Title Understanding the Language of Evaluati l Theory Author(s) Tokunaga, Anne Editor(s) Citation 言語と文化. 2010, 9, p.1-18 Issue Date 2010-03-31 URL http://hdl.handle.net/10466/12726 Rights http://repository.osakafu-u.ac.jp/dspace/

Understanding the Language of Evaluation : An Introduction to Appraisal Theory Anne Tokunaga Abstract Expressing interpersonal meanings such as glvmg opldlons, making judgments and moderating evaluative meanings are a fundamental part of any language and consequently, students are regularly asked to respond personally to give their opinions on various issues and topics in English. However, despite the significance of evaluation and its role in daily life, this aspect of language is rarely explicitly taught. Moreover, within linguistics it is only in recent years that much of the research on evaluation been done. This paper provides an introduction to evaluation as well as to Appraisal Theory which has been developed within the tradition of Systemic Functional Linguistics. The paper explores the appraisal framework and identifies some of the resources that may be used to express feelings, make judgements and give valuations, as well as moderating opinions and expressing degrees of certainty and inclination. Keywords: Appraisal; Attitude; Evaluation; Interpersonal Meanings; Systemic Functional Linguistics Introduction Halliday (2004) considers what he terms interpersonal meaning, and which includes the expression of one's attitudes, as one of the major functions of language, while Thompson and Hunston (2000: 6) describe the expression of the writer's opinion as an important feature of the language, adding that "it needs to be accounted for in a full description of the meanings of texts." Yet despite the importance of evaluation, linguistic research has in the past focused more on referential meaning and how language represents states of affairs or things in the world, with the result that this aspect of language has been rarely explicitly taught in the classroom. It is only in recent years that the expression of a speaker's opinion has received more attention with the recognition that we are not just engaged in communicating in order to download information about the world around us. A number of linguists using a wide range of terms that have included 'connotation' (Lyons 1977), 'affect' (Besnier 1993), 'attitude', 'modalization' and 'modulation' (Halliday 1994), 'attitudinal stance', 'epistemic stance', 'style' (see Biber and Finegan 1989 and Conrad and Biber 2000) 'evaluation', and 'appraisal' (Martin 2000)1 have shed light on the fact that along with expressing meanings about the real world, language works to I see Thompson and Hunston 2000: 2-5 and Martin and White 2005: 38 for a discussion on the terminology used, and an overview to the various approaches to evaluation. -1-

simultaneously allow us to communicate our interpretations of our world and volunteer these evaluations in order to position ourselves in relationships with others. Evaluation Defining evaluation Thompson and Hunston while recognizing that the many different terminologies that are used to describe evaluation are a consequence of the fact that it is a complex phenomenon with many facets and functions nevertheless choose to use the superordinate term 'evaluation' and define it as The broad cover term for the expression of the speaker's or writer's attitude or stance towards, viewpoint on, or feelings about the entities or propositions that he or she is talking about. That attitude may relate to certainty or obligation or desirability or any of a number of other sets of values. (Thompson and Hunston 2000: 5) Entities and propositions An important distinction that Thompson and Hunston (2000: 3-4) make concerning evaluation is how language works to express opinions about entities and opinions about propositions. While entities generally involve positive or negative feelings and are realized lexically, opinions about propositions involve degrees of certainty and are realized grammatically (see also Martin and White 2005: 38-39). Traditionally, Systemic Functional Linguistics has explored interpersonal meaning more in terms of grammaticality, where aspects of grammar are used to signal a writer's degree of commitment to the truth of a particular 'knowledge-based' proposition or their commitment to the obligation or necessity of a particular 'activity-based' proposal. These meanings can be realized through metaphorical expressions of modality, for example, I think and there is no doubt that; modalization, otherwise termed 'epistemic' modality which is used to express meanings of probability and usuality; and modulation, sometimes referred to as 'deontic' modality and which involves terms like should, must and required to express notions of obligation and inclination. While Thompson and Hunston (2000: 20) maintain that "the grammar of modality has been very fully explored", they make the point that affective evaluation; that is, evaluation about entities "on the other hand, has been far less investigated" with Martin (2000: 144) concurring, adding that what has been elided in studies of interpersonal meaning in Systemic Functional Linguistics "is the semantics of evaluation - how the interlocutors are feeling, the judgements they make, and the value they place on the various phenomena of their experience." -2-

