1.1Trump’s negativity on Twitter

Scholarly studies in media and communication have shown that there is a correlation between heavy Twitter use and negatively-charged emotional language aimed at garnering the attention of online audiences (Stieglitz and Dang-Yuan 2013Stieglitz, Stefan, and Linh Dang-Xuan 2013 “Emotions and Information Diffusion in Social Media – Sentiment of Microblogs and Sharing Behavior.” Journal of Management Information Systems 29 (4): 217–248. ; Ott 2017Ott, Brian L. 2017 “The Age of Twitter: Donald J. Trump and the Politics of Debasement.” Critical Studies in Media Communication 34 (1): 59–68. ). ‘Negativity’ is regularly observed in political discourse where it tends to focus on the alleged faults and weaknesses of opposition candidates (Dolezal et al. 2017Dolezal, Martin, Laurenz Ennser-Jedenastik, and Wolfgang C. Müller 2017 “Who Will Attack the Competitors? How Political Parties Resolve Strategic and Collective Action Dilemmas in Negative Campaigning.” Party Politics 23 (6) 666–679. ), or on negative emotions and opinions toward people and events (Ott 2017Ott, Brian L. 2017 “The Age of Twitter: Donald J. Trump and the Politics of Debasement.” Critical Studies in Media Communication 34 (1): 59–68. ). This study explores the linguistic realisation of such negativity by considering how it is realised in negative evaluative language using the Appraisal framework, a social semiotic model of evaluation (Martin and White 2005Martin, J. R., and P. R. R. White 2005 The Language of Evaluation: Appraisal in English. Basingstoke: Palgrave/Macmillan. ; Hunston 2000Hunston, Susan 2000 “Evaluation and the Planes of Discourse: Status and Value in Persuasive Texts.” In Evaluation in Text: Authorial Stance and the Construction of Discourse, edited by Hunston, Susan and Geoff Thomson: 176–207. London: Oxford University Press.). Twitter seems to privilege negative communication which can breed “dark, degrading, de-humanizing discourse” (Ott 2017Ott, Brian L. 2017 “The Age of Twitter: Donald J. Trump and the Politics of Debasement.” Critical Studies in Media Communication 34 (1): 59–68. , 62). There have been several studies focusing on the negativity of Trump’s tweeting practice (c.f., Ross and Caldwell 2020Ross, Andrew S., and David Caldwell 2020 “ ‘Going Negative’: An APPRAISAL Analysis of the Rhetoric of Donald Trump on Twitter.” Language & Communication 70: 13–27. ; Ceron and d’Adda 2016Ceron, Andrea, and Giovanna d’Adda 2016 “E-campaigning on Twitter: The Effectiveness of Distributive Promises and Negative Campaign in the 2013 Italian Election.” New Media & Society 18 (9): 1935–1955. ). Lee & Quealy (2016)Lee, Jasmine C., and Kevin Quealy 2016 “The 282 People, Places and Things Donald Trump has Insulted on Twitter: A Complete List.” The New York Times 25. have noted the unprecedented degree of personal attacks and insults that characterised Trump’s election campaign. After analysing 66,463 tweets from the primary season leading up to the 2016 election, Conway-Silva et al. (2018)Conway-Silva, Bethany A., Christine R. Filer, Kate Kenski, and Eric Tsetsi 2018 “Reassessing Twitter’s Agenda-Building Power: An Analysis of Intermedia Agenda-Setting Effects During the 2016 Presidential Primary Season.” Social Science Computer Review 36 (4): 469–483. also concluded that Trump produced more lying accusations than any other candidate. Krugman (2016)Krugman, Paul 2016 “Donald Trump’s “Big Liar” Technique.” The New York Times. labelled Trump’s online behaviour as the ‘big liar’ technique, arguing that the accuracy and truthfulness of most of Trump’s tweets have become irrelevant to their discursive power and influence. Also pervasive in Trump’s discourse is repeated derogatory labelling of mainstream media as “fake news”. This is a rhetorical strategy aimed at spreading misinformation by delegitimising news outlets (e.g. CNN and the New York Times) while indicating allegiance to outlets that support Trump’s agenda (e.g. Fox News) (Ross and River 2018Ross, Andrew S., and Damian J. Rivers 2018 “Discursive Deflection: Accusation of “Fake News” and the Spread of Mis-and Disinformation in the Tweets of President Trump.” Social Media+ Society 4 (2): ).

