Donald Trump, speaking to the United Nations General Assembly on September 25, 2018, contrasted “the ideology of globalism” with the “doctrine of patriotism,” thus invoking the traditional understanding of ideology as necessarily nefarious. “Doctrine,” of course, is blessed by religious collocations and connotations thereby sanctifying “patriotism.” This contrast is in keeping with his administration’s withdrawal of the US from international human rights institutions. This withdrawal suggests that Trump regards these universal rights as elements of globalism, so we must assume that he regards them also as ideological and thus nefarious. As nationalism is just as certainly ideological as globalism, it is clear that one person’s ideology may be another’s sacred duty, and an ideology that is nefarious to some may be liberating to others.
Though most current scholars would claim that ideologies may be positive or negative and anything in between, most current research on ideological discourse takes as its object of study discourses that the authors find objectionable, for example, van Dijk’s (Reference Van Dijk1998) studies of racism in D’Souza’s The End of Racism (Reference D’Souza1995), and his study of the denial of racism (Reference Van Dijk1992) and ideology and discourse (Reference Van Dijk, Freeden and Stears2013); Blommaert and Verschueren’s (Reference Blommaert and Verschueren1998) study of anti-migrant themes in Belgian public discussion and government publications; Wodak’s (Reference Wodak, Anthonissen and Blommaert2007) study of the ideological constructions of the “German Wehrmacht” and her (2011) study of far-right populist and racist rhetoric; Verschueren’s (Reference Verschueren2012) study of the coverage of the 1857–58 Indian Mutiny in late nineteenth-and early twentieth-century French and English history textbooks; to name but a very few. These studies reflect the tradition that assumes that ideologies are characteristically injurious in that they typically benefit some more or less clearly demarcated groups and disadvantage other more or less clearly identified groups: Nazi and Neo-Nazi ideologies are intended to benefit those of “pure White blood” and disadvantage all others, but particularly Jews and homosexuals (Point 4 in 25 Points of American National Socialism).Footnote 1 It is much less clear who is disadvantaged by the ideologies of human rights activism: while Amnesty celebrates the release of Prisoners of Conscience (PoCs), it does not exclude representatives of governments that imprison PoCs from the protections afforded by human rights.
Besides focusing on objectionable ideologies, these studies assume that the target ideologies are hegemonic and that their goals are at least partially hidden in their texts and must be unearthed by careful linguistic analysis, employing methods from various strands of discourse analysis, linguistic pragmatics, and sociolinguistics, and informed by cognitive, historical, and sociological studies.
The analysis presented in this book departs from the usual assumptions about ideology and takes as its object elements of the discourse of human rights activism, a discourse that many would regard as positive and thus non-ideological, a position assumed in Amnesty International’s Letter Writing Guide,Footnote 2 which directs members and volunteers not to “discuss ideology or politics” in their appealsFootnote 3 in support of human rights. Another version of this Guide is provided in Appendix 1.Footnote 4 I demonstrate in Chapter 3 that the Amnesty International (AI) documentsFootnote 5 display the characteristics of ideology identified in Eagleton (Reference Eagleton2007) and Verschueren (Reference Verschueren2012) and are thus unquestionably ideological. Briefly, these documents are normative in that they invoke international declarations, for example, the UN Universal Declaration of Human Rights (UN UDHR);Footnote 6 international covenants, for example, the International Covenant on Civil and Political RightsFootnote 7 and the International Covenant on Economic, Social, and Cultural Rights;Footnote 8 and international enforcement mechanisms, for example, the International Criminal CourtFootnote 9 and the International Court of Justice.Footnote 10 They are also designed to unify their addressees and readers, especially AI members and volunteers, motivate them to act, and provide them with a universal rationale to naturalize and legitimize their beliefs and actions.
The AI documents explicitly espouse the ideology of human rights and make it the basis for the evaluation of certain governmental and non-governmental actions as violative and the basis for redress of those violations. Additionally, as elements of activist resistance, of counterpower (Featherman Reference Featherman2015; Gee Reference Gee2011), the AI documents and the discourses of human rights in general, though representing universal laws, represent the weaker party in the situations in which they operate, so that their hegemonic aspirations are held firmly in check.