Parameters of evaluation In Thompson and Hunston's definition of evaluation above, the "sets of values" refer to 'evaluative parameters' (see Bednarek 2006: 35-37) and it is this, they argue, that accounts for the wide range of terms that are used for evaluation. The fact is that "the act of evaluating something can be done along several different parameters," (Thompson and Hunston 2000: 22) and not just the two mentioned above: the good-bad or positivenegative parameter, which is used to express opinions about entities; and the certainty parameter which is used to express opinions about propositions. In total, Thompson and Hunston (2000: 25) identify four parameters of evaluation which together with the good-bad and certainty parameter include an expectedness or obviousness parameter and an importance or relevance parameter. Bednarek (2006) in her parameter-based framework of evaluation, which draws on work done by Lemke (1998), identifies at least nine semantic categories, arguing that evaluative statements appraise in terms of importance, usefulness, comprehensibility and reliability and so forth. On the other hand, there are some approaches to evaluation which don't use the term 'evaluative parameter'. In the case of Appraisal Theory (White 1998: 75), there is a differentiation between three 'axis' or 'systems' of evaluation: Attitude (loosely corresponding to the good-bad, positive-negative, emotivity parameters) Graduation (corresponding more or less to the importance and reliability parameters) and Engagement (corresponding roughly to the possibility, necessity and evidentiality parameters). The relationship between parameters of evaluation and genres Thompson and Hunston (2000: 24) make the important point that "different parameters of evaluation can have specific roles to play in a discourse," which means that, "certain genres will prioritize either evaluations of good-bad or certainty." For genres where the central function is to assess the worth of something, such as book reviews, the "good-bad parameter is more significant" (ibid.). Typical words such as 'good', 'bad', 'wonderful', 'dreadful' provide explicit evaluation along the good-bad or positive-negative parameters. In the case of genres such as academic research, the parameter of certainty is important since this genre has the purpose of building knowledge claims, which then must be assessed in regards to the degree of certainty that may be attached to any particular claim. The basic features of evaluation Although Thompson (2005: 76) states that there are other parameters of evaluation which can help to reveal what "kinds of values are established in any particular genre" he also maintains that the most simplest and basic one, under which all the others may be subsumed, is the good-bad parameter. This parameter reflects the fact that we classify evaluation as any statement that conveys a negative or positive assessment and although we often use the general terms of 'good' and 'bad', we do have more restrictive evaluative -3-

terms that convey positive and negative evaluations such as 'delicious' which means 'good-tasting', and beautiful which means 'good-looking.' Moreover, whether something is good or bad is often assumed when evaluating. For example, in a book review statements such as 'this is an important book' (evaluating on the importance parameter) and 'this is a useless book' (evaluating on the relevance or usefulness parameter) nonetheless imply that the book is good or bad. While a consumer products' review of a car may determine that reliability and efficiency (evaluating on the reliability parameter) are important criterion for establishing whether a particular car is good or not. As can be seen, the explicit term 'good' may be applied to many things and unlike terms which express one's feelings towards something such as 'I love it', the term 'good' when applied to a given entity invokes criteria (and values) regarding how well that entity meets or fulfills the qualities or standards of the category to which it belongs. Thus an examination of the purpose or goals of an entity and the category to which it belongs, together with whether or not the entity achieves those goals can help to define 'goodness' and 'badness'. In doing so, 'what achieves our goals' and 'what impedes the achievement of our goals' can be glossed respectively as 'what is good' and 'what is bad' (see also Hunston (1985; 1989 in Thompson and Hunston 2000: 14), notwithstanding the fact that standards invoked in a particular context may vary from culture to culture. Alternatively, actions tend to be evaluated using the terms 'right' or 'wrong' and so a passenger in a taxi may say, "That's the wrong tum," to direct a driver not to tum that way. In contrast to the terms 'good' and 'bad', which indicate whether an entity meets the purposes of their category, 'right' and 'wrong' can signal the 'correctness' of an action or whether some action is appropriate based on ethical grounds. Another consideration concerns statements of deontic modality in which we express whether something should or should not be done, as in the case above where the passenger can add, 'You ought not to take that turn'. Such expressions of obligation convey elementary beliefs and values that we assume should be evident and imply that acting on the basis of these values IS a desirable, 'good' or the 'right' thing to do (see Fairclough 2005: 171). The context-bound nature of evaluation Rather than simply being the expression of a single individual's comments on the world, Thompson and Hunston (2000: 6) argue that each evaluation that is expressed tends to build up a socially shared value-system or ideology that may reflect what a sub-group or even a whole society counts as good or bad, true or untrue or what should or should not happen. This according to Thompson and Hunston (2000: 22) inevitably means that, "evaluations of good and bad are dependent on the value system underlying the text" and this suggests that in assessing whether an instance of discourse is evaluative or even how it -4-