This paper explores Trump’s negativity on Twitter using the Appraisal framework (Martin and White 2005Martin, J. R., and P. R. R. White 2005 The Language of Evaluation: Appraisal in English. Basingstoke: Palgrave/Macmillan. ), to understand how patterns of evaluation are targeted at certain groups and are involved in creating alignments with the ambient audience. While there has been a limited concentration of work applying Appraisal to Trump’s tweets, there is less work considering his political discourse from the perspective of affiliation (although see Zappavigna (2018)Zappavigna, Michele 2018 Searchable Talk: Hashtags and Social Media Metadiscourse. London: Bloomsbury Publishing.). Ross and Caldwell (2020Ross, Andrew S., and David Caldwell 2020 “ ‘Going Negative’: An APPRAISAL Analysis of the Rhetoric of Donald Trump on Twitter.” Language & Communication 70: 13–27. , 14) analysed a corpus of Trump’s tweets around the time of his political campaign and subsequent victory, using Appraisal to reveal his tendency to “go negative” in attempts to delegitimise his political opponent, Hillary Clinton. Another study drawing on Appraisal found that Trump made use of negative evaluation and judgement much more frequently than Clinton (Hoffman 2018Hoffmann, Christian R. 2018 “Crooked Hillary and Dumb Trump.” Internet Pragmatics 1 (1): 55–87. ). Other work has noted that Trump tends to communicate informally, directly, and provocatively, in his attempts to appeal to supporters by constructing an image of a “homeland” threatened by dangerous “others” (Kreis 2017Kreis, Ramona 2017 “The “Tweet Politics” of President Trump.” Journal of Language and Politics 16 (4): 607–618. ). He tends to use personal pronouns such as I, we, and membership terms such as nation and people to construe an ingroup of nationalistic Americans. Outsiders are referred to with pronouns such as them and they, for example in tweets about ‘Muslim bans’. Kreis concluded that this may contribute to the normalisation of right-wing populist discourses.

The present study will consider the kinds of affiliation strategies adopted by Trump that may contribute to this kind of in/out grouping as Trump attempts to create a community of shared values that excludes outsiders. It builds on previous work on social bonding via ‘ambient affiliation’ in social media discourse (Zappavigna 2014Zappavigna, Michele 2014 Coffeetweets: Bonding Around the Bean on Twitter. In The Language of Social Media, edited by Seargeant, Philip and Caroline Tagg: 139–160. London: Palgrave Macmillan. , 2018Zappavigna, Michele 2018 Searchable Talk: Hashtags and Social Media Metadiscourse. London: Bloomsbury Publishing.; Zappavigna and Martin 2018Zappavigna, Michele, and James R. Martin 2018 “# Communing Affiliation: Social Tagging as a Resource for Aligning Around Values in Social Media.” Discourse, Context & Media 22: 4–12. ) to demonstrate the ways in which negatively appraising an ‘other’ can be used to foster social alignments that have serious and negative implications for particular social groups.

