Introduction
The term hate speech gained prominence in the early 1980s through the work of U.S. legal scholars associated with Critical Race Theory (CRT), who sought to expose the systemic racism embedded within American society and its legal institutions. Rather than limiting hate speech to isolated expressions of individual hostility, these scholars framed it as a discursive mechanism that sustains racial hierarchies and marginalises minority groups (Delgado & Stefancic, 2017, pp. 3–10).
Building on this foundation, legal definitions have attempted to operationalise the concept, with one of the most widely cited provided by the European Commission against Racism and Intolerance (ECRI). Hate speech is defined as:
The use of one or more particular forms of expression–namely, the advocacy, promotion or incitement of the denigration, hatred or vilification of a person or group of persons, as well as any harassment, insult, negative stereotyping, stigmatization or threat…based on a non-exhaustive list of personal characteristics or status…including “race”, colour, language, religion or belief, nationality or national or ethnic origin, as well as descent, age, disability, sex, gender, gender identity and sexual orientation.
Despite its influence, this definition has drawn criticism for its semantic and legal ambiguities. Faloppa (2020) notes that it remains unclear whether all behaviours listed under the definition warrant equal condemnation, leaving room for interpretative inconsistencies. Efforts to refine the concept within national contexts, such as in Italy, have encountered similar challenges. The Jo Cox Parliamentary Commission on Hate, Intolerance, Xenophobia, and Racism (2016–2017) expanded upon the ECRI framework by introducing the pyramid of hate model, conceptualising hate as a progressive process: beginning with shared stereotypes, escalating through systemic discrimination, and culminating in verbal aggression and violence (Commissione Jo Cox, 2017). However, despite its dynamic structure, the model has been critiqued for insufficiently differentiating between forms of expression at various stages and for its underlying assumption of a linear, cumulative progression in the development of discriminatory behaviours (Faloppa, 2020, p. 31). These examples reflect the broader struggle to define hate speech in a manner that captures its complexity across legal, social, and communicative domains.
In addition to the limitations of existing policy frameworks, further complications arise from the conceptual underpinnings of the term itself. The phenomenon poses significant challenges, and prevailing definitions often fail to account for the varied motivations behind hate acts or the differing experiences of targeted groups. Furthermore, the very terms ‘hate’ and ‘speech’ can be misleading: ‘speech’ suggests a focus on oral communication, while ‘hate’ implies that hostility stems solely from personal emotions, obscuring the structural and strategic dimensions of discriminatory discourse (Faloppa, 2020, pp. 26–27; Perry, 2001, p. 10).
In Italy, hate speech entered public discourse more widely in the early 21st century, though it remains absent from most standard dictionaries. The online dictionary Treccani offers one of the few entries, defining it as: ‘Espressione di odio rivolta, in presenza o tramite mezzi di comunicazione, contro individui o intere fasce di popolazione’ (Hate Speech, Treccani, 2018). However, this definition remains imprecise in specifying the forms of expression involved, the full spectrum of social groups and personal characteristics that may be targeted, and the various media channels through which hate speech can be disseminated. This imprecision reflects the ongoing conceptual ambiguities that continue to surround the term.
This study doesn’t aim to propose an alternative definition of hate speech; rather, it investigates how its discursive dynamics manifest within a specific case study: La Zanzara, a talk radio programme broadcast by Radio24 and hosted by Giuseppe Cruciani, David Parenzo, and, until 2021, Alberto Gottardo.1 The programme adopts a deliberately provocative style, presenting itself as a space where socio-cultural and political opinions can be voiced without censorship. A defining feature of its self-presentation is its explicit opposition to political correctness, a stance that positions its discursive practices in contrast to mainstream media norms and facilitates the expression of controversial or marginalised viewpoints.2
This analysis adopts a linguistic perspective on hate speech, with particular attention to the role of DMs in constructing and modulating hostile discourse. Instead of concentrating on explicit lexical items or overtly derogatory language, the study foregrounds pragmatic features that subtly influence tone, stance, and interactional dynamics. In doing so, it seeks to demonstrate how hate speech may be mediated through interactional cues that enhance both its persuasiveness and its social acceptability.
The analytical framework draws on a pragma-dialectical rhetorical approach, with a particular focus on the informal fallacy of the ad hominem argument (Iacona, 2010, p. 133; Yap, 2013, p. 99). This fallacy occurs whenever a speaker targets the personal characteristics or circumstances of an interlocutor rather than engaging substantively with their argument. By applying this framework, the study offers a rhetorical interpretation of the ECRI’s definition of hate speech, situating personal denigration within broader argumentative structures designed to discredit or marginalise opponents. The central research question guiding this inquiry is as follows: How and to what extent do DMs contribute to the construction of hate speech?