functions evaluatively, one must take into account the fact that the evaluation is bound both to the culture and its particular context of situation. To illustrate this the expression, 'practically deserted', may seem more referential in content but when it appears in the context of situation of a beach holiday, it conveys positive evaluation, whereas used in the context of a description of an underground car park in an unfamiliar city it would suggest a negative evaluation (Thompson and Hunston 2000: 6). In another example Thompson and Hunston (2000: 22) describe a small subgroup of humans who enjoy the sport of ice-diving (throwing oneself into icy water on a winter's day) and who positively evaluate the activity as an aesthetic experience dismissing the obvious physical unpleasantness that this activity would certainly have. A further example provided by Channell (2000: 43) is the adjective 'fat', which also could be interpreted purely as referential but in many post-industrial cultural frameworks when applied to humans, particularly females, provides a negative evaluation, but in cases where the referents are animals, the evaluation is positive. Finally, in some discourses within certain cultures some statements may become evaluative. For instance, 'he's an atheist' would likely have an evaluative meaning if it were used to refer to a candidate within an American presidential campaign and would categorize that person as being 'undesirable' for a large number of voters, whereas it may be simply a statement of fact in other discourse contexts, for example the filling out of a census form. Therefore, as Firth (1935: 37 in Channell 2000: 40) noted many years ago "the complete meaning of a word is always contextual, and no study of meaning apart from a complete context can be taken seriously." It is thus only "through consideration of occurrences in context" (ibid.) that one can determine the evaluative status of a word. The role of co-text in evaluation Not only is the context of situation and context of culture crucial in understanding how to construe evaluation, co-text also plays a critical role. Evaluative meanings can be expressed by almost all classes of words and grammatical structures. Although attitudinal meanings are often expressed using explicitly lexical items, according to Martin & Rose (2003: 54) speakers are able to distribute both explicit and implicit evaluative meanings throughout the whole text, which then work together to form patterns "running through the text" which construct "the 'stance' or 'voice' of the appraiser" (ibid.) and which then position the audience towards a particular point of view or interpretation of the content. For example, while the word 'student' may be a purely referential term, in some contexts it can also have either positive or negative connotations. The co-text is important in showing the presence of evaluation, since comparators like just can signal a negative evaluation as in the following exchange: A: Who is he? B: He'sjust a student... (Thompson and Hunston 2000: 17) -5-

Moreover, Hood (2004 in Hood and Forey 299) point to the use by speakers of intensifiers to grade up a particular meaning within a clause, for example, "an awful lot of things have been happening." The intensifier gives the clause a subjective slant and tips off the reader to interpret the meaning of the clause as evaluative in some respect. Related to co-text is what Halliday (1979: 66-7 in Martin and White 2005: 19) identifies as the. prosodic structure of interpersonal meaning. This is where "evaluation tends to be found throughout the text rather than just being confined to a particular part of it" Martin and White 2005: 24). This is called 'prosodic' "since the meaning is distributed like a prosody throughout a continuous stretch of discourse" (Martin and White 2005: 19). Examples of prosody: It's a dirty, rotten stinking lousy bloody low filthy two-faced lie. What an amazing album. 'Love Struck Baby' starts it off and is one of their most famous song. I suppose he might possibly have mightn't he. (Martin and White 2005: 20) Evaluation and the negotiation of social relationships In communicating evaluations, speakers are doing more than just interpreting the world and proffering their attitudes and opinions for the sake of expressing something. Even though this process of professing short-term feelings and judgements may lead to them being established as long-term values, which in turn may contribute in the construction of a speaker's persona, speakers by proffering their evaluations are as well simultaneously engaged in constructing a relationship between themselves, and their audience. It is through the sharing of opinions, judgments and values that we align ourselves to one position or another, influencing our audience's personal reaction which enables the construction of a relationship between us and our audience in which solidarity is built and negotiation can take place. For example, if you are watching a game of soccer and someone says, "That was a lucky goal!" it means something quite different from "That was a fantastic goal!" The first clause aligns the speaker with the opposition while the second clause aligns the speaker with the team that scored the goal. Similarly, words such as 'brave' or 'foolhardy' may be used to describe someone who undertakes to complete a risky act. If a speaker chooses the word 'brave' they are aligning themselves with people who positively value the person's act. In describing politicians, words like 'clever' and 'astute' can be used, and if a speaker chooses to use the word 'astute' they are positioning themselves with people who are more positively inclined towards that politician. Moreover, effective speakers and writers can manipulate their audience towards a particular point of view or ideological position by building up particular meanings in a text -6-

(as with the use of prosody); characterizing situations in certain ways; and through placing evaluation strategically in a text so as to persuade people to see things in a particular way. The example below, characterizes a situation as a problem: 'My husband runs his own business and is a workaholic'... he kept putting off our plans for a break and I got so fed up I went away with four friends... How can I get him to see I'd rather go with him? (Woman's Own, summer 1998: 32. In Thompson and Hunston 2000: 6 & 8) Thompson and Hunston make the point that by presenting this situation as a problem (even though there are two participants - the husband and wife - and there are potentially not one, but two problems) it is very hard for the reader not to accept it as such. A third technique that a writer can use to manipulate their audience is to place the evaluation at the start of the clause, which is the part which does not express the main point of the clause. Information here is seen as 'given' and unlike 'new' information (which is found at the end of the clause) it is harder to challenge. An example is the use of the evaluators, 'frivolity' and 'triviality' in the following text: Similarly, the frivolity and triviality of much modern discourse... is a product of a decade... (The Guardian, 30-5-98, p. 23 in Thompson and Hunston 2000: 9) A fourth technique is the use of 'discourse labels' (Francis 1986; 1994 in Thompson and Hunston 2000: 9) such as 'this claim', 'this admission', 'this nonsense' which can also evaluate previous text, and since "they occur as given information the reader is unlikely to question their validity" (Thompson and Hunston 2000: 9) By building up these kinds of meanings in a text, effective writers can align readers around a set of values about communal activities to produce a sense of belonging and community. In other words, the audience is drawn to a particular point of view concerning the content or ideological position that is presented as being the most natural one to hold. The modeling of Evaluation in Systemic Functional Linguistics Tenor In Systemic Functional Linguistics, evaluation falls under the register variable of tenor. Register functions to "describe[s] the impact of dimensions of the immediate context of situation of a language event on the way language is used" (Eggins 2004: 9). Halliday (1985) characterizes tenor as follows: Tenor refers to who is taking part, to the nature of the [communicative] participants, their statuses and roles; what kinds of role relationship obtain, including permanent and temporary relationships of one kind or another, both the types of speech roles -7-

