1Although the number of scientific works devoted to conspiracism and conspiracy theories has increased significantly since the early 2000s, they do not provide a stable definition of the phenomena. This difficulty can be explained primarily by the number of disciplines that have taken on “conspiracism”, making it difficult to develop a shared conceptual and transdisciplinary framework (Boullier et al. 2021: § 2; Butter and Knight 2015: 21-39). We thus speak interchangeably of conspiracy “theory”, “narrative”, “discourse”, “style” or “rhetoric”, “causality”, “reading framework”, “belief”, “mentality”, “historiosophy” or “conspiracist worldview”, depending on the disciplinary approach. This plurality, if not scattering, of definitions and approaches is made even more confusing by several peripheral notions that gravitate around them: the notion of “fake news”, “rumor” or “disinformation” (Giry 2022: 43-64), “post-truth” (Danblon 2020: § 4) or “populism” (Bergmann 2018: 3).
- 1 In the rare cases where the label “conspiracy theorist” is claimed, it is usually either misused or (...)
2Another difficulty when working on conspiracism lies in the constitution of a corpus of “conspiracy” theories and discourses. On the one hand, the authors of these discourses rarely describe themselves as “conspiracy theorists1” and tend to reject such a qualification as infamous: it is, therefore, complicated to rely on self-categorization. On the other hand, the absence of a consensual definition of conspiracy theory means that the corpora come up against a problem of “false positives” and “false negatives” (Nera and Schöpfer 2023: 6-8). The corpora built around the definitional traits selected are either too broad and group together discourses whose common points are difficult to grasp (Paparouni 2010: 99); or too restrictive and miss theories that are nonetheless commonly accepted as “conspiracy theories”. In the absence of a stabilized definition, the “conspiracy theories” selected to serve as a basis for work are often chosen for their prototypical nature, to the point where they sometimes tend towards a caricature of “conspiracy discourse”, with the risk of producing equally caricatured conclusions (Hagen 2018: 22). This difficulty in finding the right definitional framework for “conspiracy discourse” makes its categorization particularly vulnerable to the biases of disqualifying labeling. The accusation of “conspiracism” is an effective argumentative strategy that is well identified by those involved in public debate as a means of discrediting a discourse or a person who is accused of a form of irrationality or credulity (Klein and Van der Linden 2010: 133). This risk is all the greater because such labeling can be used to justify disqualifying (or even censoring) discourses critical of political and economic power (Giry 2017: 6).
3To mitigate the risks of unjustified labeling, we opt for a case-by-case empirical approach to “conspiracy discourse” (Giry 2017: 7). Without such an approach, another risk for the analyst is to fall into a circular logic that consists of searching a corpus of “conspiracy discourse” for the elements that previously enabled him to categorize it as such – which will inevitably lead to confirming its presence. To avoid these biases, we have worked on a set of mixed discourses – bringing together productions likely to be considered “conspiracy theorists” and others that are not – to see whether or not examining their linguistic materiality using the tools of discourse analysis and argumentation can enable to draw a line between them, without presupposing from the outset whether such a dividing line will emerge, and if so, where it will pass.
4This article compares frequent defining features found in the literature with a corpus of positions taken on vaccine policy during the Covid-19 pandemic, to check whether these features prove relevant in distinguishing between “conspiracy” and “non-conspiracy” discourses. We look at how the speakers point out inconsistencies they perceive in the institutional discourse on vaccination, and whether the way these inconsistencies are interpreted makes it possible to characterize a discourse as “conspiracy”. This study is based on argumentation analysis and, more broadly, on discourse analysis to identify (if present) the language markers likely to justify such a labeling. To this end, we have built up a corpus of discourse critical of vaccine policy. Although these critical discourses are not all “conspiracy”, numerous studies have shown the penetration of “conspiracy” arguments, particularly within anti-vaccine groups. This choice of corpus therefore allows us to bring together discourses that share the same critical position but are based on different argumentative and discursive constructions.
5After presenting the elements of the definition of the term “conspiracy” present in the literature on which we base our work, we show how they can echo ways of dealing with incoherence in discourse. We explain the protocol governing the collection of our corpus and the methodology used for its analysis. We then present a case study comparing three ways of dealing with incoherence and suggest that the differences uncovered constitute clues that can be used to identify the “conspiracy” nature of a specific discourse and distinguish it from a “non-conspiracy” one.