To address this question, the analysis examines the use of DMs by both hosts and audience participants in La Zanzara, focusing on how these markers facilitate rhetorical aggression, ideological positioning, and interactional alignment. Given the programme’s confrontational ethos, the central hypothesis is that DMs extend beyond their conventional role in managing conversational flow, functioning additionally as instruments for sustaining in-group cohesion, excluding opposing perspectives, and regulating power dynamics within the interaction.
The empirical scope of this study is limited to broadcasts from the second week of March 2020, a period during which Italy was under national lockdown due to the COVID-19 pandemic. This timeframe was selected on the premise that heightened social tensions and ideological conflicts associated with the pandemic would amplify discursive confrontations within the programme. By concentrating on this context, the study explores how DMs interact with rhetorical strategies, particularly ad hominem fallacies, to intensify and normalise hate speech. Therefore, the analysis seeks to contribute to a deeper understanding of the role of DMs in the enactment of hateful discursive processes, highlighting the complex interplay between language, ideology, and mediated public discourse.
In this article, the term offensive speech is used as a broader pragmatic category that includes, but is not limited to, forms of hate speech as defined by institutional frameworks such as ECRI. While hate speech provides a key conceptual foundation — particularly in its legal and political dimensions — the analysis also draws on offensive speech to account for a wider range of rhetorical and interactional strategies. These include mockery, exclusion, and personal denigration, which may not meet formal legal criteria, yet still contribute to symbolic violence and ideological polarisation.
Theoretical background
Hate speech in the linguistic paradigm
Hate speech has been widely studied across disciplines, with linguistic research increasingly framing it as a form of symbolic violence and as a means of constructing and reinforcing social hierarchies through language (Bianchi, 2021; Fairclough, 2013, 2015). This study contributes to this body of work by examining the linguistic mechanisms that sustain hateful discourse, focusing particularly on DMs that enhance its impact. While previous research has largely concentrated on lexical features, this analysis aims at integrating pragmatic and discourse-analytical perspectives to offer a more dynamic understanding of hate speech.
Within Italian linguistics, considerable scholarly attention has been devoted to the ways in which language perpetuates gender-based inequalities. Foundational studies by Sbisà (1976), Sabatini (1987), and Lepschy (1987) critically examined gender biases embedded within grammatical and lexical structures, demonstrating how language reflects and reinforces broader patterns of social discrimination. This line of inquiry was subsequently expanded and refined by later scholars, notably Sapegno (2010) and Baldi (2023a), who continued to focus on the mechanisms of sexism in language. In parallel, De Mauro’s (2016) classification of parole per ferire – terms that target ethnicity, geographic origin, disability, or moral character – broadened the understanding of how ostensibly neutral words can acquire derogatory connotations, contributing to a broader reflection on linguistic discrimination beyond the domain of gender.
Following these foundational investigations into gendered language, scholarly attention increasingly turned to other forms of linguistic discrimination. Research began to address racialised and xenophobic discourse. Faloppa (2000, 2004, 2015, 2020) systematically documented the lexicons of racism and otherness in Italian, while Cepollaro (2015a, 2015b) explored the semantic-pragmatic dynamics of pejorative epithets, illustrating how such terms function within specific interactional contexts to reinforce exclusion.
More recently, scholars have examined the intersection of hate speech and political discourse, particularly within populist rhetoric. Studies by Dell’Anna (2016), Gualdo (2022), Ondelli (2014, 2017, 2018, 2021), Petrilli (2020), Trifone (2022), Ambrisi (2021) and Cortelazzo (2024) reveal how populist communication employs simplified language, binary oppositions, emotive appeals, and metaphorical framings to construct out-groups and legitimise exclusionary ideologies (Baldi, 2023b). These mechanisms are further intensified by the spectacularisation of political discourse (Novelli, 2006; Gualdo, 2022) and morphological innovations that heighten the emotional resonance of exclusionary rhetoric (Cedroni, 2017).
This interplay between language, politics, and ideology is especially evident in the personalised discursive strategies of individual politicians. Analyses by Antonelli (2017), Ondelli (2017, 2018), Baldi et al. (2019) and Baldi (2022, 2023b), and Combei et al. (2022) demonstrate how specific lexical and pragmatic choices shape political messaging, resonating with audiences while embedding discriminatory narratives. Cortelazzo’s (2024) observatory on neopolitical language further tracks these evolving patterns, particularly within media-saturated environments.
Although much of this scholarship has concentrated on lexical features, recent work has advocated for a more integrated approach. Bianchi (2021), drawing on speech act theory (Austin, 1975) and performativity (Butler, 1997), emphasises that hate speech not only transmits harmful content but also enacts social hierarchies,3 reinforcing exclusionary mechanisms through illocutionary force.
The digital sphere introduces additional complexities, as discriminatory discourse circulates rapidly across online platforms, transcending traditional media boundaries (Citron et al., 2011; Citron, 2014; Brown, 2018). Italian linguists have increasingly addressed these dynamics. Gheno (2018) identifies neologisms, altered spellings, and abbreviations that bypass content moderation while fostering in-group solidarity. Similarly, Bonomi and Morgana (2020) document stylistic markers such as irregular punctuation, unconventional capitalisation, and politicised language that both express group identity and signal ideological alignment. These findings align with broader research on toxic online communication (Grotti, 2024).