they are taking on in the dialogue and the whole cluster of socially significant relationships in which they are involved. According to Halliday's description above, an examination of tenor can answer the following questions: Who is involved in the interaction? What are the roles or subject positions of the participants? For example, the role of student/lecturer, customer/salesperson and fan/fan. What type of speech role are they adopting? For example, the speaking positions of the participants such as who is giving information, who is demanding information, and who has the right to talk for longer periods. What are the relationships between the participants? For example, is it a relationship where one participant has more knowledge or capacity than the other? What are the statuses of the participants? In the 1980s, tenor was further mapped out by Poynton (1985), who influenced by the work of Brown and Gilman (1960/1971, in Poynton 1985), added two key tenor variables, which answer the following questions: What is the power relationship between the interactants? Here the situation can be characterized by the degree of control one participant can exercise over another. According to Martin and White (2005: 29), roles between interactants therefore are either equal or unequal in respect to the following five factors: age, gender, ethnicity (concerned with racial, religious and other cultural divisions), capacity (disabilities/abilities of various kinds) and class (based on the distribution of material resources ). What is the social distance or degree of involvement between interactants? Here the situation can be characterized in terms of 1) the degree or depth of contact; 2) whether the roles we are playing are those that bring us into frequent or infrequent contact; 3) how regularly the contact occurs; and 4) whether the contact is during work or leisure activities (see Martin 1992: 528-9). So factors outside the text such as: how much contact participants have; the equalities or asymmetries within social relationships; their social roles and speaking positions all have an impact on how language is used and realize meanings at the interpersonal level. For example, when chatting with friends in an informal situation (-power, +contact) we invite empathy by using words that express our feelings explicitly, while when we are in more formal situations (+power, -contact) we keep our attitudes to ourselves or use more objective language realizing that we can easily run the risk of being excluded and alienated through expressing our opinions too freely with people we don't know well. -8-

Interpersonal meaning In accord with Halliday (1994), Eggins (2004: 11) states that, "the fundamental purpose that language has evolved to serve is to enable us to make meanings with each other." She goes on to state "People interact in order to make meanings: to make sense of the world and of each other. The overall purpose of language, then can be described as a semantic one, and each text we participate in is a record of the meanings that have been made in a particular context" (Eggins 2004: 11). Evaluation is related to the construction of interpersonal meaning. Within each text one can find strands of interpersonal meaning running throughout it which expresses the speaker's subjective presence in texts as they adopt stances towards both the material they present, and those with who they are communicating; feelings, judgements or valuations towards the subject matter and how they position their readersllisteners to do likewise; particular persona and how they align or disalign themselves with actual or potential respondents (see Martin & White, 2005:1). As shown in figure 1 below, there is a relationship between the subject matter and its expression which is described as one of realization or encoding (Eggins 2004: 14). This realizational relationship extends all the way from the Context of Culture (ideology) through to the very concrete words, structures, sounds and graphology of text. Here, "tenor is realized through interpersonal meanings, which are in tum realized through the grammar of the clause as exchange" (Eggins 2004: 187). The realization relationship, in figure ], can be read in both a predictive and a deductive direction. For example, if one has the details concerning the context, it is possible to predict with reasonable accuracy the typical linguistic choices which will characterize a text, while if one has a text, it is possible to "deduce the context within which it was produced and of which it is a realization" (Eggins 2004: 328). Martin and White (2005: 34) regard evaluation for which they use the term 'Appraisal' as one of "the three major discourse semantic resources available for construing interpersonal meaning alongside involvement and negotiation." While the focus of negotiation is on the interactive aspects of discourse speech function, and the exchange structure; and the focus of involvement is on resources such as terms of address, expletives and slang; appraisal focuses on the "semantic resources used to negotiate emotions, judgements, and valuations, alongside resources for amplifying and engaging with these evaluations. The[se] -9-