This study takes as context the methodologies proposed in Verschueren (Reference Verschueren2012) for the systematic, controlled, accurate, empirical, and cross-disciplinary study of ideological discourse.Footnote 11 The study adopts Verschueren’s guidelines, and where necessary, adapts, augments, and challenges them. In investigating the relations among the AI documents, the study invokes the extensive literature on the analysis of multi-modal documents (e.g., Bateman Reference Bateman2018; Forceville Reference Forceville and Machin2014, Reference Forceville2017; Forceville and Clark Reference Forceville and Clark2014; Jewitt Reference Jewitt2009, Reference Jewitt2014, Reference Jewitt, Georgakopoulou and Spilioti2016; Machin Reference Machin2014; Roque Reference Roque, Tseronis and Forceville2017; Tseronis and Forceville Reference Tseronis, Forceville, Tseronis and Forceville2017; van Leeuwen Reference Van Leeuwen, Tannen, Hamilton and Schiffrin2015).
The Amnesty International webpages, emails, and appeal letters that constitute the primary data for the analysis presented in this book are a convenience sample collected over the period 2013–16. The appeal letters came to me in emails I received from Amnesty as an active member of the organization. These emails encourage their receivers to forward the appeals to government or other officials with the power to improve the situations of the PoCs and others who are the subjects of the appeals. Amnesty members who receive the emails may edit the appeal letter or may just click a SEND button. As the appeals analyzed for the study were collected over more than a decade ago, some of the individuals on whose behalf the appeals were created have been released, others may still be imprisoned, some have died, and others may have revoked their consent for Amnesty International to campaign on their behalf.Footnote 12
This book presents an analysis of a corpus of forty AI appeal lettersFootnote 13 on behalf of “Prisoners of Conscience”Footnote 14 and other individuals and classes of individuals whose human and/or humanitarian rights have been violated, and in support of specific human and humanitarian causes, for example, freedom from torture, freedom of expression, arms control treaties, protection of LGBTQIA+ (lesbian, gay, bisexual, transsexual, queer, intersex, asexual, collectively) people, and rights to abortion.
The appeals are sent to AI members/volunteers by AI staffers, embedded in an email that provides background information on the situation that is the subject of the appeal, along with links to AI webpages with further information. These appeals can also be found on AI’s website embedded in webpages that are similar to the emails. Ostensibly, the appeals are to be electronically forwarded by the volunteers to their “Message Recipients,” who are generally powerful government functionaries, by clicking on a button marked SEND.
The analyzed letters were collected into a single Word document (Appendix 2); a further forty letters are collected in a separate corpus (Appendix 3). These allow other researchers to compare their analyses with mine, the kind of counterscreening recommended by Verschueren (Reference Verschueren2012: 22).
Verschueren (Reference Verschueren2012) presents “pragmatic guidelines for empirical research” into ideology in language use. While Verschueren’s book presents a well-organized framework for a description of an ideologically infused discourse genre, it is not, as he says, a theory. Consequently, its framework must be augmented with a broad range of linguistic, pragmatic, and discourse analytic theories in order to provide explanations for the specific phenomena encountered in the data identified by Verschueren.
Even though Verschueren’s (Reference Verschueren2012) book is “designed for training purposes” - its “target audience includes … historians, sociologists, anthropologists, and political scientists, to name but a few broad categories” (xii) - Costa (Reference Costa2014) deems it a “major book for students and advanced scholars in pragmatics and discourse analysis” (310). The current book is organized according to Verschueren’s Guidelines, Procedures, and Caveats for empirical ideological research and thereby presents an analysis of the discourse and language of the AI appeal letters by engaging a thoroughly thought-out framework by an important scholar in the study of pragmatics and ideology.
The analysis I present here is derived from several approaches I have adopted in studying the AI letters. This multiplicity of approaches is in keeping with the precept that discourse analysis is a multi-disciplinary, multi-perspectival enterprise. These approaches include initial studies of aspects of the letters: genre, speech acts and (im)politeness (Delahunty Reference Delahunty2013), ideological conflict (Delahunty Reference Delahunty2015), ideological positioning of multiple audiences (Delahunty Reference Delahunty2017), participant positioning (Delahunty Reference Delahunty2018a), rhetorical moves and steps (Delahunty Reference Delahunty and Götzsche2018b), and the re-entextualization, re-contextualization, and re-genrefication of information from other texts and genres (Delahunty Reference 283Delahunty, Bhatia and Tessuto2021, Reference Delahunty2023).