Beyond stylistic and lexical choices, digital environments also reshape the pragmatic dimension of hate speech. The anonymity and reduced accountability typical of online spaces (Suler, 2004) intensify illocutionary force, facilitating silencing and illocutionary distortion, which marginalise certain voices (Bianchi, 2021). These conditions reveal the performative power of hate speech in sustaining discursive injustices. Despite moderation policies, discriminatory language adapts rapidly, often evading detection (Gagliardone et al., 2015), underscoring the need for continued linguistic inquiry into its textual and contextual variability.
Adopting a pragmatic and discourse-analytical lens, this study addresses hate speech by focusing on DMs, an often overlooked but crucial element in shaping its structure, stance, and emotional resonance. DMs modulate argumentative flow, enhance in-group cohesion, and amplify illocutionary force, playing a pivotal role in the reproduction of social hierarchies and the normalisation of exclusionary ideologies.
While this study does not fully adopt the methodological framework of Critical Discourse Analysis (CDA), it draws selectively on its conceptual foundations — particularly those developed by Van Dijk (1993, 2006), Fairclough (2013; 2015), and Wodak (2015a, 2015b, 2019) — to explore how language contributes to the reproduction of power relations and the discursive construction of group boundaries. Rather than conducting a comprehensive CDA in the sense of genre and interdiscursive analysis, the focus here is on pragmatic interactional mechanisms, especially DMs, that modulate stance, alignment, and the rhetorical force of offensive speech in political media contexts.
By foregrounding these interactional mechanisms, the analysis aims to offer a more nuanced account of hate speech, one that moves beyond content to examine the dynamics that sustain its impact.
The pragma-dialectical approach to the argumentum ad hominem
The study of argumentation spans disciplines ranging from classical philosophy to contemporary communication theory. This research adopts an informal framework, not to reject formal paradigms, which remain valuable for analysing argumentative structures, but to address their limitations in accounting for the multifunctionality of DMs. In this perspective, interpreting DMs within argument construction requires integrating pragmatic and contextual dimensions.
This approach draws on the pragma-dialectical model developed by van Eemeren and Grootendorst, who define argumentation as ‘a verbal, social, and rational activity aimed at convincing a reasonable critic of the acceptability of a standpoint by putting forward a constellation of propositions justifying or refuting the proposition expressed in the standpoint’ (van Eemeren & Grootendorst, 2004, p. 1). Thus, argumentation is understood as a complex speech act: verbal in its form, social in its interaction, and rational in its appeal to shared reasoning.
The pragma-dialectical approach integrates both descriptive and normative dimensions. Descriptively, it situates argumentation within speech act theory, whereas normatively, it emphasises standards of reasonableness that validate argumentative moves. Arguments, understood as ‘constellations of propositions’, are embedded within communicative contexts and gain value only when oriented towards resolving a difference of opinion. As a form of ‘normative pragmatics’, this model stresses contextual analysis, particularly in identifying implicit premises shaped by interactive norms and background knowledge (van Eemeren & Grootendorst, 2004, pp. 1–3).
A central feature of this framework is its treatment of fallacies. Whereas formal logic defines fallacies as structurally flawed but superficially valid arguments, the pragma-dialectical model reframes them as violations of the procedural norms that govern the resolution of disagreements. In this view, a fallacy obstructs the resolution process itself (van Eemeren & Grootendorst, 2004, p. 162). Particularly relevant to the present research is the ad hominem fallacy, which arises whenever a speaker targets an opponent personally rather than addressing the substance of their argument. This strategy seeks to discredit claims by focusing on personal characteristics rather than on content (Iacona, 2010, p. 133; Yap, 2013, p. 99). Such moves also violate the ‘freedom rule’, which stipulates that ‘discussants may not prevent each other from advancing standpoints or from calling standpoints into question’ (van Eemeren et al., 2009, pp. 85–86; Yap, 2013, p. 100). Within this framework, three subtypes of the ad hominem fallacy are distinguished: the ‘abusive variant’, involving direct personal attacks; the ‘circumstantial variant’, which questions an opponent’s motives or interests; and the ‘tu quoque variant’, which highlights inconsistencies in the opponent’s behaviour or statements. Each of these undermines credibility and shifts attention away from substantive issues, thereby impeding the resolution of argumentative discussions (van Eemeren & Grootendorst, 2004, p. 177).