resources all involve grading, which is to say that the meanings involved can be adjusted by degree to reflect the strength of the evaluation" (Martin 2000:145). Context Text Context of Culture Context of Situation -------- -~-~... ~~ ': -------- Discourse Semantic Stratum ('meaning') Lexico.grammatical Stratum (,wording') Ideology Genre I~ -----------------------------------------------~~~~~!~~ ------ power (status) Negotiation speech function solidarity (contact) Tenor Interpersonal,.... Meaning, ~ I Appraisal I Involvement I. affect!. naming I judgement I technicality!. appreciation I. abstraction exchange I. engagement I i: anti language m~~- - -~-:-!~::;: lex;' -;,:~:""~ tagging!. modal verbs I. technicallexis I. modal adjuncts '['. specialized I polarity I lexis I pre/numeration I. slang I!. intensification I. repetition i. manner; extent I. logico I semantics i. vocation f-----------t-~------ ---: ~-~- - ---~.. -_t----~~~~-i Expression Stratum tone (&key)!. loudness j 'accent' ('sounds and letters') I. pitch movement I whisper Graphology!Phonology I i voice quality I. actronyms I '. phonoaesthesia I. 'pig latins' I. formatting I. secret scripts Figure 1. Interpersonal semantics in relation to lexicogrammar and phonology. Adapted from Martin and White 2005: 35 and Eggins 1994: 113., I. taboo lexis I. grammatical metaphor Background to Appraisal Theory Appraisal theory was developed in the early 1990s by linguists working in conjunction with Jim Martin in Sydney, beginning with work on narratives and responses by Martin (1996), Rothery (1994) and Rothery and Stenglin (1994), work on media by Iedema, F eez -10- ---

and White (1994), and later work work on conversation by Eggins and Slade (1997) and history by Coffin (1997). As explicated above, Appraisal Theory has been remodeled from work in Systemic Functional Linguistics on interpersonal meaning. In studying Appraisal we are studying how interpersonal meaning gets made. Through an examination of the choices that speakers make regarding the use of appraisal resources, Eggins (2004: 184) suggests that "We can see speakers making meanings about such interpersonal dimensions as: the power or solidarity of their relationship, the extent oftheir intimacy, their level of familiarity with each other and their attitudes and judgements." An introduction to Appraisal Theory Appraisal refers to the grammatical and lexical resources for expressing evaluation which involves the expression of attitudes, including reacting emotionally (Affect), judging morally (Judgement), and evaluating aesthetically (Appreciation); the modulation of meanings by either intensifying or by softening them (Graduation); the indication of the degree of commitment made by the speaker and the stance which the speaker adopts towards the value positions and their relationship with the audience (Engagement). Attitude Attitude, together with Graduation and Engagement, forms one of the three sub-categories of Appraisal. Expressions of Attitude "playa key role in aligning people in relationships of solidarity" (Martin and Rose 2003 in Hood and Forey 2005: 297). In tum, Attitude is constituted by three dimensions, which are referred to as 'Affect', 'Judgment' and 'Appreciation'. Affect is concerned with the expression of an individual's feelings and emotions; Judgement with the evaluation of people's behaviour and actions according to various normative principles; and Appreciation deals with valuations about the things, events or phenomenon around us. In more detail, Affect is modeled as "a semantic resource for construing emotions" (Martin 2000: 148) of not just the author, but also the responses attributed to other social actors (White 2001: 10). As a fundamental characteristic of evaluation in general, Affect is made up of positive or negative assessments, in this case, about feelings or emotions. These feelings or emotions can be divided according to whether they are felt by the participant as an ongoing general mood (e.g. the boy was happy) or directed at or reacting to someone or thing (e.g. The boy liked the present). Most feelings offer lexicalizations that can grade along a scale of low, median and high intensity, as in the following words: like - love - adore; dislike - hate - detest (Martin 2000: 149). Furthermore, Affect can be classified according to whether the emotions are to do with: well-being (anxiety, fear confidence and -11-

trust etc.); the pursuit of goals (dis/pleasure, curiosity, respect, dis/comfort, etc) or to do with what Martin (2000: 150) calls the 'affairs of the heart' (sadness, anger, happiness, love, etc) (see also Macken-Horarik 2004: 297). The use of Affect in a text is a strategy for positioning listeners and for negotiating solidarity with them, as Macken-Horarik (2004: 297) notes "Because Affect focuses on the feelings and emotional dispositions of individuals, they are especially important in creation of reader empathy." Judgement has to do with the explicit evaluation of people regarding how they should or shouldn't behave, and as these evaluations are made by reference to socially determined expectations regarding behaviour, this area of analysis is highly culturally specific. The framework for judgement divides into two categories: social sanction which has to do with the assessment of human behaviour in regards to society's system of moral and legal norms. It relates to firstly, how truthful or 'real' someone is, and secondly, how ethical someone is. The second category is social esteem which has to do with evaluations of admiration and criticism which may effect someone's social standing in their community, but would not lead to any legal consequences. It has to do with how normal or unusual someone is, how capable they are and finally, how resolute or tenacious they are (see Martin 2000: 156). Appreciation involves once again, making positive or negative evaluations of qualities of processes and products in culture, and can also include evaluations of human beings when they are objectified. It comprises three variables - reaction, valuation and composition, which are each aligned with one of the three metafunctions of language: interpersonal meaning, experiential or ideational meaning, and textual meaning (Rothery and Stenglin 2000: 238). According to Martin (2000: 160), "Reaction has to do with the degree to which the text/process in question captures our attention", it therefore describes the emotional impact of the work on its audience and is aligned to the interpersonal meaning. On the other hand, valuation concerns "our assessment of the social significance of the text/process" (Martin 2000: 160) and is related to ideational meanings. Finally, composition, according to Rothery and Stenglin (2000: 238) describes "the texture of a work in terms of its complexity or detail" and concerns "perceptions of proportionality and detail in the text or process" (Martin 2000: 160). It is therefore related to textual meaning. Martin (2000: 147) makes the following important point that all the dimensions of Attitude encode feeling, although Affect is the most basic system, while Judgment and Appreciation represent an 'institutionalization' of the language into 'uncommon sense discourse'. He goes on to elaborate this point: As Judgement, Affect is recontextualized as an evaluation matrix for behaviour, with a view to controlling what people do. As Appreciation, Affect is recontextualized as an evaluation matrix for the products of behaviour (and wonders of nature) with a view to valuing what people achieve. In Halliday's terms judgement institutionalizes -12-

