A sign of The Times : an APPRAISAL analysis of the imagined community’s bonding in letters to a South African newspaper

This article investigates how writers to The Times , formerly a daily national newspaper, used APPRAISAL strategies in letters to the editor to form the bonds that unified an imagined community of readership. The couplings of interpersonal and ideational meaning around which the community affiliated are revealed by an APPRAISAL analysis. Conclusions drawn from this information indicate the nature of the imagined community in terms of how the members viewed their agency, and the cohesion of the group. Main findings show that the community affiliated around evaluations of South Africa’s ruling party, the African National Congress (ANC), and its leader at the time, Jacob Zuma, but did so very differently. The Times ’ writers also tended to be individualistic and did not rely on other community members to solve the problems they identified.


Introduction
This article reports on the qualitative and quantitative combination of APPRAISAL and basic corpus analysis respectively to identify the couplings of interpersonal and ideational meaning that writers to The Times newspaper use to reinforce the bonds around which the community of readers affiliates. These bonds are a "sign" of what The Times readers are interested in and will rally around. Although the letters in the data set were published in 2012, when Jacob Zuma was the president of the African National Congress (ANC) and South Africa, the specifics of the data are secondary to the intricate deconstruction that a blended APPRAISAL and corpus analysis makes possible. Our research aims to answer the following questions: 1. How do APPRAISAL choices in letters to the editor construe the bonds that unify the community of readership of The Times, and how do the authors attempt to align the reader with the proposed bonds? 2. What is the nature of The Times' imagined community and how do the APPRAISAL strategies help to construct it?
Our approach to this research is one of Systemic Functional Linguistics, and we rely mainly on Martin and White's (2005) APPRAISAL framework to identify, codify, and interpret the meanings that the letter writers construe to their anticipated audience of fellow community members. Appraisal is concerned with evaluation, by which people express interpersonal meanings which are intended to align an audience. The alignment to an interpersonal evaluation creates a feeling of community and shared values and positions the audience in solidarity with the writer. In the case of The Times' readers, the subjects of their evaluations show in whom or what the community is interested, and the writers' choices of evaluations show how they feel about these people or things (and how they expect fellow readers in the "community" to feel).
While the APPRAISAL framework is sufficiently detailed to ensure a suitably nuanced coding of meaning, another of its strengths is its flexibility with regard to the context in which the research occurs. Consideration of both the context and co-text in meaning identification is vital, and, while it is never possible to be one hundred percent certain of the writers' purposes, we have carefully read the co-text and appreciate the context in order to code as congruently (Martin 1995) with the writers' intentions as possible.
Section 2 of this article provides context on the source of the data: The Times newspaper and its letters page. Section 3 outlines the concept of an imagined community which is represented on the letters page. Section 4 provides an overview of the systems in the APPRAISAL framework, using data from readers' letters to illustrate resources available to writers in the Attitude, Graduation, and Engagement systems. Section 5 further explains how interpersonal evaluations contribute to 'couplings' that construct bonds aligning members of the community. The data collection and the analysis methods used are summarised in Section 6, which also reveals which topics (ideational meanings) were most salient in the data, based on basic corpus analysis. Section 7 details the results of the APPRAISAL analysis, providing evidence of the interpersonal meanings which the writers couple with the ideational meanings to form the bonds around which the community affiliates. The writers use resources from the system of Attitude to evaluate their topics, but Graduation and Engagement strategies assist them in aligning the reader with their evaluation. The nature of the community is illustrated by means of key utterances in Section 8, and Section 9 concludes with what the analysis reveals about the imagined community of The Times readers. When quoting from letters in the data set, we refer to them by the initials of The Times, followed by a number from 1-20, which references the sequence in which the letters were collected. For example, a quote from TT12 comes from the twelfth letter we collected from The Times.

2.
The Times: The paper for the people

The Times' readership
The Times was a daily offshoot of South Africa's largest weekly newspaper, the Sunday Times.
Initially The Times was delivered on weekdays to Sunday Times subscribers, later becoming available commercially in limited quantities. In November 2017, after a decade of publishing, the decision to shut down the paper was publicised (Trench 2017). It was replaced with Times Select, a daily digital-only offering. In its printed version the newspaper was promoted as South Africa's first interactive newspaper, publishing its stories with added multimedia content on TimesLIVE, the online home of The Times and the Sunday Times. Scant information is available about the readership of The Times, but in 2012, the Wikipedia 1 page for The Times said the following: "Times LIVE caters to a loyal South African audience hungry for breaking news, quality journalism and the opportunity to engage and to be more informed" (Wikipedia entry for The Times [South Africa] 2012). Wasserman (2008:25) states that the Sunday Times is an example of "elite media", and is one of South Africa's most influential newspapers, so, by extension, The Times was arguably more of the same. Data from the South African Advertising Research Foundation (SAARF) All Media Products Survey (AMPS) at the time of the letters' publication showed that the majority of The Times' readers did fall into the upper LSMs, 2 especially LSM10, the highest category:

