1. Introduction

Recently, there has been an attempt to expand the theoretical and descriptive focus of systemic functional linguistics (SFL) to move beyond clauses toward text and context (Martin, 2014). For instance, traditionally, interpersonal meaning within SFL was explored through mood and modality which are grammatical systems at clause level within the dialogic texts (White, 2015). The appraisal theory, thus, was developed within SFL to expand and explore the interpersonal meaning beyond clause level within monologic texts (Martin & White, 2005). In other words, appraisal system is a complementary perspective on interpersonal meaning which is concerned with evaluation, i.e., how interlocutors express their feelings, attitudes and judgments in monologic texts through lexis (Halliday & Matthiessen, 2014).    

In other words, the appraisal theory is a robust theory based on the solid theoretical ground of Systemic Functional Linguistics (Ngo & Unsworth, 2015) which enjoys a comprehensive and systematic analytical tool (Hyland, 2005). This has turned the appraisal theory into a viable research tool. The fine-grained description in the appraisal theory has recognized it as “the only systematic, detailed and elaborate framework of evaluative language” (Bednarek, 2006, p. 32), and “[t]he most fully developed model of evaluation” (Thompson, 2014, p. 48). Since its development, appraisal theory has been applied to research in various contexts and a wide range of topics such spoken genre (Eggins & Slade, 1997; Ngo & Unsworth, 2015; Page, 2003 among others); EFL academic writing (Jalilifar & Moazzen, 2014; Jalilifar et al., 2012; Liu, 2013 among others), politics (Jalilifar & Savaedi, 2012) and media (White, 1998). However, the majority of appraisal research has focused primarily on the English language and despite Martin and White’s (2005) invitation to work on the appraisal across languages and cultures, little research has attempted to explore it in Persian language (e.g., Jalilifar & Savaedi, 2012; Jalilifar & Hemmati, 2013). In fact, these studies have focused on the appraisal in Politics and academic writing and no research has investigated the appraisal in recounts whether in the English or Persian. Therefore, the current investigation intends to fill this gap by exploring how Persian male and female webloggers express their feelings and judgments in their personal recounts. Hence, the following research question is addressed: 

RQ: Is there a difference between the way male and female use attitude system in the appraisal theory in their recounts? 

1.1. Recounts

Recount is a storytelling genres in which a set of related events experienced by a narrator are retold in chronological order. Anderson and Anderson (2003) argued that in a recount text, past experiences are listed and described in chronological order. This type of text is used to retell past experiences based on the order to which they appear and it is used to inform or entertain the audience with the past experiences. However, Hyland (2003) believed that in recounts, past experiences are retold in its original sequences.

Sitorus and Sipayung (2018) contended that recounts can be categorized into various forms; the first form is a personal recount in which the author's personal experience is retold. The second form is the factual recount in which reports of actual events such as science experiment reports or police reports are provided. Finally, the authors introduced imaginative recount which serves to present an imaginative story. The language features of a recount text indicate that the language is written in the simple past tense, and connectives that link events in time, such as next, later, when, then, after, before, first are frequently used (Mediska & Adnan, 2019). 

Boardman (2008) stated that there are several language features usually found in the recount text. First, nouns and pronouns are used to identify people, animals or things involved. Second, recount texts use the past action verbs to refer the events. Third, past tense is used to locate events in relation to speaker’s or researcher’s time. Fourth, recount texts use conjunctions and time connectives to sequence an event. Finally, recount texts also use adverbs and adverbial phrases to indicate place and time and use adjectives to describe nouns. Based on the above explanations, it can be concluded that recount is a text that tells events of past experiences in chronological order to entertain the audience.