6Now that we have established that there is no consensual, stable definition of the terms “conspiracy”, “conspiracism” and “conspiracist” and warned against their possible disqualifying exploitation, we will lighten the rest of our article by abandoning the systematic use of quotation marks. Indeed, even if they have not been articulated in a consensual definition, several traits presented as characteristic of conspiracism have been discussed in scientific literature. We are drawing on Marc Angenot’s definition of “conspiracy logic”, which he describes as “a cognitive and hermeneutic device, an exclusive way of deciphering the world that has, above all, a history that can be followed in Western modernity” (2010: 28; translated from French). We call “conspiracy discourse”, and more precisely “conspiracy argumentation”, the linguistic manifestation of this logic, understood as a set of empirical phenomena (markers, modes of structuring arguments, positioning vis-à-vis counter-discourse, etc.). We also consider, at the very minimum, that conspiracism always revolves around the notion of plot, understood as the action of a group of actors acting in secret for their own interests and outside the legal and institutional framework: one does not plot alone, and one does not plot in broad daylight (Douglas et al. 2019: 4-5).
7This core definition is supplemented by several other criteria that are more or less consensual. The group of conspirators tends to produce an “official” and false narrative to conceal its actions (Keeley 1999: 117) and their conspiracy extends to a certain scale (often institutional). In general, conspiracy theories are characterized by their scale: this may be historical (the conspiracy extends over several centuries, or even several millennia), geographical (the conspiracy extends across the whole world), or demographical (hundreds or even thousands of people involved [Campion-Vincent 2005: 7], or a small group of people sometimes verging on omnipotence). In the most systematized conspiracy theories, all elements of reality, even the most trivial, can be analyzed coherently as clues or evidence of the existence of a hidden “plan”. For Loïc Nicolas, the explanation through conspiracy “enables the ordering of very diverse, sometimes contradictory, elements of social reality within a single, totalizing narrative, in order to hold them together” (2014: § 3; translation from French). The viability of a large-scale conspiracy requires the group of plotters to be very well organized (and therefore highly homogeneous, even essentialized) and convinced of the importance of the project (Swami et al. 2014: 572, cited by Räikkä 2018: 208).
8Furthermore, conspiracy theories are often Manichean in nature and focus on denouncing a particularly malevolent adversary (Hall 2006: 207, Bonnet et al. 2022: § 3) or even, in an esoteric or theological reading of the world, a literally demonized one (Poliakov 2006: 56-57). Conspiracism also tends to produce “self-sealing” discourses (Cassam 2019: 71, Nicolas 2010: § 14; Sunstein and Vermeule 2008: 5): conspiracist logic is ultimately unfalsifiable because it is total, capable not only of overturning objections but also of integrating a priori unconnected elements so as not to disregard any of them. The absence of proof of a conspiracy becomes proof of a dissimulation; any counterargument against the existence of a conspiracy can be interpreted as proof of the conspirator’s complicity with the plotters.
- 2 Campion-Vincent, Taguieff and Danblon and Nicolas are translated from the French.
9This totalizing dimension of the discourse can be found in the main defining features of conspiracy rhetoric: “Nothing happens by accident”; “Nothing is as it seems”; “Everything is connected” (Barkun 2013: 3-4); “it’s all connected. Nothing happens by chance” (Campion-Vincent 2005: 11-12); “everything that happens is the result of hidden intentions or wills” (Taguieff 2021: 76-79); “everything must be meticulously scrutinized” (Danblon and Nicolas 2012: 38-40)2. The totalizing explanation refuses to attribute anything to chance and interprets everything as resulting from the action of the plotters. Angenot (2012: § 38) defines conspiracy logic as a process reconstructing the coherence of events, an alternative historiosophy that competes with the more orthodox or more commonly accepted visions of history and can respond to them point by point. To unfold, the conspiracy logic requires a pre-existing historiosophy against which to oppose a competing historiosophy.
10We have chosen to abandon the features that characterize conspiracy theories as false, or conspiracy theories as resulting from a series of errors in reasoning or cognitive biases. While these traits may be relevant for identifying certain prototypical conspiracy theories, they are less so when the conspiracy theories are irrefutable theories (Paparouni 2010: 102) or when the nature of the data analyzed (oral interviews) does not allow us to reconstruct conspiracy reasoning in its entirety. We favor a descriptive and comprehensive approach to discourse, seeking to show how meaning can be derived from its argumentative structuring (Plantin 2016: 7-9).