1.2Background on Iran-US relations

There is a long history of hostility between the United States and Iran, dating back to 1953 when the United States and the UK were behind a coup d’état which led to the toppling of the Iranian Prime Minister, Mohammad Mosadegh (Pollack 2004Pollack, Kenneth 2004 The Persian Puzzle: Deciphering the Twenty-five-Year Conflict Between the United States and Iran. New York: Random House.). While the two countries have never been officially at war, they have not had any diplomatic ties since November 1979 and the American hostage crisis. In the hostage crisis standoff, some Iranian university students, angry at US support of the dethroning of the Shah of Iran, stormed the US embassy and took 60 US citizens hostage, with the permission of the Supreme Leader of the time, Ayatollah Khomeini. This led to an international crisis with US President Carter cutting all diplomatic ties between the two countries and imposing sanctions against Iran (Blakemore, 2020Blakemore, Erin 8 January 2020 “U.S.-Iran Tensions: From Political Coup to Hostage Crisis to Drone Strikes.” https://​www​.history​.com​/news​/iran​-nuclear​-deal​-sanctions​-facts​-hostage​-crisis. Accessed 9 January 2020. https://​www​.history​.com​/news​/iran​-nuclear​-deal​-sanctions​-facts​-hostage​-crisis). The closest step to any official arbitration was the 2015 nuclear negotiations during which former US Secretary of State, John Kerry, and Iran’s Minister of Foreign Affairs, Mr Javad Zarif met on a few occasions. Together with leaders of China, France, Russia, the UK, and Germany, they signed a Joint Comprehensive Plan of Action (JCPOA) in order for Iran to suspend its nuclear activity in exchange for sanction relief from the US and other European countries. However, in May 2018, Trump unilaterally withdrew from the agreement and vowed harsh penalties for companies trading with Iran. Following this decision, tensions between the two countries escalated and led to instability in the Middle East region.

The US has always criticised Iran for its financial and military support of militia-based groups, as it considers them a source of volatility and insecurity in the region. The two countries were on the brink of a full-scale war in early 2020 when the Pentagon confirmed the killing of a high status Iranian General, Qasem Suleimani, Commander of Quds Force, in Baghdad airport. Iran fired more than a dozen missiles toward US military bases in Baghdad, in response. A more tempered rhetoric and the possibility of mutual dialogue between the two countries has largely been annihilated, and the two countries continue to blame each other for all instability in the region. Iranian Supreme Leader, Ayatollah Khamenei, has ruled out the possibility of any dialogue between the US and Iran after the killing of General Suleimani. This is indicative of the high level of hostility between the US and Iran at the present time, an extremely dangerous context for the kind of negatively-charged discourse produced by Trump about Iran that we will explore in this paper.

1.3The structure of this article

This paper begins by introducing the Appraisal framework (Martin and White 2005Martin, J. R., and P. R. R. White 2005 The Language of Evaluation: Appraisal in English. Basingstoke: Palgrave/Macmillan. ), a framework for analysing the patterning of evaluative language developed within Systemic Functional Linguistics. It then details the model of ambient affiliation (dialogic and communing affiliation), explaining how this model arose out of Knight’s (2010)Knight, Naomi 2010 “Laughing Our Bonds off: Conversational Humor in Relation to Afiliation” PhD Thesis, University of Sydney, Sydney, Australia. work on conversational exchanges. Affiliation considers how social alignments arise out of attitudinal stances targeted at things and experiences, referred to as ideation-attitude couplings. The paper then details the corpus of Trump’s tweets about Iran and explores the specific affiliation strategies deployed by Trump to align or de-align from certain stances and voices, and to generate solidarity or dissonance with respect to certain groups

2.1The Appraisal framework

Martin and White’s (2005)Martin, J. R., and P. R. R. White 2005 The Language of Evaluation: Appraisal in English. Basingstoke: Palgrave/Macmillan. Appraisal framework offers an account of evaluative language aimed at describing three systems of meaning, attitude, Engagement, and Graduation,11.All appraisal terms and sub-types are written in small caps. that are used in this paper to explore the attitudinal positions in the corpus of Trump’s tweets. attitude is a system for mapping how feelings and opinions are construed in texts and is comprised of three regions of meaning: affect (expressing emotions e.g. love, hate, fear), judgement (assessment of human behaviour based on social and ethical norms e.g. competent, stupid) and appreciation (valuing of things and events e.g. beautiful). Attitudes can also be graded. The graduation system models the up-scaling and down-scaling of an evaluation in terms of intensity and amount (force) or prototypicality and preciseness (focus) (Martin and White 2005Martin, J. R., and P. R. R. White 2005 The Language of Evaluation: Appraisal in English. Basingstoke: Palgrave/Macmillan. , 135–160). Another dimension of Appraisal is the engagement 22.The main distinction in engagement is between utterances which engage with dialogic alternatives (heterogloss) and those which do not (monogloss). monoglossic statements do not acknowledge any other positions while heteroglossic statements whether entertain other propositions and viewpoints (dialogically expansive) or challenge/reject alternative viewpoints (dialogically contractive). system, influenced by Bakhtin’s (1981)Bakhtin, Mikhail M. 1981 “The Dialogic Imagination: Four Essays, ed.” Michael Holquist, trans. Caryl Emerson and Michael Holquist (Austin: University of Texas Press, 1981) 84 (8): 80–2. notion of “dialogism”, which deals with intersubjective positioning and whether a proposition includes or excludes other voices and alternative viewpoints (Martin and White 2005Martin, J. R., and P. R. R. White 2005 The Language of Evaluation: Appraisal in English. Basingstoke: Palgrave/Macmillan. , 92–134). If a proposition entertains other value positions and voices, then it is “dialogically-expansive” but if it actively challenges or constrains the scope of other alternative positions, the proposition is “dialogically-contractive” (Martin and White 2005Martin, J. R., and P. R. R. White 2005 The Language of Evaluation: Appraisal in English. Basingstoke: Palgrave/Macmillan. , 102). Figure 1 below shows the Appraisal framework with examples from Trump’s tweets about Iran to illustrate the most delicate choice in each system, with Appraisal features shown in bold font.