Several studies investigated recounts, its functions and applications. Yusnita et al. (2012) tried to improve students’ writing of recounts using Picture Series; Saragih et al. (2014) considered recount texts as a tool to improve Grade III Students of Elementary School writing abilities. Muflikhati (2013) used recount texts in the form of dialogue journals of the Tenth Grade Students to enhance the students’ writing abilities. Sari (2017) examined students’ problems in writing recount texts and Mediska and Adnan (2019) taught recount texts in order to help students improve their speaking for mastering the spoken form of recount texts. Purwanto (2019) used semantic mapping to improve writing recount texts by second-grade students. The majority these previous studies used recounts as a tool to improve other skills. Only, Nurohmah (2013) investigated students’ recount from the perspective of SFL. The data were six pieces of recount texts written by the eighth semester students of English Department who were treated as respondents. As illustrated by the short report above, previous studies primarily utilized recounts as a research tool and have not focused on them as an original source of analyses from the perspective of different theories. Furthermore, the extant studies focusing on recounts per se have rarely differentiated narratives and recounts and have used the two terms or genres interchangeably. Accordingly, in the present study, the focus was on recounts as operationalized by Eggins and Slade (1997). They defined recount genre as a sub-genre of storytelling that consists of a temporal sequence of events in which the narrator spreads his or her interpretation and evaluation through the text making it worth telling. This characteristic of recounts and their frequency of appearance in the personal weblogs written by Iranians led us to investigate their attitudinal resources by drawing on the attitude subsystem of the appraisal framework for the first time.

1.2. Appraisal Framework

The appraisal can generally be defined as the evaluative use of language. Through its main system (i.e., attitude), the appraisal explores the way language is used to evaluate people or things through expressing emotive reactions (affect), passing judgment on people’s behavior according to social rules (judgment) and expressing opinions about material objects and utterances (appreciation) to form alliance and solidarity with those who share these views and distance themselves from those who do not share the same views (Martin & White, 2005). Figure 1 (adopted from Martin & White, 2005) shows an overview of the appraisal systems and subsystems.

Figure 1. An Overview of the Appraisal Resources

Since the focus of this study was on the attitude subsystem, the following parts provide some definitions and examples of this subsystem based on Martin and White (2005) and White’s (2001b) models. As it is shown in the above figure, the appraisal has three systems: 1) Attitude, 2) Engagement, and 3) Graduation. The attitude system itself has three subsystems: 

A) Affect: It is concerned with both positive and negative emotional responses. Types of affect and their English examples are shown in Table 1. Affect is typically realized through

      a) Mental processes of reaction e.g. like, hate, love, adore

                  This pleases me, I hate chocolate, …

      b) Attributive relations of Affect e.g. 

                  I’m sad/happy/disappointed

                  She is frightened.

      c) Through ideational metaphors which are usually realized as nouns e.g. 

                  His fear was obvious to me.    

Examples of affect are found in table 1 in the appendix.

B) Judgment: It is concerned with evaluating human behavior positively or negatively by reference to a set of institutionalized norms. These social norms are either rules or regulations (i.e., social sanction as illustrated in Table 2 below) or less defined social expectations and value systems namely social esteem. They are presented in table 3 below. Thus, under judgment, one may assess behavior as moral/immoral, legal/illegal, as socially acceptable or unacceptable, as laudable or deplorable, and as normal or abnormal (White, 2001b). Examples of judgment are provided in table 2 in the According to White (2001b), judgments can be realized linguistically as adverbials, attributes, nominals and verbs: 

1) Adverbials – justly, fairly, virtuously, honestly, pluckily, indefatigably, cleverly, stupidly, eccentrically

2) Attributes and epithets – a corrupt politician, that was dishonest, don’t be cruel, she’s very brave, he’s indefatigable, a skillful performer, truly eccentric behavior

3) Nominals – a brutal tyrant, a cheat and a liar, a hero, a genius, a maverick

4) Verbs – to cheat, to deceive, to sin, to lust after, to chicken out, to triumph

It must be noted that like other subtypes of the appraisal, judgment can be expressed overtly usually utilizing particular lexical items such as skillfully, corruptly, lazily etc. which is called ‘inscribed’ or explicit expressions of Judgment. It can also be expressed indirectly which is called evoked or implied rather. In such cases, speakers trigger particular kinds of judgment in their interlocutors by using some neutral-looking words (White, 2001b). This is a useful strategy that-empowers the speakers to say what they want without being held responsible. e.g.

Inscribed Judgment: Jack is lazy.

Invoked Judgment: Jack does not work hard.

C) Appreciation: It is the system by which we evaluate products and processes. In other words, while judgment evaluates human behavior, appreciation typically evaluates natural objects, manufactured objects, text as well as more abstract constructs such as plans and policies, e.g. His speech was interesting. Various types of appreciation are illustrated and exemplified in table 4 in the appendix.