feelings as proposals (about behaviour) whereas appreciation institutionalizes feelings as propositions (about things). Even though feelings are encoded in all three dimensions of Attitude, Hood and Forey (2005: 298) make the case that the expression of Attitude as Affect or Judgement is more personal than expressing an Attitude as Appreciation, arguing that since "Affect encodes feelings and emotions, and Judgements encodes the evaluation of people and their behaviour", expressions of Affect or Judgement in a discourse contribute to a more personal stance than Appreciation, which "is to do with the institutionalization of Affect in the realm of aesthetics, and the valuing of things rather than people" (ibid.). Realization of Attitudinal meanings Evaluative statements are in most cases realized as relational processes where the attribute, which may be an adjective or a noun phrase, realizes the evaluation. In the case of Affect, Martin and White (2005: 58) point out that the grammatical frame that characterizes it "is a relational attributive process with a conscious participant involving the verb feer' (ibid.): Affect {person feels affect about something} {it makes person feel affect that [proposition]} I feel happy (about that/that they've come). It makes me feel happy that they've come. For judgement, Martin and White (2005: 59) characterize the frame as a relational attributive process ascribing an attitude to some person's behaviour. They (ibid.) provide the following example: Judgement {It was judgement for person/of person to do that} {(for person) to do that was judgement} It was silly of/for them to do that. (For them) to do that was silly. Martin and White (2005: 59) distinguish the grammatical frame of Appreciation as a mental process ascribing an attitude to a thing. They (ibid) provide the following example: Appreciation {Person consider something appreciation} {Person sees something as appreciation} I consider it beautiful. They see it as beautiful. -13-

Along with the canonical realization of Attitude through the use of the relational attributive process (e.g. with an adjective, 'This book is good,' or a noun phrase, e.g. 'This is a bad novel'), there are a number of realisations for a given attitude that "diversify across a range of grammatical structures" (Martin and White 2005: 10). Evaluative statements can be realized through mental processes where the evaluative element is a verb. (e.g. 'He liked the book,' 'The novel interests me', 'I was bored by the story.') through other types of processes such as 'behavioural' verbs (e.g. 'He wept.') through nominalised realisations of qualities (turning adjectives into nouns), e.g. joy, confidence, insecurity, and processes (turning verbs into nouns), e.g. grief, despair. with adverbs which are typically circumstances of manner, (e.g. 'The story was put together well') or comment adjuncts (e.g. 'Sadly, the story didn't end well') through exclamations which are alternatives to evaluative statements, e.g. ('What a wonderful book!' instead of 'This book is wonderful.') (see White 2009, Fairclough 2003). Implicit and explicit evaluation Evaluations can be made directly in the text or implicitly. Explicit, otherwise called inscribed appraisal is clear and overt (and as such is harder to ignore or resist) and uses non-metaphorical language with lexical items that, as White (2009: 11) points out are stable attitudinally and which therefore typically have a fixed attitudinal value or meaning across a wide range of texts. Examples of explicit attitudinal lexis include, 'skillfully', 'lazily' and 'happy'. While most cases of appraisal are inscribed, appraisal can also be evoked. In the appraisal framework, implicit evaluation are termed "invocations" or "tokens" and occur when the ideational meaning - the 'non-evaluative' meaning has an emotional or evaluative impact (see Lipson 2006: 94). As White (2009: 11) maintains, 'factual' or ideational meaning has "the capacity in the culture to evoke a judgement or an appreciation on the part of the reader/listener {depending upon the reader's social/cultural/ideological reading position)." For example, sunny weather and beautiful places are associated with happiness. This isa token of positive Affect shared across many cultural groups. However, even in this case one needs to be careful about reading position when analyzing ideational meaning as tokens of affect. As Martin (2000: 154) notes "it's easy to imagine scenarios in which rain would bring great joy" as is the case for farmers facing a drought. Graduation Additionally, it is recognized that evaluation depends on resources that are used to modulate meanings by either intensifying or by softening them. Graduation is divided into categories: 'force' (showing how strong or weak the intensity is) and 'focus' {adjusting -14-