The discourse markers
This study investigates the construction of hate discourse in Italian by analysing the pragmatic use of DMs in a radio programme. Adopting a functional approach, it draws on Bazzanella’s framework, first outlined in the Grande grammatica italiana di consultazione (1995) and refined in later works (2001, 2006a, 2006b). According to Bazzanella, DMs are linguistic elements with stable, though often attenuated, meanings that fulfil multiple roles in structuring discourse and regulating interaction. They contribute to textual cohesion by linking phrasal, interphrasal, and extralinguistic elements, and they position utterances within the interpersonal dimension of communication (Bazzanella, 1995, p. 225; 2006b).
Bazzanella categorises DMs’ multifunctionality into two principal domains: interactive and metatextual. Interactive functions convey stance and facilitate conversational dynamics such as turn-taking, phatic bonding, fillers, hedges, and managing attention and comprehension. Metatextual functions organise discourse internally, including structuring, reformulation, and focus (Bazzanella, 1995, pp. 225, 232–249; 2006b). These roles are summarised in Tables 1 and 2, 4.
Table 1
Interactional functions of DMs.
| SPEAKER | ADDRESSEE |
|---|---|
| Turn-taking devices | Interrupting devices |
| Fillers | Back-channels |
| Attention-getting devices | Attention confirmed |
| Phatic devices | Phatic devices |
| Hedges and boosters | / |
| Checking comprehension | Comprehension confirmed; requests for clarification |
| Requesting agreement, confirmation | a. Agreement, confirmation, support; b. Partial or complete disagreement |
| Yielding the turn | / |
Table 2
Metatextual functions of DMs.
| Textual markers | 1. Structuring the parts 1.1. Introduction (as a frame device) 1.2. Transition 1.3. List 1.4. Digression 1.5. Ending 2. Quotation and indirect speech markers |
| Focusing devices | 3. Local 4. Global |
| Reformulation markers | 5. Paraphrase markers 6. Correction markers 7. Exemplification markers |
Bazzanella’s framework is well-suited for this study both conceptually and methodologically. Conceptually, its focus on the situated, performative realisation of DMs, rooted in Saussure’s parole, aligns with this study’s emphasis on context-specific discourse practices. Methodologically, it offers the flexibility to capture the DMs’ multifunctionality in Italian, where such markers are pervasive across spoken and written registers. The model has informed numerous empirical studies and remains a central reference in Italian discourse analysis (e.g. De Cristofaro et al., 2022; Fiorentini & Sansò, 2017; Ghezzi, 2018; Mastrantonio, 2019; Sansò, 2020; Thaler, 2016).
Within her compositional-pragmatic model, Bazzanella identifies propositional and distributional features of DMs. A DM’s propositional exteriority refers to its syntactic and semantic detachment from the sentence: it does not affect truth conditions, nor can it be questioned or negated. This is evident in partial inter-substitutability, whereby different DMs may fulfil similar pragmatic functions despite lexical differences. Distributionally, DMs exhibit positional flexibility, occurring across the sentence and forming chains with overlapping or complementary roles. Shifts in position or prosody often entail functional shifts, underscoring prosody’s role in meaning-making (Bazzanella, 1995, pp. 228–231; 2001, p. 45; 2006b).
Bazzanella also incorporates the concept of ‘contextual anchoring’, distinguishing between the ‘a priori context’, defined by sociolinguistic variables, and the ‘activated context’, which emerges during real-time interaction and includes cotextual as well as extralinguistic factors (Bazzanella, 2001, pp. 47–48; 2006a). In alignment with scholars such as Schiffrin (1987) and Mosegaard Hansen (2006), she emphasises the importance of considering prosodic and paralinguistic cues in cotextual analysis. She further divides the activated context into two dimensions: the ‘extralinguistic dimension’, which involves physical objects and actions present in the interaction, and the ‘interactional dimension’, referring to discourse sequencing and emotional dynamics. Together, these factors illustrate how both structural and contextual parameters shape the use and interpretation of DMs (Bazzanella, 2001, pp. 48, 52; 2006b).
This comprehensive approach makes Bazzanella’s framework especially apt for analysing hate discourse, where the interplay of interactional cues and argumentative strategies is key to understanding how hostility is constructed and sustained.
The Hate-pandemic corpus
Methodology and corpus of analysis
This section outlines the data selection and methodological procedures for analysing five episodes of La Zanzara, broadcast on Radio24 between 9–13 March 2020, a politically and socially charged period. These episodes were selected for their thematic relevance and high audience participation, providing a representative sample of the programme’s discursive practices.
Transcripts were initially generated using the automated speech-to-text tools TurboScribe and Happy Scribe. Given the limitations of automated transcription, particularly with overlapping speech, dialectal variation, and informal registers, a thorough manual revision was subsequently undertaken. This revision process followed three principal stages:
Error correction: addressing inaccuracies in speaker identification, punctuation, and segmentation.
Standardisation: correcting typographical errors and normalising orthography while preserving features of informal speech, such as phonetic reductions.
Annotation: classifying DMs according to their pragmatic functions, based on an adapted annotation scheme from established Italian linguistic classifications (Bazzanella, 2006b; Ghezzi & Molinelli, 2014; Sansò, 2020).