Research Questions and Data
This section addresses the preliminary methodological issues Verschueren (Reference Verschueren2012) raises in his chapter 2 on “Pragmatic rules of engagement” and presents as a series of rules. These rules deal with the formulation of research questions, the selection of data, and pragmatics-derived rules for the study of ideological discourse to “enable interpretation with due regard for what can be intersubjectively established to count as evidence” (21, emphasis in original; footnote omitted). The research questions may be prompted by the analyst’s “involvement” (emphasis in original) with the discourse and perhaps a sense that something is “wrong” (21) with it, characteristically, that there is a contradiction between its overt and covert messages. My initial interest in analyzing the AI letters derives not from any sense of contradiction, but from my desire to know what happened to “my” letters when I clicked the SEND button. Did they go to their Message Recipients? If so, what happened to them there? Who, if anyone, reads them? If they did not go to the Message Recipients, then what happened to them? Was clicking SEND merely a signal to AI’s computer systems to send a “Thank you. Please donate” email? This puzzle about the fate of my letters led to questions about their roles in AI’s activism and fundraising, which led to whether the letters were elements in an ideological discourse, and if so, by what criteria.
Rule 1: Formulate researchable questions, i.e., questions to which answers can be given, supported with empirical evidence and susceptible to counterscreening.
Verschueren prefers “research questions” rather than hypotheses, as the latter require the articulation of “an assertion that can be verified or proven,” or as Karl Popper claimed, can only be falsified (Andersen and Hepburn Reference Andersen and Hepburn2021). Analysis of ideological discourse should be “data-driven” and “theoretically grounded in [a] methodology for letting the data speak” (Verschueren Reference Verschueren2012: 23, emphasis in original), though “criteria for data to count as (carriers of) evidence” are essential (23, emphasis in original). In the coming chapters, I ask and answer the following questions: What ideological functions, meanings, frames of reference, and interpretation undergird the appeal letters? What linguistic choices enact these ideological positions? How do those choices enact a specifically oppositional, and hence explicitly dialogic, genre? Can the analytic methods developed by scholars to study the discourses of nefarious ideologies to expose their hidden claims be profitably applied to the discourses of liberationist ideologies? Are the two discourse types similar and, if they are, in what ways, for example, do liberationist discourses also hide important claims?
Rule 2: Before an aspect of meaning can be seen as an ingredient of ideology, it should emerge coherently from the data, both in terms of conceptual connectedness with other aspects of meaning and in terms of patterns of recurrence or of absence.
While this book argues that the AI appeals fulfill all the functions of ideology proposed in Eagleton (Reference Eagleton2007), here I briefly address Verschueren’s more general methodological concern. I claim that human rights, “an aspect of meaning,” are “an ingredient of ideology” that “emerge[s] coherently from the data.” The AI letters consistently and systematically derive their evaluations of the problems they address and the solutions they propose by referring to human and/or humanitarian rights and justify their appeals for redress by reference to those rights, though it is only rarely that reference is made to specific instruments, for example, the UN UDHR; they are typically generic, for example, to “the right to life” or “human rights.”Footnote 15
The letters occasionally offer concessions to the Message Recipients that rights violations are based on competing exigencies, for example, governments’ responsibilities to bring criminals to justice (#4 Al-Qahtani, Iraq),Footnote 16 which are then rebutted by characterizing the violations as local failures to meet universal and international standards (of legal procedure, and the like). This is not to say that the discourse of human and humanitarian rights is in general internally consistent, coherent, and settled (see Chen’s Reference Chen2015: 434–435 remarks on the “evolution” of the UDHR), merely that the AI letters adopt consistent positions on the issues they address and express these in consistent ways, thus allowing the ideological meanings of the letters to be seen to emerge coherently from the data.