Appraisal also accounts for whether or not the evaluative meaning is realised in the text through explicit evaluative language, that is, whether it is inscribed, or through less direct means, that is, whether it is invoked. Appraisal distinguishes between three systems of implied evaluation: provoke, flag and afford (Hood 2010Hood, Susan 2010 Appraising Research: Evaluation in Academic Writing. London: Springer. ). For example, attitude may be provoked via lexical metaphors or flagged via graduation resources. Finally, attitude may be ‘afforded’ via ideation; for instance when something can be said to have shared significance. An example is the Whitehouse which is generally a shared positive symbol in American culture, although which it may have positive or negative connotations depending on the ideological position of the author/speaker relative to the political party currently in government.

2.2Dialogic affiliation

The Appraisal framework provides insight into attitudinal meaning and how social values are constructed in texts. However, sociality is more complex than simply expressing feelings or opinions: “[we] don’t after all simply affiliate with feelings; we affiliate with feelings about people, places and things, and the activities they participate in, however abstract or concrete” (Martin 2008Martin, James R. 2008 “Innocence: Realisation, Instantiation and Individuation in a Botswanan Town.” Questioning linguistics: 27–54., 58). Thus, the key unit of analysis for exploring affiliation is the ideation-attitude coupling; how feelings or opinions are associated with experience in texts, or in other words, how ideational and interpersonal meanings are related in discourse to form a value that is available to be negotiated. Knight’s (2010Knight, Naomi 2010 “Laughing Our Bonds off: Conversational Humor in Relation to Afiliation” PhD Thesis, University of Sydney, Sydney, Australia., 2013Knight, Naomi K. 2013 “Evaluating Experience in Funny Ways: How Friends Bond Through Conversational Hum.” Text & Talk 33 (4–5): 553–574. ) work on the role of ideation-attitude coupling in the negotiation of bonds in conversational exchanges forms the basis of the social semiotic approach to affiliation adopted in this paper. This work suggested that in dialogic exchanges, interactants tend to rally around, condemn, or defer social bonds

An example of an ideation-attitude coupling is shown below in a tweet where Trump criticises Hillary Clinton via negative judgement (incompetent), creating a disdainful tone as a disqualification strategy:

This tweet features a coupling of Secretary of State, Hillary Clinton and negative judgement, which is notated, following the convention adopted by Zappavigna and Martin (2018)Zappavigna, Michele, and James R. Martin 2018 “# Communing Affiliation: Social Tagging as a Resource for Aligning Around Values in Social Media.” Discourse, Context & Media 22: 4–12. , with ideation shown in underlined italics and attitude in bold. The square brackets and / are used to suggest the fusion of attitude with ideation, forming a value that is open to potential response by other tweeters. This coupling offers to the ambient audience ‘bad Hillary’ as a potential bond which they can choose to embrace, challenge or laugh off, depending on their political orientation or some other variable in the context. In this example inscribed (explicit) negative attitude has been coupled with Hillary Clinton. However, one might argue that providing the title of “Secretary of State” in this context also invokes negative judgement of capacity targeted at Clinton, in the sense that Trump seems to use the title sarcastically or with a degree of disdain. In our data analysis we have focused on inscribed couplings since these have clear realisations through the system of attitude and have only considered invoked attitude where it is very clear from evidence present in the co-text. Most tweets contain attitudinal meanings that are invoked via various strategies, however justifying their annotation usually requires abducing contextual parameters that are not always available in ambient context such as Twitter (see Section 2.1 for the types of invoked attitude that are considered in Appraisal).