The finalised transcripts, prepared in both plain text and XML formats to ensure compatibility with corpus analysis tools, were securely archived on the Sunet Drive provided by Stockholm University. These files are accompanied by comprehensive documentation detailing the transcription guidelines, annotation scheme, and data storage protocols, ensuring that future researchers have access to all necessary resources for replication, verification, or further enquiry. Both transcription and annotation were carried out solely by the author, adhering to a rigorously developed protocol to ensure consistency and methodological transparency.
The annotated data were analysed using a qualitative, interpretative approach that prioritises linguistic practices and interactional dynamics over quantitative measures. This pragmatic perspective explores correlations between the use of DMs and speakers’ control over the progression of debate. In line with Bazzanella’s framework, DMs are shown to play a central role in structuring interaction by linking phrasal, interphrasal, and extralinguistic elements, while positioning utterances within the interpersonal dimension of communication.
The show, on air since 2006, is renowned for its provocative engagement with contemporary political and socio-cultural issues. Its title references a historic student newspaper from Milan’s Liceo Parini, which gained notoriety in 1966 for publishing La posizione della donna nella società italiana, an article featuring female students’ perspectives on premarital relationships and sexuality (Schiavi, 2016). This reference underscores the programme’s intent to challenge societal norms and provoke debate, encapsulating its deliberately confrontational, unsettling style aimed at stimulating reflection on contentious topics.
The analysis includes contributions from the three hosts as well as audience members participating via live calls and voice messages. Audience identities were anonymised to ensure ethical compliance. The hosts’ ideological positions significantly shape the programme’s discursive dynamics: Giuseppe Cruciani, the main host, is known for pro-right-wing populist views; David Parenzo aligns with upper-left bourgeois positions; and Alberto Gottardo mediates between populist and leftist perspectives.
Analysis
The following section presents and discusses the findings from the linguistic analysis, organised into two parts. The first part addresses DMs that are integral to the nature, construction, and content of argumentation. The second part examines markers specifically associated with the expression of hate speech, with particular attention to the phenomenon of ad hominem arguments.
To highlight the multifunctionality of each marker, I have classified them into their respective functional categories and provided concise definitions. Illustrative examples drawn directly from the analysed texts are included to demonstrate their usage.
To ensure analytical consistency, each DM is examined through a recurring structure that includes a discussion of its pragmatic function based on existing literature, followed by one or more illustrative examples, and a critical analysis linking its use to the interactional context and broader ideological dynamics. This approach is designed to facilitate comparability across markers and to clarify the relationship between theoretical framing, empirical data, and interpretative insight.
The role of DMs in argumentation
Reformulation devices
Cioè
In Italian spoken discourse, ‘cioè’ is widely recognised as a highly polyfunctional DM, central to reformulation, elaboration, and interactional management. Traditionally conceptualised as a metatextual device indicating clarification or exemplification, it signals the speaker’s intent to rephrase, explain, or expand upon prior content, thereby enhancing textual cohesion and mutual understanding (Bazzanella, 1995, 2006a, 2006b; Ghezzi, 2018; Sansò, 2020).
Beyond reformulation, it mediates shifts in epistemic stance, reframing discourse through shared knowledge and inferential reasoning (Rossari, 1997). Furthermore, it functions as a cognitive planning device, marking hesitation and processing during speech production (Bazzanella & Miecznikowski, 2007), and serves to modulate utterances through strategies of boosting or hedging. These multifunctional roles underscore its significance as an interactional resource in Italian conversation.
Within the analysed dataset, ‘cioè’ appears frequently in audience-generated voice messages, where it plays a key role in managing emotional intensity and discourse coherence. While traditionally associated with reformulation and clarification, in La Zanzara it functions as both a cognitive placeholder and an emotional intensifier. Its repetition links fragmented utterances, enabling speakers to retain control over speech production and structure affect-laden content. This strategic use of ‘cioè’ underscores its value as an interactional resource in managing stance, coherence, and engagement in spontaneous, high-intensity discourse.
Empirical Example
The strategic use of ‘cioè’ in emotionally charged audience messages demonstrates its role in structuring discourse and managing affective intensity. Example (1) illustrates this pattern, with the marker appearing repeatedly to organise fragmented speech and maintain interactional control.
| (1) | Voice message: Cioè come cazzo si fa a pensare che se uno è giovane allora non muore, quindi chissene frega, non è un problema. Cioè contagia almeno altre 10 persone, tra cui vecchi, gente che ha bisogno di cure. Cioè quello è proprio un deficiente, perché ci crede cazzo, sembra proprio che non lo faccia neanche apposta. Cioè ci crede che se lui non muore, che forse è l’unico che dovrebbe morire, per quanto mi sta sui coglioni che non si capisce una minchia, allora non muore lui va tutto bene. Ha ragione Parenzo non pensavo di dirlo nella mia vita. Ha ragione Parenzo, cazzo. Arrestiamo Gottardo, arrestiamo Gottardo! (La Zanzara, 11-03-2020) |
In Example (1), ‘cioè’ operates not only as a reformulative marker but also as a cognitive and emotional anchor. Its repeated use connects fragmented segments of speech, sustaining cohesion where narrative structure becomes unstable. By layering clarification and emphasis, the marker enables the speaker to manage coherence while navigating high emotional intensity.