The letters page
To find the readers' 'voices' to give us a sign of their interests and values, we consulted the letters page, because, according to Conboy (2006), this is where the readers' identities can be found. Joe Latakgomo, then the public editor of the media company to which The Times and the Sunday Times belong, states that the letters to the editor of a newspaper are a "useful barometer of the newspaper's relationship with its readers" (Latakgomo 2012:6). They have also been described as the community in conversation with itself, he notes. Reading the letters page of a local newspaper in a foreign country, he adds, will give you a good idea of the community the newspaper serves. Although the figures for The Times were unavailable, Latakgomo (2012) was informed that the Sunday Times receives about one thousand letters per week. In response to criticism from a Sunday Times reader whose letter was not published, he reminded writers that it is not possible for every letter received by the newspaper to be published. In some cases, reveals Latakgomo (2012:6), the writers "even draw the conclusion 1 We would usually be sceptical about quoting Wikipedia; in this case, however, we believe that this entry was written by someone promoting the newspaper, so presumably the author would have got this quote from an editorial or managerial source. Since no equivalent information appeared on the newspaper's official website, we treat Wikipedia as the next most appropriate source. 2 After the abolition of apartheid, SAARF developed a new classification system that did not divide people in terms of race, as had previously been the case, but, instead, in terms of their living standards, introducing a number called the living standard measure (LSM). The scale ranges from 1-10, with 10 being the highest standard of living. The LSM is calculated based on "access to services and durables, and geographic indicators as determinants of standard of living" (SAARF 2010). 3 Percentages of total readers below LSM 7 were omitted as there were too few reader observations within these categories for data to be reliable. that their letter was not published because the editor does not like the views they are expressing". Readers must be reminded, he states, that publication of a letter is a privilege, not a right. Only in the event that a writer has directly been involved in something that has been reported on in the news, will the letter be prioritised. This is proved by a letter in the data (TT6) from a teacher who was present at the University of Johannesburg when a clamour for admissions resulted in the death of a parent. Latakgomo (2012) also explains the existence of "a broad policy that most newspaper editors subscribe to: letters have to be credible, and accurately reflect an issue. Also, the letters page forum should not be used for personal attacks.
If an attack has to be levelled at an individual, a politician, or even an institution, it should be based on verifiable fact -and must not be libellous" (Latakgomo 2012:6).
While The Times did not respond to requests for comment on the letters page production process in 2012, when the data set was compiled, it is still clear from the comments made by Latakgomo (2012) that letters are an important part of the newspapers' role in their communities, allowing readers to communicate with the newspaper, as well as publish their viewpoints to fellow members of the imagined community of The Times.

The imagined community
Although Anderson (1983:15) uses the term "imagined community" in the definition of a nation as an 'imagined national community', it has been used by Steenveld and Strelitz (2010), for example, as a term for audiences of media, and this is the understanding that we have. Despite this difference, Anderson's original reasons for the term are easily applied to newspapers. He calls the nation "imagined" because "the members of even the smallest nation will never know most of their fellow-members, meet them, or even hear of them, yet in the minds of each lives the image of their communion", and he calls it a "community" because its members are perceived as comrades, despite the fact that they occupy different social strata (Anderson 1983:15-16). It is highly unlikely that each member of the community will meet all of their fellow members, yet they have "complete confidence in their steady, anonymous, simultaneous activity", according to Anderson (1983:31). Similarly, individuals in newspapers' audiences are all engaged in the activity of reading their newspaper at some time during the day, and other readership community members know that they are doing this. Few of these people know the personal details of their fellow members, because the reading process is a private ritual. According to Gripsrud (2008:42), this ritual perspective considers the media as "modern society's communication with itself about itself" that inspires the feeling of community in its members. Following this, The Times community members would write letters to the newspaper about current events which concern its readers. To find out the concerns of the imagined community of The Times, the APPRAISAL framework is employed.