11While these characterizing features alone do not fully define a conspiracy theory, they can be related to empirically observable linguistic markers that would be their correlates. To ascertain whether these markers can be used to distinguish conspiracy from non-conspiracy discourse, we look at sequences that are comparable in terms of their themes and structures. We begin by justifying the attention we have decided to pay to identifying and dealing with incoherence in “official” discourse; we explain the method used to compile our corpus and then proceed with three case studies.
12The strong argumentative dimension of conspiracism has already been noted by Steve Oswald and Thierry Herman (2016: 296); it is related to the fact that to exist, conspiracist discourses necessarily oppose an “official” discourse. In this respect, it is significant that the conspiracy argument can be led to stage or reconstruct such a discourse when it does not clearly exist (for example, the media and politicians are accused of holding the same discourse, even when there are press and opposition groups). This denunciation of official discourse is the product of the competitive relationship in which conspiracy theorists automatically feel caught up: because revealing a hidden truth means opposing those who wish and have the means to keep it hidden, the conspiracy discourse implies a confrontation between (at least) two mutually exclusive projects. It is therefore antinomic to a competing and dominant discourse; this makes it a counter-discourse, based on “stigmatized” knowledge that contradicts one form of orthodoxy or another (Barkun 2015: § 3). This mode of constructing conspiracy discourse requires it to be prepared to assume the burden of proof in the face of a doxa it intends to challenge. It thus corresponds to the definition of an argumentative discourse proposed by Marianne Doury (2003: §8):
We will consider argumentation as a way of constructing discourse to make it more resistant to contestation. For argumentation to exist, it is not enough for there to be a confrontation of opinions (“I love the music hall”, “I hate it”); there must also be a “crystallization of disagreement”, to use V. Traverso’s term (1999: 76), a crystallization that leads speakers to construct their positions and support them with propositions-arguments whose relationship to the conclusion determines specific modes of refutation (translated from French).
13An individual begins thus to argue when disagreement is likely to emerge. Therefore, the production of an argumentative sequence functions as an indication of disagreement with a stated thesis. One of the specific features of the conspiracy discourse is that it can offer an alternative reading of the “official” discourse and the doxa, so any event, even when its interpretation seems consensual, can be the subject of a competitive reading. In other words, any official or institutional discourse can become the object of contestation and see the emergence of an alternative discourse in disagreement, the “friction” between the two discourses favoring the opening of argumentative sequences.
14The production of such sequences can be analyzed at several levels. On the one hand, when produced as a counterpoint to elements of doxastic discourse, they are likely to introduce an alternative and total interpretation of events. The more consensual the doxastic data, the more plausible it is that the resources mobilized to contest them are conspiracy theories (for example, assuming that the Earth is not round requires all the world’s space agencies to be lying and therefore coordinating to plot together). Furthermore, since the “Earth is round” thesis is part of the doxa, it is not polemical a priori, so the conspiracy counter-discourse must make the “it’s a lie” disagreement explicit to develop its own analysis. On the other hand, the antinomic opposition between “official” and conspiracy discourses leads to epistemic competition that makes any form of concession or search for agreement more difficult, if not impossible. Caught up in this competition, one strategy used by conspiracy theorists to gain an advantage is demonstrating their ability to shed light on certain grey areas not dealt with by their opponents. “Conspiracy theories always explain more than competing theories, because by invoking a conspiracy, they can explain both the data of the received account and the errant data that the received theory fails to explain” (Keeley 1999: 119).
15The question of the superiority of one explanatory system over another therefore crystallizes around blind spots, what Brian Keeley calls “contradictory data” and “unaccounted-for data”. These are respectively elements of reality that contradict the “official” theory or are ignored by it because they are deemed embarrassing or inexplicable. David Zarefsky shows that conspiracy rhetoric is particularly effective when it can make sense of an ambiguity or inconsistency that is not otherwise resolved in doxastic or “official” discourse”: “Conspiracy arguments become widely accepted when they explain an otherwise ambiguous evil” (2014: 205). On the other hand, conspiracy rhetoric loses its effectiveness when there is a simpler or more accessible competing explanation for the same facts. In other words, one of the most effective rhetorical strategies used by conspiracy theorists is to point out contradictions, inconsistencies, and grey areas in opposing arguments to disqualify them, while proposing their own explanations based on an interpretation of events that leaves nothing to chance.