Analysis
These findings are consistent with previous research that identifies ‘cioè’ as a key marker of reformulation and cognitive processing in real time (Bazzanella, 1995, 2006a; Ghezzi, 2018; Sansò, 2020). However, they also extend existing insights by highlighting its role in asynchronous, user-generated formats such as voice messages. In these contexts, the absence of immediate listener feedback creates communicative gaps that ‘cioè’ helps to bridge, enabling speakers to monitor their discourse, manage simulated turn-taking, and sustain coherence in extended monologic sequences.
This study builds on earlier analyses that described ‘cioè’ as a filler or reformulative marker, while engaging more recent accounts of its multifunctionality in signalling spontaneity, affect, and argumentative structure. By focusing on informal digital discourse—particularly in environments characterised by delayed or absent feedback—it contributes to a broader understanding of ‘cioè’ as a flexible and adaptive discourse resource. The findings reaffirm its central role in Italian interaction, while demonstrating its relevance across evolving communicative settings.
Modulation devices
DMs functioning as hedges and boosters modulate utterances by attenuating their force or precision, enabling speakers to hedge, express uncertainty, or reduce epistemic commitment. This role is particularly important in discussions involving sensitive or complex topics, as speakers balance caution, politeness, plausible deniability, and argumentative force (Bazzanella, 1995, pp. 238–240; 2006a; 2006b; Sansò, 2020). These DMs are typically realised through modal adverbs that function interpersonally, mitigating speaker stance while managing audience expectations regarding the reliability of statements.
In the corpus analysed here, modulative DMs appear frequently across both host and caller speech, often clustered around moments of heightened emotional intensity or ideological tension. Their strategic placement—particularly in utterance-initial or final position—enables speakers to manage audience expectations, mitigate the risks of face-threatening acts, and maintain discursive legitimacy in contentious exchanges.
Empirical Examples
The following examples illustrate how modulative DMs are deployed by both hosts and callers in La Zanzara to navigate ideological tension, manage accountability, and position speaker stance with varying degrees of certainty.
In Example (2), Cruciani uses ‘evidentemente’ to generalise and reduce the strength of his epistemic commitment, as illustrated in the following excerpt:
| (2) | Ho capito ma ragazzi evidentemente qualcuno se se se [sic] lo scrive pure una persona, non mi ricordo chi… (La Zanzara, 9-03-2020) |
Here, ‘evidentemente’ permits inference without full endorsement or accountability, while the hesitation expressed through the repetition of ‘se’ underscores cognitive processing and rhetorical caution. Together, these elements frame the discourse as speculative rather than assertive.
A comparable use of modulation appears in Example (3), where a caller introduces a previously contemplated idea using ‘praticamente’, a marker that summarises a more complex internal thought process:
| (3) | Allora, dicevo praticamente mi è venuta in mente un’idea cioè sono quattro, cinque giorni che ci sto pensando. (La Zanzara, 11-03-2020) |
This usage corresponds with Ghezzi’s (2022) analysis of vagueness markers in Italian radio discourse, wherein ‘praticamente’ functions as a metadiscursive marker, foregrounding the relationship between speaker and utterance, and signalling the speaker’s stance toward both content and audience (Ghezzi, 2022, pp. 82–83, 213).
The same speaker continues in Example (4) with ‘giustamente’, acknowledging a potential challenge to his credibility:
| (4) | Mi possono anche prendere per pazzo giustamente. (La Zanzara, 11-03-2020) |
Here, ‘giustamente’ modulates the statement by acknowledging potential judgement while maintaining distance from it, functioning as a stance-taking device, particularly when speakers anticipate their ideas might seem unconventional. This strategy of pre-emptive mitigation aligns with Ghezzi’s (2022) and Bazzanella’s (2006a) findings on the social functions of modal DMs.
Analysis
Taken together, the examples above demonstrate how modulative DMs function as rhetorical buffers, enabling speakers to present controversial ideas without fully assuming epistemic or social risks. Within Italian radio interaction, these DMs play a dual role by modulating epistemic stance while also functioning as face-saving devices. Especially in emotionally charged or adversarial contexts, these devices allow speakers to balance active engagement with self-protection, thereby minimising the risk of face-threatening acts.
This careful balancing act corresponds closely with Bazzanella’s (1995, 2006a, 2006b) conception of the modal function of hedges and boosters. As she argues, these markers enable speakers to participate in contentious discourse while avoiding full propositional commitment, an essential strategy in interactions where ideological friction is high, and accountability must be managed tactically.