The APPRAISAL framework
According to Martin (2000), the APPRAISAL system consists of resources to evaluate the kinds of emotions that are negotiated in a text, the intensity of the feelings involved, and how readers are positioned in alignment with the stances advanced by the text producer. APPRAISAL uses three subsystems -Attitude, Graduation and Engagement (see Figure 1 below)to achieve this (the names of the subsystems are capitalised to prevent confusion with the non-technical use of the terms). In the sections that follow, subsystems of the APPRAISAL framework are defined to show our understanding of the system's resources and to give the reader an idea of how they are realised in the data. Only those subsystems realised in the data, however, are exemplified. We have used instantiations from the letters to The Times to illustrate the subsystems so that a localised, accurate perspective is provided, instead of relying on arbitrary international examples without context. This also gives the reader a sample of the kinds of evaluations and alignment strategies employed in the data, as well as how the writers feel about their topics.

Attitude
Attitude is the evaluation of feelings, people's behaviour, and things such as processes, texts, and natural phenomena (Martin and Rose 2007). Attitude comprises three subsystems: Affect, Judgement and Appreciation, which Martin and White (2005: 42) link to "emotion, ethics and aesthetics" (see Figure 2 below). Each of these options can be classified as positive [+] or negative [-]. Affect, which construes emotion, consists of another three subsystems: Happiness, Security and Satisfaction (Martin and White 2005). There are a number of reasons to be worried about the state of entrepreneurship… Satisfaction encompasses feelings about "the pursuit of goals… displeasure, curiosity, respect" (Martin and White 2005:49). An example of this appears in TT4 ([+Satisfaction]): … it is very encouraging to see parents involved in their children's education.
The resources of Judgement evaluate behaviour according to the writer's personal norms (Normality, Capacity and Tenacity) and legal norms (Veracity and Propriety). Normality evaluates whether someone is special in a way that others would want to emulate ([+Normality]) or possesses qualities that would be deemed strange ([-Normality]). Capacity involves a person's ability to do something -their "capability" (Martin 2000), as exemplified by the writer of TT3 ([+Capacity]): The ANC has exceptional ability to see us beyond today's limitations.
Whether the person or entity is reliable enough to carry out a task is covered by Tenacity. An example of [-Tenacity] appears in TT6: … the security personnel failed to control the situation.
Veracity deals with honesty, as shown in TT10 ([-Veracity]): … they then make false claims of pregnancy… Propriety assesses whether a person's behaviour is ethical. In the following example from TT1, the writer praises the ANC's behaviour ([+Propriety]): … it has played a leading role in helping to free the country.
Appreciation, the final subsystem of Attitude, is concerned with the value of 'things'. Reaction, the first type of Appreciation, construes how much the writer likes something (evaluation of quality) or how captivating it is (evaluation of impact), according to Martin and White (2005). The examples below are from TT2 and TT5 respectively.
[+Reaction] (impact): Its interesting findings include that entrepreneurial leaders are made… According to Martin and Rose (2007), the second type, Composition, indicates how logical, coherent and balanced something is, as shown by the instantiation of [-Composition] in TT6: I was shocked, knowing how chaotic this would be.
Valuation, the final type, describes how useful, worthy or significant the writer deems something to be (Martin 2000). The following example of [+Valuation] appears in TT4: Education is vital for personal growth and the development of a nation… Identifying the writers' evaluations of topics is the first step in understanding the nature and values of the imagined community. This is where the community 'shows its colours', so to speak, in that it declares what its views are on a topic, person, or entity and the editors choose to publish this letter. Attitude is thus the most overt strategy of alignment, showing the other community members that if they are part of the readership of this particular newspaper, their views should align with the letter writers'. This renews their affiliation with, or inclusion in, the imagined community of The Times readers.

Graduation
Graduation is more subtle than Attitude in its alignment function, upscaling or downscaling the intensity of an Attitudinal evaluation to make it more convincing. This system of resources, the second in the APPRAISAL framework, is used to modify meaning according to prototypicality, amount, or intensity. Writers can achieve this in two ways: Focus, which grades meaning in terms of prototypicality, as it applies to meanings which are not scalable (e.g. Doctors are real heroes); and Force, which grades evaluations according to their intensity (using the system of Intensification, e.g. I am extremely disappointed) and their amount (using the system of Quantification, e.g. Our debt is huge) (Martin and White 2005). The full system is represented in Figure 3 below. Focus can be sharpened or softened (and so more or less scaled), depending on how strong or weak the writer wishes the categorization to be (Martin 2004). The following example of Softening occurs in TT11, where the writer makes the term 'borderless' less precise: We live in an almost borderless economy… Intensification of evaluations (up or downscaling) can be done in terms of Quality and Process. Process refers to the vigour with which a process is carried out, whereas Quality is the degree to which a quality is achieved (Martin and White 2005). Intensification of Process is shown by the writer of TT12: … had it not purged from its ranks… Intensification of Quality (from TT8): … the decision to severely reprimand the CEO… The array of resources available in the Graduation system helps to align the reader with the writer's opinion by making it more convincing. To illustrate, in the example above where Extent: Distribution: Space is upscaled in TT1, the writer says that the ANC is the oldest liberation movement on the continent. As a continent is a large space which contains, assumedly, many liberation movements, the opinion that the ANC is good is upscaled, as it is compared with all of those on the continent. The reader is more likely to be convinced of the ANC's greatness, as it must be something special to have survived longer than the other liberation movements in all of Africa. However, the writer is not forceful in his or her convincing of the reader; the positioning is more subtle. The sentence is phrased in a factual, objective way, leaving less space for readers to argue with it, especially if they do not have a positive opinion of the ANC.