In the few letters that include them, concessions are based on local, often legal, considerations. Where concessions might be made on the basis of religious ideologies, the letters are silent. For example, in the case of Raif Badawi (#18 Badawi, Saudi Arabia), who was flogged by Saudi Arabian authorities in front of a mosque after Friday evening prayers, the appeal might have acknowledged that the verdict and punishment are based in Islamic (Sharia) law. Certainly, one might reasonably infer an implicit criticism of that law and the religion it derives from. In the case of Beatriz (#8 Beatriz, El Salvador), who was denied an abortion even though continuing her pregnancy was endangering her life, the appeal might have conceded that El Salvador’s absolute ban on abortion, like that of other countries, for example, Argentina, is based on Catholic tenets. The letters’ appeals to universal rights avoid a direct confrontation between religious and secular values, even if one might argue that those secular values ultimately derive from religious notions of the transcendent value and dignity of individual human beings, for example, Quinlan (Reference Quinlan2011), and thus avoid inviting the response that values deriving from a god, however partial and capricious, outrank values deriving from mere human deliberations.
With a few exceptions, the letters typically also omit reference to, or discussion of, broader political contexts. In this, the letters appear to follow the AI letter-writing guidelines that explicitly prohibit politicizing an appeal, thus implying that appeals based on human rights are not political discourse. The exceptions are two letters to President Obama urging him as representative of the US government to lead “the effort to stop the unregulated flow of weapons” (#2 Arms Trade, USA) by supporting “the highest possible human rights standards within a new Arms Trade Treaty.” This treaty was adopted by the UN and came into force on December 24, 2014.
Rule 2.1: Types of data must be varied horizontally and vertically.
“Horizontal variation refers to genre differences” (26, emphasis in original). As we will see in Chapter 2, several distinct though related genres are relevant to the study of AI appeal letters. The most immediately relevant to the AI appeal genre is the email sent to volunteers in which the appeal is embedded. This email is addressed to the volunteer, provides information on the case, which is characteristically reproduced in the appeal, and urges the volunteer to SEND the embedded appeal. Other relevant documents include bulletins on AI webpages, AI reports on violative situations and the case information on which these reports are based, as well as the laws, legal decisions, treaties, covenants, and declarations that define the various rights invoked in the letters. And on a larger scale, discussion of human rights in a broader range of contexts.
Verschueren’s vertical variation refers to levels of linguistic structure, for example, “patterns of word choice, presupposition- and implication-carrying constructions, interaction patterns, and more global constructs (such as argumentation patterns)” (Verschueren Reference Verschueren2012: 27). In response to later guidelines, I address all of the items mentioned in Rule 2.1, and considerably more.Footnote 17
Rule 2.2: An appropriate amount of data is required.
For Verschueren, an “appropriate amount of data … depend[s] on the specific nature of one’s research question” (28). Clearly my corpus is far smaller than those referred to in Verschueren’s book: his studies of discourses on migrants in Belgium and of discourses on colonization in English and French history textbooks from the late 1800s and early 1900s. Nonetheless, my database is sufficient to provide reliable results which can be further tested by counterscreening and by the study of relevant related genres, for example, the embedding emails from AI, AI webpage documents and reports, and the legal, academic, and journalistic polemics surrounding human rights and their violations. The letters in my corpora are similar enough to each other in function, rhetorical organization, and patterns of linguistic choice for me to be confident that other researchers’ analyses of the letter corpora in Appendixes 2 and 3 would be consistent with mine. Moreover, the size and generic consistency of the AI letters suggest that my analysis may avoid the risks Verschueren (Reference Verschueren2012: 28) associates with “sizable” corpora – that given corpus enough and time an analyst will “find – or construct – the ‘desired’ patterns of meaning.”Footnote 18
Rule 2.3: Whatever is found throughout a wide corpus should also be recoverable in (at least a number of) individual instances of discourse.
Rule 2.3 is formulated to minimize the likelihood of researcher bias by requiring that claims about the corpus be manifested in individual instances of the target discourse. For example, Blommaert and Verschueren (Reference Blommaert and Verschueren1998) devote chapter 7 (162–189) “to the analysis of a single leaflet that was widely distributed by the official Belgian information bureau Inbel …” (Verschueren Reference Verschueren2012: 29).
Rule 2.4: The quality of the data must be carefully evaluated in view of the precise research goal.