16Considering the highly argumentative dimension of conspiracy discourse and the existence of an argumentative strategy consisting of pointing out the inconsistencies of the opponent while providing an alternative resolution, we made the following hypothesis: in an interaction situation with an individual adhering to a conspiracy logic, the introduction into the discussion of linguistic elements from an “official” discourse is particularly likely to encourage the deployment of a conspiracy argument centered on the search for, denunciation of and strategic treatment of inconsistencies in this official discourse.
17The extracts analyzed in the rest of this article are borrowed from a corpus of interviews conducted in 2022 and 2023 as part of doctoral research. To identify elements of “conspiracy” discourse by comparing them with “non-conspiracy” discourse one needs the ability to compare, in the same corpus, discourses considered to fall into these two categories, without prejudging whether they belong to one or the other. To this end, we carried out a series of semi-structured interviews with people who take a critical view of the French government’s health policy in the context of the fight against the Covid-19 pandemic.
18This type of discourse is not in itself conspiratorial, but several studies have shown the links that can exist between opposition to vaccines and certain so-called “conspiracy theories” (Muric et al. 2021: 8). We can therefore assume that our corpus includes both types of discourse. The other challenge in compiling such a corpus of interviews was to bring together the conditions for the emergence of a conspiracy discourse in interaction. The interviewees can be reluctant to take on certain statements that might be identified as conspiracy theories by the researcher: there is, therefore, a risk that they would spontaneously censor their speech, aware that it might be despised or criticized, a fortiori in the face of a researcher representing scientific and academic institutions (Franks et al. 2017: 3). In order not to be denied access to what the interviewees have to say, the interviewer must act as a “third party” in the discussion (Plantin 2016: 527-528): he or she must not take sides with either of the two camps so as not to be identified as an adversary, while at the same time problematizing and highlighting possible disagreements in the form of questions or reported speeches to allow argumentative sequences to emerge. This positioning issue means the interviews should be conducted using a comprehensive approach to reassure the interviewee of the legitimacy of his or her discourse (at least initially) while avoiding offering it academic validation.
19The interviews are divided into two parts: the first part is open-ended, during which the respondent is invited to express his or her experience of the pandemic, his or her opinion of the health policy, and his or her assessment of it. In the second stage, the interviewer becomes more involved in the interview, acting as the spokesperson for a doxastic point of view. He also presents in the form of reported speech positions likely to be considered “official” in a conspiracy logic, prototypically those of the government or major international institutions (WHO, European Union). This second part of the interview seeks to create the conditions for the emergence of argumentative sequences that can be compared with each other.
- 3 In France, for example, non-vaccinated people were denied access to many public places (bars, resta (...)
20The first example illustrates a case of argumentation displaying none of the characteristics of conspiracy speech. The respondent (E1) points out an inconsistency in her opponent’s discourse but prioritizes the hypothesis of his stupidity to account for it and does not infer a conspiracy. The extracts analyzed, reproduced below, appeared in the interview after E1 had repeatedly expressed the importance to her of leaving everyone free to choose whether to be vaccinated. For her, the choice is a personal one. She has nothing against other people getting vaccinated but doesn’t want it imposed on her. For the health authorities, combating Covid-19 required mass vaccination, which meant introducing incentives, or even coercive measures, such as requiring a full vaccination pass for certain activities or to practice certain medical professions3. This argument has been used by the public authorities, and particularly the former Health Minister Olivier Véran, to criticize the choice not to vaccinate:
TR – what do you think then of an argument we heard, including from Olivier Véran, who said/ in the context of an epidemic, sanitary measures only make sense if they are massive/ and that here in this situation, typically, forms such as collective immunity or things like that\ only make sense if they affect a very large majority of the population
- 4 All the quotations borrowed from the interviews are translated from French.