Moreover, the data underscore the prominence of modulative DMs in the genre of political-entertainment radio. Their frequent and reciprocal use by both hosts and callers points to a shared pragmatic competence and to co-constructed discourse norms that govern public mediated speech (Sansò, 2020). In this highly performative communicative environment, modulation emerges as a key resource, shaping a style of interaction that is both spontaneous and strategically adaptable. It serves as a tool not only for rhetorical flexibility but also for maintaining legitimacy and securing audience engagement.
Taken as a whole, these findings refine our understanding of modulative DMs by highlighting their crucial role in navigating complex interactions and in projecting authenticity under contentious conditions. Far from being peripheral linguistic features, they constitute a central mechanism through which speakers manage conflict, build rapport, and position themselves ideologically within polarised media spaces.
Turn-taking devices
Allora
Among the most frequent DMs in spoken Italian, ‘allora’ plays a crucial role in structuring interaction and regulating conversational flow. Positioned at the start of utterances, it performs various pragmatic functions, including signalling turn-taking, introducing topics, prompting elaboration, and managing transitions (Bazzanella, 1995; 2006a; 2006b; Molinelli, 2018; Sansò, 2020). This multifunctionality makes ‘allora’ particularly effective in dynamic contexts like live radio, where maintaining rhythm and coherence is essential.
While previous studies have examined the versatility of the DM ‘allora’ in everyday conversation (Bazzanella & Miecznikowski, 2007; Molinelli, 2018; Sansò, 2020), its function in mediated, high-stakes contexts such as political talk shows remains underexplored. This study addresses that gap by analysing its use in La Zanzara, focusing on how ‘allora’ structures discourse and negotiates authority under pressure. Building on research into its polyfunctionality in interactional management, this analysis extends those insights to media discourse, showing how ‘allora’ not only maintains coherence but also manages rhythm, frames arguments, and asserts or challenges speaker authority, with these functions being especially salient in adversarial and asymmetrical settings.
Empirical Examples
In the corpus, ‘allora’ features prominently in both host and audience speech, fulfilling multiple interactional roles. Consider the following Example (5):
| (5) | Cruciani: Allora? Caller: E allora vuole dire che questo virus è altamente evolutivo… (La Zanzara, 11-03-2020) |
In this instance, ‘allora’ is used interrogatively by the host to prompt the caller to elaborate. Standing alone with rising intonation, it functions as a turn-allocating cue, minimally invasive yet assertive. This allows Cruciani to exert control over the interaction without openly interrupting, maintaining his authority while inviting response.
| (6) | Allora ragazzi, io volevo parlarvi di Radio Maria… (La Zanzara, 13-03-2020) |
In Example (6), ‘allora’ signals a clear topical shift, marking the transition from informal back-and-forth interaction to a more structured monologic turn. This shift often functions as a preparatory move toward argumentation or the expression of a personal stance. The vocative ‘ragazzi’ further reinforces the speaker’s alignment with the audience, framing ‘allora’ as a soft, affiliative transition device within a participatory context.
Building on this, Example (7) illustrates a more intricate use of ‘allora’, wherein the DM plays a central role in the management of turn-taking, the reclamation of the floor, and the organisation of discourse under interactional pressure.
| (7) | Allora, aspetta, fammi finire, per favore, allora l’8 gennaio Burioni ha dato il primo allarme… (La Zanzara, 9-03-2020) |
In this case, the caller employs ‘allora’ twice: first, to reclaim the floor following an interruption, and subsequently, to resume a previously suspended line of reasoning. The DM thus functions both as a floor-holding device and as a temporal-organizational cue, enabling the speaker to structure their argument under pressure and maintain coherence despite competing turn-taking dynamics. Positioned both at the beginning and within utterances, ‘allora’ contributes to the construction of a discourse chain that brackets reformulations, signaling a return to earlier points and facilitating the speaker’s navigation of interactional tension.
Analysis
Taken together, these examples demonstrate the multifunctionality of ‘allora’ in the context of live political-entertainment radio. In Examples (5) and (6), the DM functions as an interactional resource for managing conversational flow: in the former, it subtly guides turn-taking while avoiding overt conflict; in the latter, it signals a topical shift while maintaining audience alignment. Building on this, Example (7) showcases the DM’s dual role as a floor-reclaiming and structuring marker, crucial for managing discursive coherence in contentious interactions.
These uses align with what Bazzanella & Miecznikowski (2007) describe as the expansion of the pragmatic range of ‘allora’, from temporal and textual functions to interactional management and authority negotiation. Within La Zanzara, ‘allora’ emerges as a strategic resource for navigating asymmetrical power dynamics: hosts use it to manage rhythm and assert dominance, while callers deploy it to maintain coherence and claim participation rights.