2. Review of literature

As it was mentioned before, robustness of the appraisal theory and its comprehensive analytical tool have led to its application in a wide range of contexts and genres. For example, the appraisal framework in conjunction with Lebov’s (1972) model was used by Page (2003) to explore ways male and female speakers express their feelings in personal narratives. To this end, 23 oral narratives about experiences of childbirth were elicited through informal interviews. The results showed that women’s narratives contained a higher proportion of affect and they usually intensified its use through the appraisal subsystem of amplification (i.e., more recently called graduation) while men used appreciation system more and they usually mitigated their expression of emotions. Furthermore, men usually talked about their emotions indirectly. Page (2003) believes that the results could support previous findings that women put more emphasis on affective meaning while men emphasize informative meaning. However, she warned against generalizing the idea that all women in all contexts emphasize emotions while men do not. 

To explore international students' ability to express their attitudes in English, Ngo et al. (2012) investigated conversational data collected from four Vietnamese postgraduate students enrolled in two universities in Sydney. The data, then, was analyzed for features of attitudinal language using the appraisal Framework developed by Martin and White (2005). The findings showed that the international students with high English proficiency test scores still needed help to be able to express their feelings and opinions effectively. The authors recommended narratives to teachers as a useful source for helping students build their evaluative language repertoire. In order to examine evaluative use of language in academic writing by drawing on Martin and Rose’s (2003) appraisal model, Jalilifar and Moazzen (2014) analyzed the discussion sections in two types of research articles, i.e. ISI journals and Non-ISI journals. They found that there were significant differences between two groups of articles on the use of attitude resources. In other words, ISI articles used evaluative resources of language more effectively hence their discussions were more interactive in nature. The researchers hypothesized that the attitudinal use of language might be an important element of an effective discussion in research articles. 

To examine application of appraisal system in Brazilian Portuguese, Vian Jr. (2008) focused only on realization of graduation system in a Portuguese novel. The researcher found that many categories of appraisal system proposed by Martin and White (2005) applied to the Brazilian Portuguese, however, in comparison with English; there were some differences in the linguistic realization of the system in Brazilian Portuguese such as the use of the diminutive suffix. The author believes that these differences are attributable to the genre and register of the local culture and are, in turn, due to the dialectic relation between text and context. In a pilot study, Liu and Thompson (2009) compared the use of appraisal resources in a Chinese student’s L1 and L2 argumentative writings to explore interpersonal meanings within the EFL context. The student was asked first to write an English essay about a topic and a week later he was asked to write another essay in Chinese about the same topic. The results showed that in the English essay, a higher number of judgment (47.1 %) and appreciation (35.6 %) were used than affect (17.3 %). This pattern of attitudinal resources is regarded as being a characteristic of argumentative genre (Lee, 2006). The Chinese essay, however, contained more appreciation items (55.1 %) than judgment ones (40.8 %). Furthermore, the affect items were the least used (4.1 %). This less use of affect and judgment resources means less disclosure of feelings and avoidance of direct ethical and moral evaluation and can be attributed to differences between western and traditional Chinese writing rhetorics (Wu & Rubin, 2000; Lee, 2008). However, the exploratory nature of such case studies must be borne in mind and their finding should be used with care.

To explore how a narrative utilizes language resources implicitly to instruct the readers to absorb its ethical values, Macken-Horarik (2003) investigated the deployment of appraisal resources in narratives. The data for this study comprised of both a short-written narrative given to sixteen-year-old Australian students in a formal English examination and two of their successful written answers. The study revealed that appraisal resources were used in narratives as strategies for the formation of empathy and discernment in an ideal reader. It also showed that successful readers attended to both explicit and implicit appraisal resources of the narrative which in turn led them to adopt the narrative axiology. In an attempt to explore relations among gender, genre and linguistic choice, Page (2008) analyzed narratives written by male and female writers. The narratives were derived from participants’ personal weblogs. The results showed that in terms of Halliday’s linguistic framework (1994), women used a greater proportion of mental processes of affection while men used greater number of verbal processes. This means that women were more inclined to reveal their emotions and consequently their narratives included more affective elements. For example, in terms of emotional disclosure, the researcher found that the narratives written by women included “evaluative actions” (Labov, 1972) like outbursts of tears, ten times more frequent than the men narratives. 