category boundaries). A speaker can either increase or decrease the intensity or force of an utterance by using words like, slightly, somewhat, rather, very, entirely; or blur or sharpen the focus of their semantic categorizations by using words like, sort of, kind of, true and pure as in, 'I'm kind of upset by what you said' and 'He's a true friend' (Martin and White 2005: 94). Engagement Engagement relates to the degree to which a text is monoglossic or heteroglossic (Bakhtin, 193511981 in Hood and Forey 2005: 295). In other words, the way the text is either opened up or closed down to other voices which allows a writer to adopt a position which indicates the degree of commitment to the appraisal being expressed (see Thompson and Hunston 2000: 142). In more heteroglossic texts the listener is seen as "having more or less potential to 'interact'" (Hood and Forey 2005: 296) and so is more inclusive. Hood and Forey (2005: 296) provide examples of resources for Engagement that allow the projection of others' words into the text, either indirectly as in "... we've heard that from a lot of previous speakers", or in the form of a direct quotation as in " X mentioned earlier that their program is... ". Other common resources for signaling degrees of openness to other unspecified voices are through the use of counter-expectancy markers ('but', 'just') which function to imply other countered voices (Martin and Rose 2003 in Hood and Forey 2005: 296) and modality markers of possibility or inclination ('suppose', 'if possible', 'would like to') "which construe propositions as in negotiation with other voices" Hood and Forey (2005: 296). Conclusion Appraisal theory provides a framework for the study of evaluative language that can help English language students better understand and recognize evaluative meanings in a text, and secondly, enable them to be more able to employ interpersonal resources available in language to express their attitudes in everyday situations and in more academic contexts. The tools provided by appraisal theory enable students to investigate what kinds of things they can comment on when responding to some entity (the 'what') such as the feelings they experienced, aspects of human behaviour (e.g. the capacity of the writer or the moral behaviour of the characters) and the qualities present in the work (the impact it has, its social value and composition). Moreover, the framework provides a clear picture of the resources that are needed to 'up-scale' or 'down-scale' these evaluations as well as the language resources that are used to align ourselves with our audience and the propositions that are expressed. References Bednarek, M. (2006) Evaluation in Media Discourse: Analysis of a Newspaper Corpus. London: Continuum. -15-

Butt, D., R. Fahey, S. Feez, S. Spinks and C. Yallop (2000) Using Functional Grammar: An Explorer's Guide (2 nd ed.). Sydney: NCELTR. Bloor, M. and T. Bloor (2007) The Practice of Critical Discourse Analysis: An Introduction. Oxford University Press. Brown, R. and A. Gilman (1960) 'The pronouns of power and solidarity' In T. Sebeok (ed.), Style in Language. Cambridge: MIT Press. 253-76. Channell, l. (2000) 'Corpus-Based Analysis of Evaluative Lexis.' In Evaluation in Text. Authorial Stance and the Construction of Discourse. Oxford: Oxford University Press. Eggins, S. (1994/2004) An Introduction to Systemic Functional Linguistics. London: Continuum. Fairclough, N. (2003) Analysing Discourse. London: Routledge. Halliday, M.A.K (1994/2004) An Introduction to Functional Grammar. London: Edward Arnold. (2004 third edition revised by Christian Matthiessen). Hood, S. (2004) 'Managing attitude in undergraduate academic writing: a focus on the introductions to research reports'. In Ravelli, L.l & R.A. Ellis (eds) Analysing Academic Writing: Contextualized Frameworks. London: Continuum. Hood, S. & G. Forey (2005) 'Introducing a conference paper: Getting interpersonal with your audience.' In Journal of English for Academic Purposes, Vol. 4: 291-306. Hunston, S. & Thompson, G. (Eds.). (2000) Evaluation in Text. Authorial Stance and the Construction of Discourse. Oxford: Oxford University Press. Iedema, R., S. Feez and P. R. White (1994) Media Literacy. Write it Right Literacy in Industry Research Project. Erskineville: Disadvantage Schools Program, Metropolitan East Region, NSW Department of School Education. Lemke, l.l. (1992) 'Interpersonal meaning in discourse: value orientations'. In M. Davies and L. Ravelli (eds), Advances in Systemic Linguistics: Recent Theory and Practice. London: Pinter. 82-194. Lemke, l.l. (1998) 'Resources for attitudinal meaning evaluative orientations in text semantics'. Functions of Language. Vol. 5.1. 33-56. Lipson M. (2006) Exploring Functional Grammar. Bologna: Quaderni del Centro di Studi Linguistico-Culturali (CeSLiC), CLUEB. Martin, l.r. (1996) 'Evaluating disruption: symbolizing theme in junior secondary narrative'. In Hasan, R. and G. Williams (eds.) Literacy in Society. London: Longman. Martin, J.R. (1997) 'Analysing genre: functional parameters'. In Christie, F. and l.r Martin, (eds.) Genre and Institutions: Social Processes in the Workplace and School. London: Cassell. Martin, l.r. (2000) 'Beyond exchange: appraisal systems in English', In Hunston, S. and G. Thompson (eds.) Evaluation in Text: Authorial Stance and the Construction of Discourse. Oxford: OUP. Martin, l.r. & D. Rose (2003) Working with discourse: meaning beyond the clause. London and New York: Continuum. Martin, J.R. & P.R.R. White (2005) The Language of Evaluation: Appraisal in English. -16-