The goal of the current research is to develop an analysis of AI appeal letters that identifies the linguistic choices their writers make and the ways in which those choices enact the ideological formations underlying human and humanitarian rights. The AI letters fit this goal because they are written by (specialist) staffers in AI in accordance with AI guidelines and reflect AI’s mission to promote human rights. They are therefore authentic representatives of the genre, even if the details of their composition are obscure. They function as directives to their addressees and enact the characteristics of ideology identified by Eagleton (Reference Eagleton2007) discussed in Chapter 3 below, and briefly re-presented here: the letters are based on a body of ideas characteristic of a particular social group, specifically AI staffers and volunteers, and, probably, staffers and volunteers of other human, humanitarian, and civil rights organizations. Because they derive from universal norms, these ideas offer legitimating epistemological and evaluative positions according to which individuals can identify themselves and others and which motivate them to act in their own and others’ social interests by engaging in a discourse of power and counterpower (Featherman Reference Featherman2015) – they include projected “voices of dissent” (Verschueren Reference Verschueren2012: 30). Their epistemological, legitimating, and evaluative functions rely on the assumption of universality undergirding the UN UDHR and of the various humanitarian conventions they invoke. Elements of these instruments are typically invoked by definite descriptions, for example, “the right to life,” thus presupposing the existence of the named rights. As these instruments have the force of law, they are normative and thus allow for the naturalization of evaluations based on them, which are typically expressed as unhedged categorizations, for example, “Flogging is a flagrant violation of the right to life” (#18 Badawi, Saudi Arabia). The genre represents the direct expression of AI’s ideological positions and range of interests and has remained consistent over the years I have been collecting instances.
Six Analytic Dimensions
Verschueren (Reference Verschueren2012) compares and contrasts the history textbooks he analyses across six dimensions (31–50); here, I briefly discuss these dimensions in relation to the AI letters.
(1) Language (Verschueren Reference Verschueren2012: 31)
The language of the letters and of the embedding emails is mainly standard (American) English, for example, “authorize” (#1 Targeted Killings, USA) rather than “authorise,” though occasionally a British spelling slips in, for example, “labourers” (#40 Cobalt, Apple Corp.). I have not found any clearly non-standard usages. The underlying assumption must be that the addressees (or their surrogates) speak or read English well enough to be able to understand the letters. This assumption reflects the global importance of English as a lingua franca and its predominance among the official languages of the UN and working languages of other international institutions and organizations (Crystal Reference Crystal2003; Hickey Reference Hickey2012). The literacy level required to write and read the letters ranges from requiring a graduate degree (#1 Targeted Killings, USA) to 11th grade (#11 Pussy Riot, Russia), though the readability formulas differ in their grade level assignments. The relatively high literacy levels suggest that AI staffers are highly literate and that AI assumes that its members and volunteers are equally so. This literacy level may also suggest that AI assumes that the Message Recipients or their surrogates will have similarly high levels of English proficiency, even if few if any of the letters will actually be read by their Message Recipients and that what is persuasive to them is the number of letters sent and/or received, not the sophistication of their arguments. While the letters are linguistically consistent, they occasionally display minor editing errors, for example, inconsistent punctuation and the omission of a definite article in “I am deeply concerned that US government’s …” (#1 Targeted Killings, USA; emphasis added), and collocations suggestive of second language usage, for example, in (#3 Satire as Crime, Egypt), “their free expression of speech,” indicating, again, that their arguments may be less important to their success than their numbers.Footnote 19
Complementary to the readability levels of the appeals are their keywords. Keywords are those that distinguish a target corpus from a general corpus, in this case the 10,311 words of the appeal letter corpus and the 10 million words of the BNC_COCA_mixed_Sp_Wr_US_UK_10_million corpus. The analysis returned eighty-one keywords whose frequencies in the appeals and reference corpus are substantially different. As one would expect, most of these are specific to legal and human rights domains, for example, deport, whose various forms (deportation, etc.) occur seven times in the appeals corpus, which is 5,819 times more frequent than they occur in the reference corpus. The least frequent keyword listed is execute, which occurs 25.52 times more frequently in the appeals than in the reference corpus. The ten most frequent words and their variant forms in the appeals corpus are: program, politic, deportation, meaning, relation, extrajudicial, detain, perpetuate, abduct, and defame. While several of these are not specific to the legal and human rights domains, most are. However, when we look at the collocations of those that are not, we find them to be human rights specific, for example: “These programs are a violation of the human rights of people across the globe” (#23 Mass Surveillance, USA; emphasis added); “They are too often unable to make decisions about who they can love, who they have intimate relations with, who they marry, and about whether to have children and how many” (#33 Contraception, Burkina Faso; emphasis added).Footnote 20
(2) Temporal Perspective (Verschueren Reference Verschueren2012: 32)
As a genre, the AI appeal letters have a history going back to 1961, when AI was founded, though elements of that genre are clearly much older, for example, protest directives. According to an AI webpage,Footnote 21 “Over the years, human rights have moved from the fringes to centre stage in world affairs,” and the categories of rights appeals have broadened from focusing on individual PoCs to include such issues as the abolition of torture and the death penalty, sexual and reproductive rights, discrimination against various categories of people, and refugee and migrant rights, among others (see Domaradzki et al. Reference Domaradzki, Khvostova and Pupovac2019; Vašák Reference Vašák1977). However, demands for human rights, in their modern formulations, go back at least as far as the European Enlightenment and the American and French Revolutions. (Chapter 2 provides a brief history of AI.)