This tweet about Clinton does not directly address anyone as part of an explicit conversational exchange. Instead, like many tweets, it has a potential “imagined audience” (see Marwick and boyd 2011Marwick, Alice E., and Danah Boyd 2011 “I Tweet Honestly, I Tweet Passionately: Twitter Users, Context Collapse, and the Imagined Audience.” New Media & Society 13 (1): 114–133. ; Litt and Hargittai 2016Litt, Eden, and Eszter Hargittai 2016 “The Imagined Audience on Social Network Sites.” Social Media+ Society 2 (1): 1–16. ) that may choose to engage with the tweet or simply read it. This audience might include Trump’s followers or general Twitter users who search for material, browse hashtags or keywords, or are directed to the tweet from a news site, amongst many other possibilities. The tweet received 9.1 thousand replies such as the following examples which both table couplings that are critical of Trump’s stance on Clinton:

The replies above both de-align with the coupling in Trump’s tweet by challenging it with competing couplings targeting negative judgement appreciation at Trump and his verbiage. The unfolding thread of replies that criticise Trump with couplings of this kind can be interpreted as mass rallying around a ‘bad Trump’ bond, tabled as a challenge to the ‘bad Hillary’ bond that Trump asserted in his original tweet.

2.3Communing affiliation

While Twitter discourse can involve direct interaction between users, interpretable as rallying around, rejecting or deferring social bonds using Knight’s dialogic affiliation framework, often a tweet will receive no reply. Such tweets can nevertheless be seen to be offering potential bonds to the ambient audience (Zappavigna and Martin 2018Zappavigna, Michele, and James R. Martin 2018 “# Communing Affiliation: Social Tagging as a Resource for Aligning Around Values in Social Media.” Discourse, Context & Media 22: 4–12. ). This is particularly the case with Trump’s tweets, where there is usually a cascade of replies but rarely any exchange between Trump and any subsequent replies. In order to account for ambient social bonding in the absence of direct interaction, Zappavigna and Martin (2018)Zappavigna, Michele, and James R. Martin 2018 “# Communing Affiliation: Social Tagging as a Resource for Aligning Around Values in Social Media.” Discourse, Context & Media 22: 4–12. and Zappavigna (2021)Zappavigna, Michele 2021 “Ambient Affiliation in Comments on YouTube Videos: Communing Around Values About ASMR.” 外国语 44, no. 1: 21–40. developed the system of communing affiliation. Communing affiliation (Figure 2) involves three strategies whereby an ideation-attitude coupling tabled in a post is made more bondable:

• convoking:44.Communing affiliation strategies are shown in small caps throughout; convoking, finessing and promoting. mustering community around a bond

• finessing: dialogic positioning amongst other potential bonds

• promoting: enhancing discursive visibility of a bond

These strategies are concerned with how an orientation towards a bond is established in a text; that is, how a particular ideation-attitude coupling is directed to a community, how it is set against other potential couplings offered by other potential voices in the social stream, and how it is made visible or more prominent in the social stream. Since Trump largely ignores replies to his tweets, communing affiliation offers a means of understanding the kinds of strategies he is using to build potential alignments with his audience without necessarily directly engaging with them.

Figure 2.System of Communing affiliation adapted from Zappavigna (2021)Zappavigna, Michele 2021 “Ambient Affiliation in Comments on YouTube Videos: Communing Around Values About ASMR.” 外国语 44, no. 1: 21–40.