Ngo and Unsworth (2015) used Martin and White’s (2005) appraisal framework to investigate evaluative patterns in international students’ oral discussion and casual conversations. The focus of this study was on refinement of attitude sub-system of appraisal framework. The participants were sixteen graduate students from universities in Sydney. They were assigned into eight groups. Four groups had Vietnamese conversations and other four groups had English conversations about the same topic. Their talks were video and audio recorded. The audio data was the main source of this research. By drawing on the findings of this study, the researchers refined the original framework developed by Martin and White (2005). They believe that the use of this framework will lead to a more delicate coding of attitudinal meaning in similar contexts which, in turn, can lead to more effective English teaching to international students with non-English speaking backgrounds.

3. Method

3.1. Design 

It was a descriptive quantitative design. It was descriptive based on the fact that the researchers investigated the problem in hand using document analysis using a theoretical model in the study and tried to describe the problem. It was quantitative based on the fact that the researchers used numerical terms and statistical analyses to examine differences in the corpus. A descriptive design allows researchers to describe the phenomenon in details without any need for intervention or manipulation. 

3.2. Materials

After an initial survey of a variety of materials found in 20 Iranian personal weblogs, 48 recounts were chosen. Twenty-two of which were written by female writers while the others were written by males^[1]. The total number of words in male and female data was almost the same. Care was taken to include only those stories complying with Eggins and Slade’s (1997) operationalization of recounts mentioned in the introduction section. Additionally, the data was restricted to a period of four years (2012 to 2016) to control for possible generic variation as a result of the chronological change. For sake of illustration, two sample recounts from the data with their English translations are presented in tables 6 and 7. 

Based on the demographic information provided on the weblogs, the writers whose recounts were chosen for this study aged between 30-40 years old. They had an academic degree of BA or higher in a particular field. The purpose of writing recounts were personal problems, family problems, reminiscence, psychological discharge, etc. The authors took an informal tone in their texts and wrote them for general audiences. The total number of words in the present corpus was 14628. 

Table 6 

A Sample Recount by a Female Weblogger 

Table 7.

A Sample Recount by a Male Weblogger 

3.3 Instrument

In the present study, UAM corpus tool software introduced in the Ahvaz Appraisal workshop (White et al., 2016), Iran, was used to analyze the recounts. The UAM Corpus Tool is a modern and powerful software for annotation of text corpora. It is a very versatile software for annotating linguistic data with the following features: 1) Annotation of multiple texts using the same annotation schemes, 2) Annotation of each text at multiple levels (e.g., NP, clause, sentence, whole document); 3) Searching for instances across levels, e.g., finite-clause containing company-np, or future-clause in introduction; 4) Comparative statistics across subsets, e.g., contrasting conversational patterns used by male and female speakers (O’Donnell, 2012). 

3.4. Procedure

The recounts were converted into a plain text format and fed into the UAM corpus tool. Since, in the present analysis, a bottom-up approach was taken, the researchers searched the data sentence by sentence and identified the realization of each token of attitude in the text first and then decided on its category based on the Martin and White’s (2005) model of appraisal. It must be noted that sentence was considered as the unit of analysis in the present study. However, due to the multifunctional and context-sensitive nature of evaluation elements, in a few cases, the researchers had to look beyond sentence levels to determine the exact type of an appraisal token. As the analysis was primarily quantitative in nature, the quantity and percentage of attitudinal tokens were calculated. To check for reliability, the researchers reanalyzed the data after a two-week interval and conducted Cronbach’s alpha and found an intra-rater reliability index of 0.93. Additionally, a discourse analysist who was familiar with the appraisal framework was asked to analyze 25 percent of the data. The result of this analysis was correlated with that of main researchers and inter-rater reliability index of 0.94 was found.

4. Result and Discussion

The main objective pursued in the current study was to find out whether males and females bloggers used the attitudinal resources differently in their recounts. The analyses of two sets of data revealed that in terms of frequency of appraisal tokens, the women recounts contained a higher percentage of appraisal elements (1.73%) compared to the men recounts (0.99%) despite the fact that the men data contained 138 words more than the women data (Table 8). This finding is in line with previous research (Eggins & Slade, 1997; Page, 2003) demonstrating that talking about emotions is more prominent in women stories (Page, 2008). This in turn might further supports Holmes’ (1998) idea that men’s communication is likely to be focused more on information while females concentrate more on expressing feelings.