While Verschueren’s history textbooks describe and evaluate events that occurred prior to, and sometimes a long time prior to, when they were published, the AI appeals deal with situations current when they are written (and, ideally, sent). As they deal with situations of considerable urgency, they direct their addressees to act immediately upon their receipt.
(3) Geographical Perspective (Verschueren Reference Verschueren2012: 33)
AI was founded in England in 1961, and its headquarters are still there, though it has branches worldwide, with an important one in the US. Currently, AI claims to be “a global movement.”Footnote 22 While the legal instruments AI appeals to were designed, negotiated, and enacted to be universally applicable, certain Muslim countries have developed analogous instruments based on Muslim (Sharia) law (e.g., the Cairo Declaration on Human Rights in Islam (CDHRI) enacted in 1990),Footnote 23 which, for example, legalize executions, though UDHR does not (UDHR Article 3).
The appeal letters in my corpus were sent to eighteen countries: Algeria, Burkina Faso, Dominican Republic, Egypt, El Salvador, Indonesia, Iraq, Malawi, Myanmar, Peoples Republic of China, Russian Federation, Saudi Arabia, South Africa, Sudan, Turkey,Footnote 24 USA, Uzbekistan, and Vietnam, as well as to Apple Corporation. Amnesty’s Country Profiles includes information on AI’s “research and campaigning across the world, with all the latest news, blogs, and reports,” as well as a link to the current year’s report on the “The State of the World’s Human Rights” “during 2023 in 155 countries.”Footnote 25
(4) Size and Degree of Detail (Verschueren Reference Verschueren2012: 33)
The letters re-entextualize, re-contextualize, and re-genrecize information in their embedding emails and in the AI webpages, reports, and case files, which may be indefinitely large, as they contain internet and other links to electronic documents which may also contain such links.
In keeping with AI’s letter-writing instructions, the appeals are typically brief: they range from about a third of a page to about a page and a half and are on average approximately 300 words. They typically provide only information deemed essential to their specific cases, including the names of individuals involved, what the subjects of the appeals have been accused of, the threats to their persons, liberties, and lives, and the specifics of the redress sought. The letters may also contain elaborations of the primary appeals; for example, an appeal initially on behalf of a single individual may be generalized to include all the individuals belonging to a relevant category.
(5) Coverage (Verschueren Reference Verschueren2012: 34)
The letters vary in the amount of coverage of the case they provide. Some letters provide little coverage, just a brief description of the situation motivating the letter and a brief directive to redress that situation, but always (just) enough to state the problem and to urge its solution. Other letters may provide somewhat more detail on both elements. A very few letters include concessions and their rebuttals. Occasionally, the letters include bulleted (or otherwise displayed) lists of demands, though “demand” occurs only once in the corpus, and even there it is used impersonally and non-directively: “The revelations demand careful analysis by Congress” (#23 Mass Surveillance, USA). In contrast, the embedding emails may direct their addressees to demand that the letter addressees redress a wrong.