E1 – I’d say to Olivier Véran: he’s got it all wrong/ eh\ ((laughs)) Actually, it’s not complicated eh you don’t need to be a doctor to say that there are plenty of facts that go against his analysis... ! the first thing I’d say to him/ you see if I had him in front of me\ I’d say frankly I thought he was more intelligent than that/ what/ and that he’s pretty rude... eh/!4
- 5 In this specific case, since the pandemic is global, there is no point in aiming for collective imm (...)
- 6 The distinction between ad personam and ad hominem has not been stabilized in the argumentation lit (...)
21In her response, E1 justifies her position that Olivier Véran “got it all wrong” in taking coercive measures to “force” the choice of French citizens, based on a contradiction between the justification put forward (only a strong incentive will make it possible to achieve a sufficient rate of vaccination coverage to roll back the pandemic) and “plenty of facts” 5 (which she will develop in the rest of the interview). This error in Olivier Véran’s analysis is presented as an obvious fact that needs no justification: “Actually, it’s not complicated.” It is the strength of this evidence that authorizes E1 to believe that Olivier Véran is discredited both as a doctor (he is a neurologist) and as an expert. E1 uses a negative axiology to describe him, together with an ad personam6 attack, deeming her opponent “not that intelligent” and “rude”. Immediately after this first extract, E1 develops further what she considers to be inconsistencies in Olivier Véran’s speech:
E1 – And I’ll them him, if I follow your reasoning /so/why/this pandemic which is a worldwide pandemic\ so why/ actually we do not take care of it on a worldwide scale\ why thus if/ you say that vaccination only matter if a majority of people get vaccinated\ it’s the solution/ (.) until now we have clearly we noticed that the virus would go further borders … ! what/ you see …with it we do not have a choice it is welcome at home …
22The inconsistency is raised by E1 in the form of a rhetorical question and presented as an error of logic, supported by markers of obviousness: the use of the deictic “until now” and the French indefinite pronoun “on” (translated by we) place Olivier Véran’s discourse at odds with common sense. The isolation of his analysis is reinforced by the repeated use of the modelizer “clearly” to emphasize interpretative disagreement (Perrin 2012: 64).
23The study of this first extract shows how having pointed out a double inconsistency in her opponent’s discourse (incompatibility with the facts and internal inconsistency), E1 infers an error of analysis. The discursive construction of the evidence of the error authorizes her to pass judgment on the intelligence of the person committing it and to conclude that he is stupid. The use of the past tense “I thought that” indicates the role played by the observation of incoherence in the re-evaluation of the opponent (Perelman 1989: 267). This reassessment of Olivier Véran is reconstructed in two stages: initially judged to be “intelligent”, it is when he is confronted with his contradictions that E1 concludes that he is irrational and therefore, by contagion, that his pro-vaccine policy is irrational. This conclusion is confirmed in the rest of the interview:
E1 – after the way in which they thought about this vaccinal campaign I had the impression that it was made in a hurry … in a hurry … somehow irrational … I mean/ I said to myself but wait but … those people do not … think properly … I mean/ they’re doing it too quickly/
24The discursive construction of an adversary taking decisions “in a hurry” and displaying irrationality makes it possible to evacuate the hypothesis of a “conspiracy” argument on the part of E1. On the one hand, the possibility of a conspiracy is rendered improbable because it would presuppose a form of control and anticipation of events (“nothing happens by chance”) as well as intentionality (“everything that happens is the result of hidden intentions or wills”). On the other hand, by building his argument around the idea of an error in his analysis of the situation (“he got it wrong”), E1 concedes to him a sincere desire to provide an appropriate response to the pandemic. The disagreement between E1 and Olivier Véran lies within the camp of those who wish to fight the virus but disagree on how to do so. They share the same goal, which leads E1 to be even more generous:
E1 –you had to try/ you have to try they’re right but I find that they … they did not target the right population …
25The concessive form “they’re right but” supports E1’s opponents on the need to develop a vaccine, while remaining critical of the methods used to distribute it.
26This articulation of discourse and counter-discourse does not correspond to a conspiracy argument, which has been described as based on an opposition between two antinomic historiosophies. On the contrary, the concessive form implies that the arguer must concede certain points to her adversary, which contradicts the claim to have a coherent and infallible grid for reading the world. Characterizing the opponent as irrational, emotional, or simply intellectually limited also contradicts the construction of a conspiracy discourse denouncing a malicious, hidden, well-organized, and manipulative power. In the interview with E1, the argumentative sequences deployed to make sense of the detected incoherence in the opponent’s discourse do not therefore take the expected forms of a “conspiracy” discourse, either in terms of categorizing the opponent or positioning vis-à-vis the counter-discourse.