Engagement
Engagement is the final subsystem of the APPRAISAL framework. It consists of linguistic resources with which authors present a stance (i.e. their position on an issue) while positioning themselves regarding alternative evaluations and other voices in a text (Martin and White 2005). While doing this, they also attempt to position readers as aligning (or disaligning) with their stance. Engagement theory draws on Bakhtin (1981), who sees all written and spoken communication as dialogic, because it is always a response or reference to something that has been written or said before. The first distinction drawn in the Engagement system is between the monoglossic, or 'undialogised', and the heteroglossic, or 'dialogised' utterance (White 2003). Monoglossic utterances allow no space for alternative positions to be considered. Heteroglossia creates dialogic space which can be Expanded to allow for alternative stances or Contracted, which constrains the scope of voices or stances allowed (Martin and White 2005). See Figure 4 below for the systemic representation of these resources. Expansion is the subsystem which contains linguistic resources used to invite alternative stances, which construe an audience of diverse stances that could be in conflict with each other (White 2003). Expansion consists of Entertainment and Attribution. The former strategy construes the writer's stance as only one of a possible range of options (Martin and White 2005). Entertainment is most commonly realised by modality, as illustrated by the example from TT7 below: … US and European governments that may not even have natural resources… Earlier literature has treated these choices as an indication that writers lack commitment to the truth of their stance, but they should rather be seen as opening dialogic space to allow for alternate positions, argues White (2003). Attribution is the incorporation of an external voice by the writer, which transfers the responsibility for an evaluation onto the external voice (Martin and White 2005). Acknowledgement and Distance are two ways writers can use Attribution in a text. Acknowledgement is the most neutral-seeming way of presenting a stance using an external voice, as it often involves reported speech, with no judgement from the writer. In the following example from TT1, the writer quotes former French president Nicolas Sarkozy's praise of the ANC: It is a giant movement and an "international inspiration", as Nicolas Sarkozy described it.
Distancing strategies are used by writers to dissociate themselves from propositions made by an external voice (Martin and White 2005). These utterances are less neutral because writers evaluate the stance of the external voice. The writer of TT8, for example, shows his doubt that the assertions made by John Bester, the then chairman of Cricket South Africa's finance committee, are true: … and he claims the administration is in a good space … Writers can choose from a variety of resources from two subsystems to contract dialogic space for alternative positions: Disclamation and Proclamation. Disclamation closes down dialogic space by discrediting other stances, while authors that use Proclamation foreground their own stances instead. Strategies of Disclamation, Denial and Countering, overrule a proposition in the text (Martin and White 2005). Denial relies on negation such as no, never, not, and rejects a stance that has been introduced. In TT14, the writer contracts dialogic space for the counterargumentation using a Denial: We are not seeking "sexual empowerment in our lifetime"… Countering a stance is the second way to Disclaim it. Counters replace a dispreferred stance (a position that the author does not approve of) with one that the writer believes to be more justifiable, and often occur together with Denials. In the example from TT2 below, the writer follows a Denial of the expected stance (not) with a Counter (but) that shows the true "essence of [the police service's] approach": … the essence of our approach is not to mourn this behaviour, but to find solutions to a complex reality … The second subsystem of Contraction, Proclamation, has three options: Concurrence, Reinforcement (White and Don 2012), and Endorsement. Concurrence consists of Affirmation and Concession. When using the strategy of Affirmation, the writer assumes that the audience is already aligned, so the stance that is presented can be taken for granted. Affirmation can be realised by rhetorical questions, as well as instantiations such as the following from TT20: … corporate South Africa and the ANC naturally have to share with each other their views on the economy.
Concessions acknowledge propositions, and are often paired with Counters. They are strategies "by which argumentative ground is given up initially… only for that ground to be retaken in the subsequent counter move" (Martin and White 2005:124). The function of these is to say, 'yes, that is possible, but this is a more valid view to have'. In the following example from TT13, the writer admits that Zimbabwe has problems as a country (the reader is expected to agree with this), but states that the children of that country receive a good education: Concede Counter Children educated in Zimbabwe, [troubled though the country is], [still] have an outstanding education and can work anywhere in the world.
According to White and Don (2012), Reinforcement is the second subsystem of Proclamation, and consists of two options: Pronouncement, where the writer explicitly intervenes to declare the strength of the stance (Martin and White 2005); and Justification, where writers add extra information to explain their arguments (White 2003). Justification construes a disbelieving audience, because writers must believe that they need to show further evidence to prove their arguments. This is illustrated by the following example from TT5: There are a number of reasons to be worried about the state of entrepreneurship… For instance, when one looks at the Gini coefficient … it is clear that inequality is worsening.
Pronouncement is the foregrounding of authorial intervention in a text, either by mentioning themselves (with I or we) or with emphasis on something, where writers choose to obscure their subjectivity, as shown by the example from TT5 below: … it is clear that inequality is worsening.
In this example, as with Justification, Pronouncement construes an audience that is not aligned, as the author has felt the need to insist that the stance presented is true.
The authorial voice can also be foregrounded in Endorsement, but in this case it is used to assert that propositions of external sources are valid (Martin and White 2005;White 2003). Endorsements engage with the stance of the external source by referencing it and aligning with it, as shown by the example from TT18 below: Studies have confirmed that South Africa imports more than it can export.
Use of Disclaiming and Proclaiming strategies is a powerful method of alignment, though it can be subtle. Disclaiming tells the readers what they should not believe, while Proclaiming tells them what is the more desirable. Thus, Attitude, Graduation, and Engagement strategies operate simultaneously to nudge the reader towards alignment with the writer's stance. This interpersonal meaningthe evaluation and the strategies to align with itare attached or 'coupled' to a topic or subject (ideational meaning), and together they create a bond, which is explained further in the next section.