Convoking is a system of meaning-making considering the resources through which a text ‘calls together’ a community or group to bond around a coupling. convocation encompasses choices in systems such as address (e.g. vocatives) which can be used to marshal a persona or group to align around a value. It can also include naming resources which can be used to designate the community to whom a value is relevant. For example, the following tweet attempts to convoke the audience, marshalled as ‘we’ around the coupling [ideation: Iran Deal / attitude: negative appreciation ] tabled at the beginning of the post:

Finessing a coupling refers to the modulation of a coupling with reference to other stances and voices which may be present. This occurs through either embellishing a coupling via discursive resources such as dialogic expansion or distilling it via resources such as dialogic contraction. For example, embellishing would open the bond to various other possibilities offered by a range of voices with differing perspectives (e.g. I guess, I think, it seems, maybe). In contrast distilling would limit this range, often to only one choice, for instance through propositions that assert bald facts with monoglossic language. The relevant system of meaning at stake is the Appraisal system of Engagement (Martin and White 2005Martin, J. R., and P. R. R. White 2005 The Language of Evaluation: Appraisal in English. Basingstoke: Palgrave/Macmillan. ) which accounts for the extent to which a proposition entertains alternative viewpoints and stances, and is either dialogically expansive or dialogically contractive in accommodating or rejecting other viewpoints. For example, the following tweet distils the coupling [ideation: paying billions to Iran / attitude: negative judgement ] through resources of contraction and modality that close down other potential perspectives on this coupling:

Finally, promoting a coupling refers to interpersonally emphasising the coupling in order to attract attention in the social stream. It performs a similar function to upscaling or intensifying attitude through the system of graduation (Martin and White 2005Martin, J. R., and P. R. R. White 2005 The Language of Evaluation: Appraisal in English. Basingstoke: Palgrave/Macmillan. ). This choice is very frequent in Trump’s tweets as he used ALL CAPS and exclamation marks very frequently in his tweeting practice (see Ross and Caldwell 2020Ross, Andrew S., and David Caldwell 2020 “ ‘Going Negative’: An APPRAISAL Analysis of the Rhetoric of Donald Trump on Twitter.” Language & Communication 70: 13–27. ; Wignell et al 2019Wignell, Peter, Kay O’Halloran, and Sabine Tan 2019 “Semiotic Space Invasion: The Case of Donald Trump’s US Presidential Campaign.” Semiotica 2019 (226): 185–208. ). For example, the following tweet fosters a series of couplings negatively judging the Iranian president through this type of typographic choice together with upscaled graduation (shown in blue and underlined):

The ALL CAPS font and the upscaled graduation both emphasise the expressed attitude both through intensifiers as well as infused emphasis via Appraisal resources themselves (e.g. the choice of ‘threaten’ rather than less evaluatively loaded lexis such as ‘pressure’). Fostering typically draws upon graduation resources for upscaling or downscaling attitude (force), whereas the alternative choice, modulate, draws upon graduation resources relating to adjusting prototypicality or preciseness (focus, e.g. a true hero). As the system network in Figure 2 shows with the brace (representing an ‘and’ relation), the three choices of convoking, finessing, and promoting are simultaneous rather than mutually exclusive, and may occur together within the same post.

5.1Convoking as a strategy for mustering Trump’s base

Trump employs convocation as an affiliation strategy in 23% of his tweets to align his base of supporters around particular values. He usually musters his supporters through attacking outsiders, which in addition to the Iranian government, include Democrats, Obama and politicians or other figures who have publicly criticised him. For example, in the tweet below, by appealing to nationalist sentiments via the #MAGA (Make America Great Again) hashtag, Trump marshals his supporters against Bob Coker, a US senator who supported the Iran nuclear deal:

In this tweet, text 24, there are several couplings negatively assessing Bob Corker and the Iran nuclear deal. Trump marshals his ambient supporters around these couplings by the use of #MAGA which convokes his supporters via their identification as Americans who desire their country to be restored to an imagined historical period of apparent magnificence. This has been a common strategy to emphasise that “our country” should be protected against the “evil”, because “We” (or ‘us’ depending on the grammatical context) are threatened by “them” (see Wodak 2015Wodak, Ruth 2015 “Critical Discourse Analysis, Discourse-Historical Approach.” In The International Encyclopedia of Language and Social Interaction, edited by Tracy, Karen: 1–14. New York: John Wiley and Sons. , 66; Kreis 2017Kreis, Ramona 2017 “The “Tweet Politics” of President Trump.” Journal of Language and Politics 16 (4): 607–618. , 613). In addition, the small caps font promotes this bond, and is a typographic strategy Trump regularly uses to garner attention. The tweet positions Trump as part of a nationalistic ingroup (‘us’), as a patriot who cares about the country and its soldiers serving overseas, while Bob Corker is presented as an outsider who does not hold this value. This distinction aims to appeal to Trump’s supporters who share nationalistic bonds about the value of military service. This ‘Us/We’ versus ‘Them’ distinction has been noted in many other studies of political discourse (see for example Sarkhoh and Khosravinik, 2020Sarkhoh, Nadia, and Majid KhosraviNik 2020 “Social Media Discourses of Arabism and the Negotiation of Self in the Middle East.” World Englishes 39 (4): 609–622. ; Khosravinik and Zia, 2014KhosraviNik, Majid, and Mahrou Zia 2014 “Persian Nationalism, Identity and Anti-Arab Sentiments in Iranian Facebook Discourses: Critical Discourse Analysis and Social Media Communication.” Journal of Language and Politics 13 (4): 755–780. for a distinction between Us and Them in Arab versus Persian discourse on social media).

There were a small number of tweets (15) where Trump attempts to muster his audience around positive judgement of Iran, however this mainly occurred as part of a rhetorical strategy to suggest his own positive capacity. For example, the following tweet implies Trump’s skill as a negotiator in the first sentence and then tables two couplings positively appraising Iran:

This tweet, Text 25, embedded the following post by The Associated Press which provides context regarding the "deal":

In this tweet, marshalling is used to construct a ‘we’ that includes Trump and Iran. It is difficult to untangle the logic in this kind of rhetoric, beyond the opportunity it provides Trump for boasting, as less than a month later, he ordered the killing of the Iranian General Suleimani in an Iraq airport, leading to an escalation in tensions.

5.2Finessing as a strategy for closing down the “outgroups”

finessing a coupling involves dialogic positioning through expansive and contractive resources that acknowledge “other communities of values” (Han 2015Han, J. 2015 “# Feminism is Not a Dirty Word’: Axiology, Ambient Affiliation and Dialogism in Discourses Surrounding Feminism in Microblogging.” Honours Thesis, University of Sydney, Sydney, Australia., 78). This includes tweets acknowledging other voices which embellish (i.e. expand) or distill (i.e. contract) a coupling in terms of other potential stances or perspectives (Zappavigna 2018Zappavigna, Michele 2018 Searchable Talk: Hashtags and Social Media Metadiscourse. London: Bloomsbury Publishing.). In Trump’s tweets, 70% of the couplings were distilled, meaning that there is less potential for dialogic alternatives. For example, the tweet below couples Trump and invoked positive judgement. This coupling is finessed through closing down the dialogic space and alternative viewpoints. The tweet begins by allowing room for the alternative perspective that Trump “called off the strike against Iran” but this proposition is immediately suppressed by “never”, distilling the coupling:

The tweet invokes positive judgement of Trump in his capacity to prevent the strike “from going forward at this time”, suggesting that he might give the order again later. Trump later indicated that he had cancelled the attack shortly before it was to begin as he was told that 150 people might be killed (Chappell 2019Chappell, Bill. June 21 2019 “Trump Says He Called Off Strike On Iran Because He Didn’t See It As ‘Proportionate’.” Accessed 20 February 2020. https://​www​.npr​.org​/2019​/06​/21​/734683701​/trump​-reportedly​-orders​-strike​-on​-iran​-then​-calls​-off​-attack​-plan). Another example of finessing a coupling is seen in the tweet below:

In this tweet “never” distills the coupling by closing the dialogic space with respect to other alternative couplings. It forms part of a prosody of tweets that construes two outgroups: Iranians and Obama’s negotiation team led by John Kerry. The positive invoked judgement of Iran in “never lost a negotiation” seems to operate only in the service of belittling Obama’s negotiation team, with the parallelism of ‘never’, an attempt at humorous ridicule.

