answers between the two studied groups was not very big. In case of alternative (b) the values of the studied groups were even equally high, i.e. 50%. The difference between the English and the Croatian group in case of the 'dual voice' alternative was 6.18%, which means that a quite similar percentage of the participants were 'in favour' of the dual voice'. The gaps between the two groups were approximately equal for the alternatives (a) and (d). Twice as many participants opted for Elizabeth's voice alone in the English group than in the Croatian group. The order was vice versa in case of the alternative (d). There were almost twice as many students in the Croatian group who thought it was impossible for them to decide about the representation of voice in the last paragraph than in the English group.

It needs to be mentioned that the difference between the two tested groups could not be tested by means of a t-test because of the nature of the questionnaire as the alternatives in question 1 are not numerical. However, the distribution of the answers did not allow chi-square test either (see Table 1) as there were less than 5 cases in cell (d) in both studied groups.

4.4. Research question 4

The aim of the fourth research question was to detect any possible influence of the translator on the readers of the translation. In 3.2.1 the authors pointed out the differences between the original English text and the Croatian translation. They thought that the word "Elizabeth in the Croatian text that was not mentioned in the English original could lead the readers to circle answers (a) or (c) and thus result in a higher percentage of these answers among the "Croatian group" than among the "English group". As we have seen in 4.3 the differences between the two studied groups were not big. In case of the alternative (a) the order was quite reverse from the one that had been expected. There were more students among the English group than among their Croatian counterparts who

opted for Elizabeth as the only represented voice. The authors think that the reason for this discrepancy might lie in the fact that literary texts in foreign language are usually read at a slower pace than those that are read in one's own mother tongue.

5 Limitaions of the study, suggestions for further studies and conclusion

Two main limitations need to be mentioned. The first concerns the characteristics of the sample. The sample was not balanced in terms of gender. There were more female than male participants, but the researchers could not do anything about this imbalance caused by social preferences of the two genders in choosing their future jobs. It needs to be pointed out that the size of the sample was not too small. There were only 38 students in the "English group". This figure might seem small to many readers. However, there are only about 200 students studying to become qualified primary school teachers of English in the whole Republic of Croatia. If one bears this in mind, then the size of the sample is quite appropriate. The second limitation lies in the fact that only one literary text was included in the experiment. Therefore, our results need to be interpreted with caution. As the authors of this paper are committed to do further research in the nature of the 'dual voice', they will do a similar experiment with this and another passage of a literary work of art with implications of the presence of the 'dual voice' with participants studying to become primary school teachers of English and core subjects in one of the neighbouring countries. At this stage they can conclude that more than a quarter of the students studying to become primary school teachers of English and primary school teachers of the core subjects hear a 'dual voice'. The participants see the reasons for the representation of the voice in the text primarily in punctuation, the use of pronouns, subjects and interjections, but some caution is needed for the proposition that the dual voice hypothesis has only a cognitive appropriateness to the reading experience, as was the case with Fludernik (1993). Some reasons mentioned by the participants in this study cannot be classified as only cognitive. Therefore, the authors believe that these reasons need to be more thoroughly investigated in their future research with new participants.

References

Austen, J. (1996 [1813]). Pride and Prejudice. Harmondsworth: Penguin.

Austen, J. (1997 [1813]). Ponos i predrasude. trans. T. Odlesic. Zagreb: Katarina Zrinski.

Bartolussi, M. & Dixon, P. (2003). P sychonarratology: Foundations for the Empirical Study of Literary Response. Cambridge: Cambridge University Press.

Bray, J. (2007). The 'dual voice' of free indirect discourse: a reading experiment. Language and Literature, 16, 37-51.

Fludernik, M. (1993). The Fictions of Language and the Languages of Fiction: The LinguisticRepresentation of Speech and Consciousness.

London and New York: Routledge. Ginsburg, M. P. (1982). 'Free Indirect Discourse: A Reconsideration', Language and Style 15(2):133-49. Leech, G. (2013). Meaning and the English Verb. Oxon and New York: Routledge.

Leskiv, A. (2009). The Literary Phenomenon of Free Indirect Speech. Studia Anglica Resoviensia, 6, 51-58. Noh, E.-J. (2000). Metarepresentation. Amsterdam and Philadelphia: John Benjamins. Pascal, R. (1977). The Dual Voice. Manchester: Manchester University Press.

Pit, M. (2003). How to Express Yourself with a Causal Connective: Subjectivity and Causal Connectives in Dutch, German and French.

Amsterdam and New York: Rodopi. Sanford, A.J. and Emmot, C. (2012). Mind, Brain and Narrative. Cambridge and New York: Cambridge University Press. Vetters, C. (1994). Free Indirect Speech in French. In: C. Vet & C. Vetters (Eds.), Tense and Aspect in Discourse (pp. 179-226). Berlin and New York: Mouton de Gruyter.

Volosinov, V. N. (1973). Marxism and the Philosophy of Language, trans. L. Matejka and I. R. Titunik. New York and London: Seminar Press. Yamaguchi, H. (1993). Unrepeatable Sentences: Contextual Influence on Speech and Thought Presentation. In: H. Parret (Ed.), Pretending to Communicate (pp. 239-252). Berlin and New York: de Gruyter.