Affiliating around the interpersonal and the ideational
Identifying 'couplings' (Martin 2000) in a text reveals the bonds that bring a community together. These combinations (of interpersonal and ideational meaning in this article) show what the members of social networks affiliate around, and how they negotiate their community identity within a larger culture (Knight 2010). People bond around the feelings that they have in common, about experiences that they share, according to Martin (2004Martin ( , 2008. In terms of interpersonal meaning, these can be shared emotions (Affect), shared feelings about behaviour and the character of people involved (Judgement) and shared preferences (Appreciation) (Martin 2004). Even though the members of The Times' community of readership are spatially distanced from each other, they are united by similar reactions to situations about which they are all aware. The distinctive ways in which they couple their evaluations of their experiences negotiate an identity that separates them from other readership communities that share different couplings. The couplings found in this analysis could be called the 'DNA' of The Times readers. For example, the positive evaluation of the ANC in TT1 (see the end of 4.2) may not be shared by readers of a different newspaper, yet it is the sign of The Times reader. The community reads the letters, and when each reader reads a familiar evaluation (interpersonal meaning) about a specific topic (ideational meaning), the reader's affiliation with the group is confirmed by this coupling. If the evaluation does not conform to the reader's opinion, the letter coaxes the reader towards a shared 'community' opinion by using strategies of Graduation and Engagement.
The interpersonal component of the couplings which signify The Times letter writers' bonds are identified with an APPRAISAL analysis, the components of which were explained in Section 4. However, how does one know what the ideational component is? In other words, how does an analyst identify what the most popular topics for the writers are? The following section provides insight into the collection of the data and its analysis.