(6) Intended Audience (Verschueren Reference Verschueren2012: 34)
Verschueren’s history books were primarily designed as textbooks for primary and secondary school students, though one was for a “general audience” and another was a “general children’s book” (see his Table 1, 38–39). As history texts dating from the late nineteenth century through the first two decades of the 20th, they display the tension between the ideologies of historiography as the objective presentation and explanation of facts and the more traditional excitation of uncritical patriotism, typically to the advantage of ruling elites, and their roles in identity formation.
The ostensible intended audience of the appeal letters must be the explicitly listed Message Recipients identified above the letter. These are typically important government functionaries, for example, presidents, prime ministers, defense ministers, and the like. Presumably these are individuals assumed to have the power to accede to the requests in the letters.
However, the letters are sent to AI volunteers, embedded in emails that present the problematic situation and exhort the volunteers to forward the letters by clicking a SEND button on the webpage. Certainly, the embedding email is designed to motivate the volunteers; however, the appeal letters themselves also seem to function in this way and, thus, seem to be designed for at least two audiences – the Message Recipients and potential letter SENDers. For both audiences, though especially for the former, it is important that the facts represented in the letters be accurate and AI is well-known for the care its staffers take to establish and accurately represent the facts of each case.
If we assume, as I suggest, that the AI volunteers who receive and are named in the emails are also an audience for the letters, then we need to be able to say something about who these people are. Presumably they are among the 10 million people around the globe that AI claims to represent and engage “to fight injustice all around the world.”Footnote 26 I do not know how many of these 10 million are on the organization’s email list, though I assume that the addressees on the email list have opted to be on it, most likely by donating to AI. However, I do not know how often that list is pruned to eliminate inactive members or people who prefer to no longer receive AI’s frequent, indeed daily, emails.
As I’ve noted, appeals can be accessed by going to the AI webpage and clicking on the link to an ongoing case. Obviously, the netizen who accesses cases in this way is not personally identified, unlike the recipients of the AI emails, who are directly addressed by first name. Nonetheless, the email recipients and those who purposefully access AI webpages are engaged in acts of identity creation or maintenance.
Verschueren raises the issue of “objectivity” in the history texts, and here I address its relevance to the appeal letters. To do so, I distinguish between objectivity and accuracy. I address the issue of accuracy in AI documents in Chapter 2, noting that the organization, having learned from its occasional errors, is recognized for the accuracy of its characterization of situations, though ultimately these characterizations are dependent upon local witnesses who typically have been affected by the situations. Accuracy simply has to do with getting the facts right. Objectivity is a stance toward the representation of a situation. It suggests a detached, impersonal, uninvolved view of a situation in contrast to an engaged, involved view. AI’s documents are accurate, personal, and involved, but not objective, in the sense I’ve just outlined. The letters’ appeals to pathos, to the readers’ empathy and sympathy, reflect this involvement and impute it to the volunteer letter SENDers. I suggest that the invocation of empathy is designed rather more for the volunteers than for the Message Recipients, whose attitudes to the cases are likely to be unsympathetic and who may respond angrily to the appeals. It is hard to imagine that a US state governor or attorney general who has refused to release a prisoner despite three court rulings to do so would be moved by appeals to their better natures (#20 Woodfox, Louisiana, USA).
As I discuss in more detail in Chapter 2, I am unclear about where my email goes when I click SEND, but I immediately receive an invitation to donate to AI, thus implicating that the letters serve at least three functions: appealing to government officials on behalf of AI’s causes; engaging and motivating AI volunteers; and fundraising. This connection between misery and money is not unique to AI; for example, according to an article in ProPublica, “Inside the Red Cross, the Haiti disaster was seen as ‘a spectacular fundraising opportunity,’ recalled one former official who helped organize the effort.”Footnote 27 See Mann and Thompson (Reference Mann and Thompson1992) and Biber, Connor, & Upton (Reference Biber, Connor and Upton2007) on philanthropic fund-raising letters, and Delahunty (Reference Delahunty and Götzsche2018b) for a comparative analysis of AI appeals and philanthropic fund-raising letters.
Finally, it is important to note that while Verschueren’s history textbooks, and presumably, history textbooks as a genre, are descriptions of prior events and reflections upon them, the AI letters and their related documents both represent current historical events and intervene in them. And just as there are “histories of history writing,” there are histories of human rights activism, for example, Neier (Reference 289Neier2012) and Sikkink (Reference Sikkink2017).