Another common way of distilling a coupling in Trump’s tweets was through rhetorical questions, which occurred in about 40 tweets. In most cases Trump first proposes a question but subsequently provides or implies the answer to the question himself, meaning that alternative positions are suppressed. He uses this strategy to portray others negatively or close down an outgroup. In the tweet below, for example, Trump couples Barack Obama with negative judgement of capacity:

This tweet was posted a few days after President Obama ordered the withdrawal of all US forces from Iraq by the end of 2011. In this post “handed Iraq over to Iran” is a reference to the likelihood that Iran will take over if US forces leave Iraq. The rhetorical question is used to distill the coupling of Obama with invoked negative judgement of capacity (in terms of his inability to handle the withdrawal of forces strategically so that a country like Iran does not take over). While the question, together with invoked negative judgement of Obama, opens up the possibility of alternative dialogistic positions, by providing an answer (and revealing the question to in fact be rhetorical), Trump closes down the dialogic space. This distills the negative stance about Obama that Trump is attempting to propagate amongst his audience.

There are also some tweets (30% using finess) where the coupling is embellished through resources of heteroglossic expansion, entertaining other possible interpretations and alternative voices. For example, in the tweet below, there are two couplings in Trump’s statements. In the first, the Iranian government (i.e. they) is coupled with negative invoked judgement stemming from Rouhani’s statement, presented as one possible interpretation among others. In the second, the coupling of “threats” with negative invoked appreciation is embellished via the use of modality of possibility (can):

The embellishing of the couplings in these tweets contributes to the overall negativity and the social sanctioning of the Iranian government as an outgroup.

5.3Promoting as a strategy for garnering attention and emphasising values

Promoting is a common strategy in Trump’s tweets (occurring with 40% of the couplings) and is used to interpersonally emphasise a coupling, in accord with Trump’s tendency toward hyperbole. It is mainly used to emphasise his arguments, point of view, and subjective opinions. fostering a coupling is the most common choice (84%) through use of all caps, exclamation marks, repetition of letters, and other various resources for the quantification and intensification of force (see Martin and White 2005Martin, J. R., and P. R. R. White 2005 The Language of Evaluation: Appraisal in English. Basingstoke: Palgrave/Macmillan. , 154; Ross and Caldwell 2020Ross, Andrew S., and David Caldwell 2020 “ ‘Going Negative’: An APPRAISAL Analysis of the Rhetoric of Donald Trump on Twitter.” Language & Communication 70: 13–27. ; Wignell et al 2019Wignell, Peter, Kay O’Halloran, and Sabine Tan 2019 “Semiotic Space Invasion: The Case of Donald Trump’s US Presidential Campaign.” Semiotica 2019 (226): 185–208. ). The couplings in the tweets are usually fostered through the use of upscaled graduation resources, applied for both emphasis and garnering attention. For example, the tweet below, Text 31, couples negative invoked judgment with President Obama’s giving money to Iran. This coupling has been promoted by the use of all caps in the lexical item, “CASH” and the exclamation mark. There is also another coupling of Iran’s deal with negative appreciation:

The fostering of these couplings emphasises the incompetence of Obama as an outgrouping move. This kind of outgrouping strategy is also seen in posts such as the following warning the Iranian administration about how to behave:

This tweet, as a whole, couples Trump with invoked positive judgement, fostered through use of all caps which also has the rhetorical effect of emphasising the threatening dimension of the statement.

While fostering through upscaling of force graduations was the most frequent choice, there are a few tweets (15% of promoted) where the coupling is modulated through choices in graduation that adjust prototypicality through focus. One example is the tweet below which couples a “statement” from media with negative appreciation. This coupling is promoted by modulating via focus to in terms of typicality:

This modulation outgroups the media as a source which usually lies and gives “false statement(s)”. This is part of an overarching rhetorical strategy employed by Trump to position the media as ‘fake news’ (see Kreis 2017Kreis, Ramona 2017 “The “Tweet Politics” of President Trump.” Journal of Language and Politics 16 (4): 607–618. ).