Basingstoke: Palgrave Macmillan. Poynton, Cate (1985) Language and Gender: Making the Difference. Geelong: Deakin University Press. Poynton, Cate (1990) Address and the Semiotics of Social Relations: A systemicfunctional account of address forms and practices in Australian English. Unpublished Ph. D Thesis. Rothery, J. & M. Stenglin (2000) 'Interpreting literature: the role of Appraisal'. In Unsworth, L. (ed.), Researching Language in Schools and Communities: Functional Linguistic Perspectives. London: Cassell. Thompson, G. (2004) Introducing Functional Grammar (2 nd ed.) London: Hodder. Thompson, G. & Hunston, S. (2000). 'Evaluation: An introduction.' In Hunston & Thompson (eds.), 1-27. White, Peter R. R. (2004) 'Appraisal outline,' The Appraisal Website. Retrieved May 2008 from www.grammatics.com/appraisall. White, P.R.R. (2003) 'Beyond modality and hedging: a dialogic view of the language of intersubjective stance'. Text - Special Edition on Appraisal. 259-84 White, Peter R. R. (2004) 'An Introductory Course in Appraisal Analysis,' 'An Outline of Appraisal,' and 'Appraisal and the resources of intersubjective stance' In the Appraisal Website. Retrieved May 2008 from www.grammatics.com/appraisal/. White, Peter R. R. (2009) An introduction to Appraisal, with the primary focus on Attitude. Workshop on Appraisal, Doshisha University 13 th June 2009. -17-

A.ppen d' IX monogloss r "a bare assertion"] 'She is a capable doctor.' contract disclaim: Deny-'no',, didn't', 'never' disclaim: Counter-'yet', 'although', 'but' proclaim concur affirm - 'naturally', 'of ENGAGEMENT course' (Resources we heterogloss concede - use to adjust and sure... [but] negotiate the arguability of our pronounce 'I contend utterances) that', 'indeed' AFFECT (People's emotions /feelings) How do you feel about it? endorse It shows /proves that..., entertain, e.g. 'This may/must', 'I expand think,' 'perhaps', 'in my view', 'apparentl ' Acknowledge: - 'He argues attribute that', 'They believe' Distance: - 'He claims to show' dis/inclination: (a feeling of intention that relates to future, as yet unrealized, states rather than present ones) 'miss', 'long for' 'demand' un/happiness How harjjjy did you {eel?-'sad' ~ in/security How secure didyoufeel?-'unsure' ~ ~ dis/satisfaction How satisfied did you feel?- fi'j 'interested', 'boredom' 'fed-up', 'fidget', 'hate'... fi'j social sanction propriety (ethics) How mora/?- - ~ ATTITUDE [involves praise 'good', 'just' 'evil' 'caring', 'fair' < fi'j "Attitudes have JUDGEMENT & veracity (truth) How believable?- to do with (People's character condemnation] 'truthful', 'honest', 'frank', 'fake', ~ ~ evaluating things, & behaviour) 'real' 'deceptive', 'bogus', ~ people's character "credible' < and their feelings" social esteem tenacity (resolve) Is s/he reliable, (Martin & Rose, [involves dependable?- 'brave' 'lazy' 'rash' 2003: 22) admiration & capacity How able?-'clever', APPRECIATION (The value of things) What do you think of it? criticism] 'weak' normality (fate) How special/ usual /destined?- 'lucky', 'odd', 'everyday' intensifiers e.g. 'really', 'very', 'merely' FORCE (Strengthen attitudinallexis e.g. 'fine', 'excellent', 'out-standing' / GRADUATION or weaken the 'nice', 'attractive', 'beautiful'. degree of comparatives e.g. 'better', 'best', 'worse', 'worst' evaluation) FOCUS sharpen (up-scale) e.g. 'a true friend' 'pure joy' soften (down-scale) e.g. 'sort of, 'kind of, 'a bit', Figure 2. Appraisal systems: an overview. Adapted from 'Appraisal systems: an overview' (Martin and Rose, 2003: 54 and White, P.R.R., 2003) reaction: quality [likeability] Did I like it?- 'ugly', 'lovely' reaction: impact Did it grab me?- 'dull', 'moving', 'exciting' 'boring', 'intense', 'engaging' composition:. complexity Was it hard to follow?- 'simple' 'detailed', 'simplistic' 'rich' composition: balance Did it go together? - 'uneven', 'shapely', 'logical', 'disorganized', 'unified', 'consistent' valuation [social significance] Was it worth-while?- 'challenging' 'profound', 'shallow' 'unique' 'deep' -18-