Table 8

General characteristics of the data 

In order to examine data from a statistical perspective, the researchers used the chi-square test as shown below and the results showed a significant difference between male and female recounts (P<0.05).  

Table 9 

Chi-square test

4.1. Affect and Appreciation

In addition to above, with regard to attitude subcategories (table 9 below), women used more affect than men (55.56 % and 43.24, respectively) while men used more appreciation (36.49 and 23.02, respectively). 

Table 10

Types of Appraisal

Again, to statistically confirm differences between the groups, the researchers used the chi-square test and its results showed that there were statistically significant differences between male and female recounts in terms of affect, judgment and appreciation (p<0.01). This finding is also similar to that of Page (2003) focusing on the appraisal in childbirth narratives told by men and women. That is, in her study, Page (2003) also found that women used more affect than men while men used more appreciation compared to women. In other words, she found that in expressing the same feelings such as excitement unlike women who presented themselves as those who experience the feeling by using affect, men preferred to attach the emotional quality to an aspect of experience itself by using appreciation. In the present study, however, men used affect and it was found that they were not afraid of expressing themselves as those who experience feelings. For sake of illustration, some examples of affect used by women and men are provided. 

As evidenced by samples of Affect used by men (Table 11), in the present research men did express their feelings directly through Affect. This finding might be attributed to the nature of personal weblogs providing its authors with a kind of anonymity mask under which they can express their feelings more freely. 

Table 11

Sample Affect Utilized by Females

Table 12

Sample affects utilized by males

4.2. Judgment

Judgment was used less in both men and women’s recounts compared to other subtypes of attitude (Table 9 above). However, there was a marked difference in the use of judgment types. That is, men used capacity more (80.00%) than women (18.52%) while women used propriety more (see Table 13 below). In judgment, the speaker evaluates someone’s behavior positively or negatively based on the speaker’s social norms (Eggins & Slade, 1997). Men in this study judged others behavior more based on their physical or mental abilities (i.e., capacity) while women judged others behavior more based on ethical morality which is the domain of right or wrong (i.e., Propriety). Whether this difference in language use can be attributed to gender or not needs further research. Moreover, further research is required to examine whether this pattern appear in other languages or not. 

Table 13

Types of Judgment

The researchers used Chi-square test to examine differences between men and women recounts and the results showed a statistically significant difference (P<0.01). 

4.3. Explicitness: Inscribed vs. Invoked 

All elements of attitude (i.e. affect, appreciation and judgment) reviewed above were analyzed in terms of whether the evaluation was worded explicitly (i.e., inscribed or explicit) or implicitly (i.e., invoked). The results demonstrated that both women and men preferred to express their evaluation explicitly rather than implicitly (Table 13). Evaluation whether in spoken or written forms, monologue or dialogue is dependent on the inter-relationship between speaker/writer and the audience (Page, 2003). Thus, women and men tendency to encode their evaluation explicitly can be attributed to the structure of personal weblogs which give their authors an opportunity to write to total strangers as if they are close friends. This, in turn, led them to express their feelings more freely and directly without being afraid of negative consequences such as negative judgment or face loss.  

Table 14 

Explicitness: Inscribed vs. Invoked 

Finally, it is worth mentioning that while analyzing the data, the researchers came across an interesting concomitant finding regarding the general pattern of evaluation resources in recounts. In fact, the notion that different modes of story-telling can be characterized by their different uses of resources of evaluation (White, 2001a) encouraged us to examine the data more closely. Interestingly, it was found that despite the aforementioned findings on the differences in the number of appraisal elements used by men and women and the difference in the use of subtypes of attitude, a general pattern was seen to be followed by both men and women concerning the frequency of affect, appreciation and judgment in each group. That is, in both groups, affect had the highest frequency (female 55.56%; male 43.24%) followed by appreciation (female 23.03%; male 36.49%) and judgment with the least frequency in each group (female 21.43%; male 20.27%). This intriguing trend might indicate that recounts follow the same pattern about the appraisal elements no matter told by men or women. This might also show the influence of genre on the realization of appraisal and further supports the idea that in various genres, various appraisal subsystems might be foregrounded (Martin, 2000) or even they might be realized in a special pattern as was shown in this study. However, further research is required to examine attitude markers in recounts in various contexts and languages before one can reach a generalization.