Book Plan
Chapter 2 briefly discusses rights and human rights. It provides a short history of Amnesty International and a discussion of the multi-modal relations among its primary document genres: the appeal letter, which is the main focus of this book; its emails to volunteers/members and the webpage in which the appeal letter is embedded; the email it sends asking for a donation in response to a volunteer SENDing the appeal; and an AI report which documents in detail an appeal situation. The chapter provides a brief analysis of the primary document genres, paying particular attention to the linguistic and visual resonances occasioned by their mutual contextualizations and proposes a Relevance Theory explanation for the intricate relations between the documents’ linguistic and visual modes and argues for the ideological study of AI documents, particularly of the appeal letters.
Chapter 3 reviews several important understandings of ideology and concludes with discussion of Verschueren’s Theses 1–4, which articulate his characterization of ideology and its manifestations and roles in discourse. These theses along with the attributes identified in Eagleton (Reference Eagleton2007) provide the bases for the arguments presented throughout the book that the AI appeals and related documents are ideologically imbued genres.
Chapter 4 addresses Verschueren’s Caveat 1 through his Procedure 2.3, which deal with data, its contexts, and the “lines of vision” connecting the documents with their relevant contexts. The chapter investigates the ideological relevancies of the appeals’ social, political, historical, and geographical contexts; their immediate context of situation; and their linguistic contexts, guided by Verschueren’s preliminary caveats, his Guidelines 1 and 2, and their associated analytic procedures. This investigation requires that Verschueren’s broad guidelines be augmented by multiple discourse analytic theoretical apparatuses: deixis; speech act theory; Gricean and post-Gricean pragmatics; contextualization cues; various theories of participant roles; positioning theory; theories of argumentation; move and step analysis; epistemic vigilance; and discussion of a broad range of specifically linguistic devices.
Chapter 5 addresses Verschueren’s Guideline 3 through Procedure 3.3 and associated caveats and procedures, which deal with language, social structures, processes, relations, and meaning. It investigates the “dynamics of meaning generation in relation to issues pertaining to social structures, processes, and relations” (Verschueren Reference Verschueren2012: 117), the roles of grammatical relations and their interactions with semantic roles, particularly the identification of agents and experiencers of violative and other actions and the ideological positioning of those referents relative to human-rights-based evaluations of the actions. It also examines the topics of the appeals and the language of the rhetorical moves and steps that constitute them.
Chapter 6 addresses Verschueren’s Procedures 3.4 to 3.7 and his Guideline 4. These deal with implicit meanings, tropes, sequential ordering of genre and interactional elements, and metapragmatic indications in the AI appeals. It investigates “carriers of implicit meanings” (Verschueren Reference Verschueren2012: 166) that function as “strategies of meaning generation” (186) “pertaining to social structures, processes, and relations” (191).
Chapter 7 provides a brief reprise of the contents of the prior chapters, of the main results of the analysis, and of the analytic methods employed. It revisits ideology and ideologies from the perspective of the analyses presented in the previous chapters and briefly contrasts AI and human rights documents with a pair of US Neo-Nazi documents. It identifies a number of “unexpected findings,” as directed by Verschueren (Reference Verschueren2012: 64), as well as changes to AI’s communicative practices since its inception. It concludes with an invitation to readers to use the data in Appendixes 2 and 3 to counterscreen my analysis.
Permission to Use Amnesty Documents
We are happy to grant you and Cambridge University Press permission to publish our material in this case, subject to ensuring that you add a paragraph in your introduction about where the material has come from, to provide some context, which will mitigate concerns about any risks to Amnesty. We agree that the following form of words will discharge this obligation.
The Amnesty International documents that constitute the data for the analysis presented in this book are a convenience sample collected over the period 2013–2024. The appeal letters came to me in emails I received from Amnesty as an active member of the organization. These emails encourage their receivers to forward the appeals to government or other officials with the power to improve the situations of the Prisoners of Conscience who are the subjects of the appeals. Amnesty members who receive the emails may edit the appeal letter or may just click a SEND button. As the documents referred to in the study were collected over more than a decade, some of the individuals on whose behalf the appeals were created have been released, others may still be imprisoned, some have died, and others may have revoked their consent for Amnesty International to campaign on their behalf.