« It’s always the same families »
- 7 The Lancet Gate is a scandal that took place during the Covid-19 pandemic. The Lancet medical journ (...)
27The following two cases illustrate an argument with discursive specificities that clearly resonate with some of the characterizations of conspiracy reasoning presented at the beginning of this article. In contrast to E1, the examples taken from interviews with E2 and E3 show the interviewees’ rejection of the stupidity hypothesis to explain the inconsistencies in their opponent’s discourse and infer the existence of a conspiracy. The first passage occurs at the beginning of the interview after the respondent (E2) has explained that the Lancet Gate7 was one of the starting points for his distrust of contemporary medical research. E2 criticizes the collusion of industry and pharmaceutical lobbies with the political world.
E2 – when you think that for other medicines such as remdesivir for example (.) two days before the European commission KNEW that it was ineffective and STILL France bought for two billions of it the day after (.) WHEN\ it was KNOWN/ that is was ineffective\ and dangerous
28The incoherence raised by E2 is at play here, between the knowledge of a drug’s ineffectiveness and dangerousness, on the one hand, and France’s decision to order it on a massive scale despite everything, on the other. Analysis of the paraverbal component of E2’s speech shows that he emphasizes the logical incoherence he is highlighting (accentuation of the connectors “still” and “when”). The contradiction is reinforced by the epistemic modalities: “knew” and “known”, which presuppose the truth of the object of knowledge, and are also emphasized. The prompting of the researcher, who plays the role of a doubtful third party (as evidenced by the use of the conditional), leads E2 to develop an argumentative sequence:
TR – and why would she do that/
E2 – the money from big pharma and there are a few families behind it/ (…) the money … it’s funny because it’s always those same families eh/ you ask then yourself why Kennedy was murdered because he wanted some money from/ the Rothschild banks and use it for the American state … and a few days later we found him assassinated there are two three other presidents that have been murdered when they wanted to start the same gesture … this it’s the money from the banks and the banks it is always the same families throughout the world so you see where everything goes smoothly it’s where there is no Rothschilds banks or relatives\
- 8 This antiviral drug is supposed to combat Covid-19. However, its efficacy has not been proven, or o (...)
29To make sense of the inconsistency between France’s decision to buy large quantities of remdesivir8 and the information at its disposal that the drug was ineffective, E2 proposes that this inconsistency be seen as evidence of the influence of a malevolent group with a vested interest in enriching itself. In other words, the decision to order this drug on a massive scale is not inconsistent with its presumed known ineffectiveness; it can be explained by the existence of a hidden variable: corruption or manipulation by “a few families”. To justify the existence of this variable, E2 argues from precedent and constructs a continuity between the present situation and past historical events. The repetition of the modelizer “always” makes it possible to attribute the same cause (money deals involving the Rothschild family and banks) to two events (the purchase of medicines today, and the assassination of American presidents in the 1960s). Both events must therefore be regarded as equally reprehensible acts: the indisputably criminal nature of the Kennedy assassination is transferred to France’s decision to purchase remdesivir on a massive scale. E2 then extends this Manichean recategorization operation to the major international geopolitical dynamics with a double present of general truth: “where everything goes smoothly it’s where there is no Rotschild’s banks or relatives”.
30This discourse corresponds prototypically to what might be described as a conspiracy resolution of inconsistencies, systematically attributing to the world’s problems the same hidden, large-scale organized, and malevolent causality. Malicious intent is also concealed behind an official, doxastic discourse, as in this extract where E2 explains that he visits “re-information” websites:
E2 – it allows you to think and open your brain to something other than the doxa/ that comes from those who are currently in power and who have the money
TR – hm hm, so that’s how it came to you? gradually?