Analysis of the data
For this research, we chose the letter that was most prominently displayed on The Times' Opinion and Letters page each day. This was usually the letter that is accompanied by a photo, or the only letter that appears between the briefs (shortened letters) and the editorial cartoon.
As The Times did not appear on weekends, the 20 letters were collected every weekday for one month, over the period 9 January to 3 February 2012. As mentioned in Section 1, the period of collection could have been in any monththe focus of the research was more of a methodological one, to investigate the capacity of the APPRAISAL framework to provide empirical evidence of the imagined community at work. Using data from any period with such a methodology would allow for the investigation of the bonds that the community affiliates around. The number of letters collected was an effort to balance the need to have a corpus that was small enough to do an intensive APPRAISAL analysis and the need to have a corpus big enough that the couplings identified would be reliable. The titles of the letters are listed in Appendix A; the authors can be contacted for the full versions of these.
After identifying the instantiations of APPRAISAL in the data using the framework suggested by Martin and White (2005), we recorded them in tables according to the subsystems of Attitude, Graduation, and Engagement, based on the format derived by Marshall (2009) from Droga and Humphrey (2002) and Martin and White (2005). Once the data was coded, the first step to answering the research questions was to find what is important to the letter writers, that is, the ideational meanings with which their evaluations are most often coupled. We ran the data through a concordancer (AntConc3.2.4w) to generate frequency lists and keyword lists. Frequency lists show which lexical items appear most frequently in the data, and, according to Baker (2006), show analysts the points of focus of the corpus. A preliminary manual grouping of these results pointed to the following topics: politics and education. However, this list merely gives the number of times the words appear in the data set. To find out which words are most salient in a corpus, one list of words can be compared with another (Baker 2006). This will give a list of keywordsthose which are key to the text. To generate a list of keywords, a concordancer does a log-likelihood (LL) test, which generates a log-likelihood score. To achieve at least 95% confidence that the word does not occur due to chance, the LL score of a word should be over 3. 83 (McEnery et al. 2006). We decided to set the LL 'cut-off' score as 3.9, ensuring that any words with a score higher than this would be significant. In accordance with our preliminary identification of topics, the significant words in The Times were: movement (referring to the ANC -10.965), ANC (9.735), entrepreneurship (9.399) and entrepreneurs (6.266), UJ and university (9.399), schools (7.832), Zuma (7.179), Malema (7.049), economic (5.483), parliament (5.483), politics (4.699), student (4.699), academic (3.916) and centenary (of the ANC -3.916). Politics was significant in the letters, but we chose to separate ANC and Zuma instantiations, since they were mentioned separately in letters to The Times. Once we had identified these topics to be the ideational meaning that the readers bond around, the other half of the bondthe interpersonal meaninghad to be explored. The following two sections present the findings of our APPRAISAL analysis of the instantiations associated with the bonds that writers to The Times construe: the African National Congress and the South African president at the time, Jacob Zuma.

The bonds that unite The Times' community of readership
In this section, the APPRAISAL resources which the writers employ to evaluate the ANC and Zuma are identified, and the instantiations thereof are quoted so that the reader has evidence of the resources that have been activated by the writers. Much of the evaluative load is carried by choices of Attitude, which is expected, because the letter is written for the writer to express an opinion. The writers seldom express their feelings with Affect (e.g. "I am impressed by the ANC", for example, which would make it one person's opinion). Instead, monoglossic instantiations of Judgement and Appreciation are given, which allow little room for argument however, as part of a community, the writer might expect readers to hold similar views (align), so they should be taken as fact. There are also Contractive Engagement resources used to close down dialogic space, and some of those which are expansive (e.g. Acknowledgement) serve to make the writers' statements more factual or objective. The writers remark on the morals and behaviour of the ANC and Zuma (Judgement), as well as the value of the ANC (Appreciation of it as an entity). Where there is Graduation, it is upscaling, making the evaluations more intense. This can help to build solidarity in the community, celebrating what they value and criticising what they disapprove of, confirming their affiliation around these topics.