5. Conclusion

To investigate the application of attitudinal resources in Persian recounts, the present research used the attitude subsystem of appraisal system as a framework. The results showed that women generally used more attitude resources in their recounts compared to men and regarding the subtypes of attitude, women used more affect than men while men used more appreciation compared to women. Interestingly, these findings were in line with Page’s (2003) study investigating appraisal in childbirth narratives told by men and women and since our study was on a language other than English (i.e., Persian recounts) and it was not suffering from similar limitations mentioned by Page (2003) such as elicited and artificial data. These findings can be considered as further supports for Holmes’ (1998) notion that men communication is likely to be focused on information while females concentrate on expressing feelings. However, since the date analyzed in the present study was limited to the Iranian context, any conclusion on the generalizability of findings need to be avoided till further research addresses the issue across various languages.

Furthermore, unlike their differences, in both women and men recounts, attitude subsystems appeared in a special order in terms of frequency, that is, affect items were more frequent than appreciation items while judgment items appeared the least in the data. This might tell us something about the generic features of recounts and supports White (2001a) proposal that different types of story-telling can be realized through different uses of appraisal resources although care must be taken not to generalize based on this small data and further research is required to examine attitude markers in recounts in various contexts. Moreover, the findings showed that both women and men predominantly encoded their evaluation explicitly which could be attributed to the contextual nature of personal weblogs in which anonymous authors can express their evaluations as if there are talking to their close friends.  

 Finally, the findings of the present study can be considered as evidence of the effectiveness and applicability of appraisal theory as a research tool across language and cultures although this domain needs further investigations. However, the analysis in this study was confined to only attitude component of the appraisal system. Hence, future research are recommended to include other appraisal subsystems (i.e., graduation and engagement) in their studies and also to explore the linguistic realizations of the appraisal system in recounts in languages other than English.

Appendix

Table 1

Types of affect (Adopted from Martin & White, 2005)

Table 2

Types of Judgment: Social Sanctions (Adopted from Martin & White, 2005)

Table 3

Types of Judgment: Social Esteem (Adopted from Martin & White, 2005)

Table 4 

Types of Appreciation (Adopted from Martin & White, 2005)

                               Positive                                                     Negative

Reaction:                        arresting, captivating, engaging …;          dull, boring, tedious …; impact ‘did it                  

                                fascinating, exciting, moving …;                        dry, ascetic, uninviting …;

grab me?’                        lively, dramatic, intense …;                      flat, predictable, monotonous …; 

                remarkable, notable, sensational …                unremarkable, pedestrian …

Reaction:                        okay, fine, good …                                       bad, yuk, nasty …; 

quality ‘did I like it?’      lovely, beautiful, splendid …;                  plain, ugly, grotesque …;

        appealing, enchanting, welcome …           repulsive, revolting, off-putting …

Composition:                   balanced, harmonious, unified,             unbalanced, discordant, irregular,

 balance ‘did it hang        symmetrical, proportioned …;              uneven, flawed …;

together?’                       consistent, considered, logical …;           contradictory, disorganized …; 

                shapely, curvaceous, willowly …                      shapeless, amorphous, distorted …

Composition:                  simple, pure, elegant …;                          ornate, extravagant, byzantine …; 

Complexity                    lucid, clear, precise …;                              arcane, unclear, woolly …;

‘was it hard to follow?’   intricate, rich, detailed, precise …           plain, monolithic, simplistic …

Valuation                        penetrating, profound, deep …;              shallow, reductive, insignificant …; 

‘was it worthwhile?’       innovative, original, creative …;            derivative, conventional, prosaic …;

        timely, long-awaited, landmark …;          dated, overdue, untimely …; 

        inimitable, exceptional, unique …;          dime-a-dozen, everyday, common; 

        authentic, real, genuine …;                        fake, bogus, glitzy …;

        valuable, priceless, worthwhile …;             worthless, shoddy, pricey …; 

        appropriate, helpful, effective …              ineffective, useless, write-off …