[…]
E2 – there were a few things that made… I mean… this story about hydroxychloroquine and the: Lancet gate/ and when you see that the Lancet belongs to important Jewish families... and the banks too and other things and all that... suddenly… and some... Buzyn, her husband, Lévy, the head of the laboratory… all these people THERE// hold one and the same discourse
31Here, E2 takes an explicit stand against a doxastic discourse coming from above (“the doxa/ that comes from those who are currently in power and who have the money”) and which he reconstructs as monolithic (“all these people THERE// hold one and the same discourse”). Since this discourse is flawed and inconsistent (the Lancet’s error in the hydroxychloroquine affair), E2 infers an attempt at manipulation on the part of “important Jewish families”. This conspiratorial historiosophy revives the anti-Semitic imaginary of the “Jewish plot” (Agnès Buzyn and her husband are also both of Jewish faith), which attributes responsibility for history’s great misfortunes to the Jewish community (Poliakov 2006: 44-45, Taguieff 2021).
32This third abstract comes after the respondent (E3) has described the vaccine as “toxic”, prompting us to ask her why the government would have chosen to encourage the population to receive a supposedly dangerous injection:
TR – but why do this?
E3 – they don’t care/ they don’t care if there are deaths/ (.) well, we would know about it if they did care/ (.) like all states that go to war, I’m sorry but we’re not trying to protect the population// normally the first thing you have to do/ is... fuck/ leave the masks on you’re getting the factory back up and running/ there, they are closing the factory the year before anyway/ the last factory that makes masks it is not normal\
- 9 The mention of “war”, which recurs several times in the interview, is an explicit reference to the (...)
33In this extract, E3 asserts her position (“they [= the French government] don’t care if there are deaths”]), which she reinforces through repetition and argues through an argument from ignorance (there is no proof to the contrary of p, so p is true). Her second line of argument is based on a comparison (we know that in a war situation, all states are indifferent to the need to protect the population; we are in [a kind of] war situation, so the authorities don’t try to avoid taking measures that could cause deaths). Finally, the third line of argument in support of her conclusion is based on the observation of an inconsistency between the seriousness of the health situation and the decisions taken by the French government9, in this case, the fact that the last mask production factory in the country was closed a year before the pandemic. The perceived evidence of incoherence is constructed by references to normality (“normally”, “it is not normal”), and by the discursive contrast between what should have been done (“you’re getting the factory back up and running”) and what was actually done (“there, they are closing the factory the year before”) – we note the association of the interlocutor with the approach that should have prevailed through the use of a generalizing “you” (“you’re getting the factory back up”), and the dissociation from the actors responsible for the incriminating measure (“there, they are closing the factory the year before”). It is based on this discrepancy that E3 builds her argument:
- 10 Renseignements Généraux: former name of the French domestic intelligence service, now the DGSI.
E3 – how do you want me to explain this? I mean guys who know exactly... who knows everything/ they know everything/ the RG10, things and all... you’ve to stop\ you can’t be in a position of responsibility like that and not know what’s going on\ guys know what’s going on \
34Here we observe that for E3, the absence of an explanation for what she perceives as an inconsistency (since she rules out error of judgment by asserting that those in charge knew what should have been done to avoid deaths) leads her to infer a form of conscious malice on the part of the government: “they don’t care”, “they’re not trying to protect the population”. This example illustrates how an inconsistency can only be resolved in one of two ways: either by assuming that the authorities are incompetent (an assumption here explicitly rejected by E3), or by restoring a form of coherence to government action by introducing a secret factor or hidden interests:
E3 – for me, it’s a total manipulation from A to Z (.) to the detriment and by sacrificing a part of the population... the elderly. Still… quite an advantage when you want to stop paying for...
TR – pensions?
E3 – take over pension funds which are still... where there’s money/ eh/ that’s where the money is, in the pension funds/ it’s crazy/ I mean, I find it... I’m hallucinating/
35The incoherence is resolved through conspiracy: the manipulation, which implies a form of malevolence towards the victims, is described as “complete” and justified by the unavowable financial interest of letting the elderly die so as not to pay their pensions. Conscious malevolence is constructed as the only alternative resolution to stupidity, the latter being declared impossible due to the level of responsibility and efficiency attributed to the secret services. The logical necessity of concluding that there is a conspiracy also seems to be reflected in E3’s expression of a form of astonishment, as she is “hallucinating” by her own deductions: “It’s crazy”. Once again in this passage, we find several totalizing procedures: the use of the modelizers “everything”, “from A to Z”. The rhetorical question “how do you want me to explain this?” suggests the absence of any alternative explanation; alternative resolutions to manipulation (incompetence, ignorance) are firmly rejected; finally, the fact that the government’s malevolent project extends (at least) over a year implies that it is anticipated and therefore intentional, further reinforcing the idea that it cannot be a question of bad decisions due to the incompetence of the authorities. Together, these features contribute to the argumentative closure of E3’s discourse.