Bond 1: The African National Congress (ANC)
Evaluations of the ruling political party in South Africa occur in six letters in the data set, and show a strong tendency for The Times readers to couple the ANC with positive interpersonal meaning. There are 33 positive Attitudinal evaluations of the ANC, compared to 6 negative ones: this is 85% positive. The author of TT1, Thabile Mange, congratulates the ANC on the celebration of its centenary, calling it "worth celebrating" (monoglossically, contracting dialogic space), but argues later that there is still progress to be made in terms of "liberating the masses from the shackles of poverty", a [-Reaction] towards the circumstances of the 'masses'. By saying that the ANC is the "oldest liberation movement on the continent", Mange upscales the evaluation of [+Valuation] of the ANC twice. Other [+Valuations] of the ANC are "played a pivotal role" and "leading role".
[+Propriety] is invoked by Mange's statement that the ANC played a leading role in "helping to free the country". The ANC is also regarded as an "international inspiration" (an upscaled [+Reaction] Acknowledged from former French president Nicolas Sarkozy -a strategy of Expansion that supports the writer's argument, as a well-known political figure is quoted). Although Mange evaluates the ANC monoglossically as displaying [+Tenacity], having "survived the storms", the same line contains an (also monoglossic) evaluation of [-Tenacity], where Mange says that "Zuma's party" has not delivered on its promises. In the next line, however, the party is referred to as the "ANC" again, not as "Zuma's party", and Mange's statement that the "masses still have confidence… that it will take them to the promised land" invokes an evaluation of [+Capacity] of the ANC. This [+Capacity] is upscaled by still, which shows that despite its shortcomings, the ANC's followers remain loyal. The focus of TT3 is also the ANC's centenary celebrations, but the author, Zukiswa Mqolomba, places more of a focus on Zuma's speech. Mqolomba evaluates the ANC as "glorious" ([+Reaction]), and praises the party for its "exceptional ability (upscaled [+Capacity]) to see us beyond today's limitations". The author also invokes upscaled [+Valuation] of the ANC, arguing monoglossically, without space for denial, that "South Africa needs the ANC like the body needs oxygen". The idea that the ANC has evolved, however, is Entertained, where Mqolomba expands dialogic space for alternative views by calling the party's trajectory "(D)evolution".
TT12 takes the stance that ANC Youth League president, Julius Malema, should be expelled from the party. 4 This must be done, according to the author, Nhlanhla Molapo, so that the party can continue to create "internal unity and cohesion" ([+Composition]). Molapo's evaluations of the ANC are of [+Tenacity]: "emerged from difficult ([-Reaction]) exile years to lead a victorious struggle against apartheid", "reached the 100-year milestone" (upscaled), and "succeeded in creating internal unity". The ANC's behaviour is seen as [+Propriety] ("does not throw its own members into the dustbin of history"), and this quote is Acknowledged as coming from Mathews Phosa. 5 However, it is also Countered with "Contrary to Mathews Phosa's assertion that the ANC does not throw its own members… the ANC, in fact, expelled the entire group of eight [dissenters]". This shows the author's belief that, although the ANC says they will not expel their own members, when members "contradict and attack key ANC policies" ([-Propriety]), it is justification for expulsion.
Cynthia Tsabedze, author of TT16, only mentions the ANC briefly, but she couples her few mentions of the party with positive evaluations. She calls the ANC "an important ally" ([+Valuation]), and says "It is a known fact" that this is so. This Pronouncement removes the author's subjectivity from the claim, making it appear more 'factual', and in so doing contracts the dialogic space for alternative stances. Readers are thus more likely to believe it and align with it because it is not presented as someone's opinion. The ruling party is also evaluated as having [+Capacity] to "help in the fight for democracy" in Swaziland. 6 TT17 is similar to TT16 in that the ANC as a party is not mentioned often, but in this case the members of the party are evaluated. The author, Thabile Mange, says that the ANC is "divided over Zuma and Malema" ([-Composition]). The other evaluation of the ANC, however, is positive: a monoglossic [+Valuation] of the party, that is upscaled twice ("oldest liberation movement on the continent"). This [+Valuation] is supported by the writer's desire that the party remain free from "politics of personalities", as this has the "potential to destroy (upscaled [-Reaction])" the ANC, and Mange argues that this is "not in the best interest of the party or the country".
TT20 is a response to a columnist's critique of the ANC's ethics. The columnist suggests that business functions are organised by the Progressive Business Forum (PBF) so that its members can liaise with ANC leaders, in order to secure business contracts with the government. 7 The author of the letter, Daryl Swanepoel (co-convenor of the PBF), argues that the columnist makes it sound "as if the ANC has stumbled across a fountain of cash, which it is determined to exploit, regardless of ethics". The columnist, therefore, is evaluating the ANC's behaviour twice as [-Propriety]. Swanepoel, in contrast, Denies that there is anything "sinister" about the functions, and uses three Engagement strategies in one line to close down dialogic space for any stances that would be aligned with the columnist: Swanepoel says, "The truth (Pronouncement), however (Counter), is that there is nothing (Denial) sinister to it. This extreme contraction of space for alternate stances is supported by Swanepoel's numerous [-Valuations] of the columnist's arguments: "rubbished", "Hardly considerable sums", "Let's get real", "inaccuracy", and "match your diligence in criticism with diligence in putting forward ideas".

Bond 2: President Jacob Zuma
Unlike those of the ANC, 100% the evaluations of Zuma are negativea sign that The Times readers affiliate around a dislike for the president at that time. After Thabile Mange, author of TT1, praises the ANC for reaching its centenary, the focus changes to Zuma. Mange differentiates between past and present ANC leadership, asking, "has Zuma's party delivered on its promises? The answer is a "resounding "no"". This Affirmation implies that the ANC's success so far has not been continued by Zuma, and highlights the evaluation of upscaled [-Tenacity] that is construed by the monoglossic words "resounding "no"", which disallows rebuttals from Zuma supporters. The writer upscales the amount of work that needs to be done by "Zuma's party" ("A lot") in order to "liberate the masses from the shackles of poverty ([-Reaction] to the circumstances of 'the masses')". TT3, also with a focus on the ANC's centenary, continues with the negative evaluation of Zuma, as well as his speech at the ANC centenary celebration, which the author sees as a reflection of the man. The author, Zukiswa Mqolomba, evaluates Zuma as having [-Capacity] to inspire "confidence in the government's political and strategic programme in response to the nation's aspirations": his speech showed him to be "in a boat without a rudder", and was full of "pauses", "misplaced full stops", "broken single words" and that four years (upscaled Extent: Distribution: Time) in office had not sharpened "the mind". Zuma's speech also causes the writer [-Satisfaction] ("lament", "misgivings"). Mqolomba argues, with no space for other viewpoints, that the speech "lacked thematic focus" and "lacked… a common thread that bound it together" (both [-Composition]). Further negative Appreciations of Zuma's speech include [-Reaction] ("lacked… an underlying message") and [-Valuation] ("useful to those oblivious of the activities", "devoid of political analysis or actionable historic lessons", "shallow", and the Affirmation "What have we done… to deserve an address unbefitting the occasion?", which takes for granted that the speech did not befit the occasion, positioning the reader as aligned with this stance  [-Propriety], making the argument more convincing and therefore more likely to be aligned with.

The Times' imagined community
The nature of the newspaper's imagined community refers to the cohesion within the group and their approach to problem solving. Who is responsible for solving these problems, and how can the community members get this done? The most salient link between community members is that they are all part of an 'in-group', an 'us', as opposed to 'them'. The frequency lists generated by the concordancer show that letters to The Times have we and our listed as the 19 th and 37 th most frequently used words, respectively. This indicates that the community members see themselves as individualistic, and do not see the need to show explicitly who is a member of the community and who is not.
Writers to The Times add to the impression of community discreteness by not addressing their fellow community members often, rather appealing to people or organisations, for example the ANC (to expel Malema in TT12), Zuma (what he should have said in his speech in TT3) and people who have control over teachers' salaries (how to adjust salaries in TT15). They suggest solutions to problems they see, but The Times writers do not advise community members, as it is not with them that they have a problem. The problems are abstract (for example, entrepreneurship, corruption, African resources benefiting Africa, the arrangement of the education system), and so finding solutions to these problems does not always involve the effort on the community's part. The Times community responds readily to events in the news, such as the mother's death at UJ, the 'bonus issue' amongst executives of Cricket SA, the ANC centenary, Enoch Godongwana's resignation, and the controversial multiracial poster from the political opposition's youth branch, the Democratic Alliance Students' Organisation (DASO). The Times writers seem to prefer offering a commentary on the situation at hand, to add their views to debates, but they also offer solutions to the problems they identify. Table 2 below summarises key utterances from the newspaper to illustrate further The Times' solutionorientated community. … the essence of our approach is not to mourn this behaviour, but to find solutions to a complex reality … TT5 We need to change mindsets and this starts with dispelling myths about entrepreneurship. TT7 As a country we must support such protests because it is our dream that African riches benefit Africans. TT8 A new start is required with the new executive putting in place checks and balances … TT11 Collaborative partnerships… could possibly provide the contextual milieu within which a transformed national education arrangement can occur. TT12 This is a good time for the ANC to rid itself of opportunists and liquidators like Julius Malema.

TT15
What is required is a curriculum with a core that is based on the 3Rs, plus a narrow selection of secondary subjects. TT18 … to prosper we need to become more innovative in our business dealings.
From Table 2, it is clear that The Times writers appeal to Appreciations, i.e. entities that can provide a solution to more abstract problems (for example, changing the curriculum, or "putting in place checks and balances" -TT8). The Times readers are independent, not relying on other community members to make a change.

Conclusion
Wasserman, H. 2008. Appendix A: Titles of letters to The Times, 9 January-3 February 2012 TT1: Happy centenary, ANC, but there's still lots to do TT2: Police suicides affect us all TT3: Zuma's speech did not take us beyond 100 years TT4: UJ has failed the disaster management test dismally TT5: Much more must be done to create entrepreneurs TT6: Greed was at the root of UJ's chaos, and tragedy TT7: Africa must unite to take back riches TT8: Entire CSA executive is at the root of the debacle TT9: No tears for Enoch after his resignation TT10: Culture, stigma drive women to kidnap babies TT11: Partnerships are key to transforming education TT12: The ANC must expel the 'petty-bourgeois' Malema TT13: Children don't master new languages at school TT14: DA youth poster reckless TT15: How to grade children and reward their teachers TT16: Rail deal means little to Swaziland's ordinary folk TT17: Politics of personality strangling ANC to death TT18: Innovation should be the name of Africa's game TT19: 'The people' angry at info bill consultation TT20: PBF's role in state-led delegations above board