36Our analysis of the various ways in which inconsistencies are treated in the discourse of our respondents allows us, based on the characterizations of conspiracism and the associated markers, to identify an initial entry point to discursive strategies likely to distinguish between “conspiracy” and “non-conspiracy” argumentation. However, this distinction needs to be tempered. Indeed, while the statements by E3 that we have analyzed here present several characteristics that tend to categorize them as conspiracy talk, other moments in the interview put these conclusions into perspective. To illustrate these moments, we have chosen one passage in which categorizing the adversary brings E3 closer to E1’s discourse. In this extract, E3 refers to Emmanuel Macron’s psychological fragilities:
E3 – in terms of personality, soul, and state of mind, I mean, he’s just like everyone else, this guy/ well, for me, whether you’re president or not, you’re not a superman/ and you’re not above ground/ well, you’re part of the planet and part of the human race\
37In this example, E3 deconstructs the operation of categorization she had previously carried out by constituting the French state (the government and its intelligence services) as an infallible group, unable to ignore anything about the health situation. The outspoken conspiracy presented here diverts from the more frequent definition of conspiracism: the conspiracy is not the result of an almighty and eternal group able to control history, but the mere result of a human being like the others. This example in no way detracts from the conspiracist nature of the previous statements analyzed from E3, but it does nuance the definition of conspiracism as a total and coherent reading of the world. We can thus suggest that conspiracism can function as an ad hoc hermeneutic device, or even as an opportunistic rhetorical strategy. In an interactional set-up where the interviewer plays the role of the third party and carries the government’s arguments, opening a conspiracy argumentative sequence polarizes the discussion by excluding any conclusion other than the interviewee’s. The processes of totalizing and generalizing the conclusion serve to disarm compromise proposals and thus the moderate position of the third party.
38A comparative study of three interviews leads to several conclusions. First, the observation of language markers linked to certain traits classically associated with conspiracism makes it possible to distinguish a conspiracy argument from a non-conspiracy argument. However, these markers are only observable in specific argumentative moments (in reaction to exposure to the opponent’s discourse), which enable the arguer to identify inconsistencies and seek to resolve them by making sense of them. The conspiratorial argumentative sequence then consists of discarding the incompetence hypothesis in favor of the conspiracy hypothesis.
39However, this mechanism needs to be clarified. On the one hand, the way in which the speaker takes charge of the reconstruction of explanatory variables (stupidity or conspiracy) is highly dependent on the context of enunciation (presence of the interviewer as a third party, request for explanations, prior trusting of the interviewee...). On the other hand, the interview framework, which allows discourse to unfold over time, shows that conspiracy “moments” can coexist with long developments on the same subject but without any of the markers usually associated with conspiracy discourse. The possibility of the coexistence of such statements allows us to formulate the hypothesis that conspiracy discourse is not always the manifestation of a reading framework of the world that runs through and conditions the discourse, but that it can be a punctual, even opportunistic, argumentative strategy. By identifying an incoherence, the arguer can assume a logical error, yet the accusatory power of such an error is not always sufficient to discredit or disqualify an opponent (after all, everyone can make mistakes). By denying the error and constructing a coherent, logical, and flawless adversary, the conspiracy argument strengthens its accusation by offering proof that its opponent is involved in a criminal conspiracy. This strategy could thus be perceived as a reverse strategy of the strawman which, instead of presenting the opposite discourse as ridicule, presents a discourse very well organized to demonize it.
40This punctual use of conspiratorial argumentative strategies could take part in the explanation of the problem presented in the introduction: the false positives. If only certain conspiracy discourses present some features of conspiracism, it is because conspiracism emerges only in some specific enunciative conditions, and therefore – we can assume – only for specific argumentative purposes.
41The study of historical conspiracy theories usually presents them as pamphlets, a polemical genre that allows argumentation to unfold throughout the discourse. This bias might have led, sometimes abusively, to the idea that conspiracism produces too coherent discourses that leave no place to chance or to doubt.
Flux RSS