This analysis underscores the DM’s centrality in Italian spoken discourse and reveals its ideological and performative load in mediated public debate. Rather than merely structuring talk, ‘allora’ becomes an instrument of stance, positioning, and discursive control. In this sense, La Zanzara illustrates the broader media-driven transformation of interactional cues into rhetorical tools that sustain authority, regulate dissent, and perform ideological alignment.
Ma
In Italian, the DM ‘ma’ serves two primary functions related to adversative contrast: signalling full contrast between opposing propositions and partial contrast as a corrective operator that revises prior inferences. In both cases, ‘ma’ functions as a multifunctional inferential operator, structuring discourse by marking clause relationships (Fedriani & Molinelli, 2019). At the interactional level, it contributes to turn-taking, often indicating the end of a turn or a shift in topic, consistent with its frequent sentence-initial position (Bazzanella, 1995, 2006a, 2006b; Molinelli, 2010). It also introduces counter-expectation, contrasting earlier content or the interlocutor’s assumptions (Borreguero, 2009). As such, ‘ma’ plays a procedural role, guiding interpretation and frequently marking contrast (Mauri & Giacalone, 2012).
In the discourse data explored in this study, ‘ma’ appears with notable frequency in both host and audience speech, predominantly in sentence-initial position. Its use extends beyond marking contrast between ideas. In emotionally charged or combative moments, ‘ma’ operates as a strategic tool for managing turn-taking, escalating conflict, and shifting attention from substantive content to personal character. The data suggest that ‘ma’ is not only central to argumentative structuring but also functions as a mechanism of interpersonal pressure and ideological positioning, especially in sequences where rational debate breaks down in favour of verbal confrontation.
Empirical Examples
In the analysed corpus, ‘ma’ often appears utterance-initially, marking adversative relations and facilitating negotiation between agreement and disagreement. It introduces counterarguments, challenges statements, and marks conversational shifts, underscoring its role in shaping interactional dynamics. Example (8) exemplifies this use, with utterance-initial ‘ma’ introducing a confrontational stance and marking a transition from alignment to opposition within the ongoing discourse:
| (8) | Ma adesso tutti quanti a elogiare i cinesi? Ma ti mangiano tutta la merda e ci contagiono [sic] da che mondo è mondo perché sempre… (La Zanzara, 13-03-2020) |
On the other hand, Example (9) illustrates how ‘ma’ extends beyond mere refutation to function as a key interactional resource in highly confrontational exchanges, where it serves to escalate conflict, interrupt turns, and assert dominance in the interaction:
| (9) | Gottardo: Ma non si capisce un cazzo di quello che dice poi! Cruciani: Mauro mi hai rotto i coglioni! Gottardo: Ma vai da un logopedista. Ma vai da un logopedista! Cruciani: Ma non si capisce quello che dici. Sei finito, sei morto. Sei finito. Gottardo: Poi tra le altre cose, tra le altre cose, il cellulare nuovo l’ha comprato con il reddito di cittadinanza il cellulare nuovo. Cruciani: Ma scusami X. L’hai comprato con il reddito di cittadinanza? Gottardo: Ma sei un mantenuto da Mantova? Caller: Certamente. Gottardo: Ma sei un mantenuto! (La Zanzara, 9-03-2020) |
Analysis
In Example (8), the DM ‘ma’ signals a shift from normative assumptions to a critical stance on a broader social issue. The caller’s utterance marks the onset of an adversative turn, in which ‘ma’ performs a classic argumentative function: introducing contrast, foregrounding disagreement, and aligning with its well-documented procedural role (Mauri & Giacalone, 2012).
Example (9), by contrast, demonstrates an intensified use of ‘ma’ within a highly confrontational exchange. Here, the repeated sentence-initial deployment of the marker not only escalates interpersonal tension but also deflects attention from the substantive content of the argument. The hosts’ use of ‘ma’ functions rhetorically to undermine the caller’s social credibility, targeting dialectal and economic attributes and shifting the interaction from rational discussion to personal disparagement.
In this context, ‘ma’ facilitates argumentative disruption through ad hominem tactics, redirecting the focus from issues to individuals. It thus reveals a dual functionality: serving both as a turn-taking cue and as a strategic device for confrontation. While typically associated with the introduction of counterarguments, ‘ma’ also enables the destabilisation of discourse through personal attack. These adversative properties significantly shape the dynamics of debate on La Zanzara, reinforcing the programme’s confrontational ethos and subverting the conditions for reasoned dialogue.
Although ‘ma’ is well established as a DM of argumentative structuring (Bazzanella, 1995; Fedriani & Molinelli, 2019), the examples above reveal an expanded interactional role. In particular, example (9) shows ‘ma’ enabling personal attacks and contributing to the breakdown of reasoned exchange. This instance suggests that ‘ma’ can, in certain interactional contexts, serve as a tool of aggression, used to undermine interlocutors, and shift the focus from substantive issues to personal identities. Nonetheless, further research is needed to explore the conditions under which this DM mediates not only argumentative structure but also the negotiation of power dynamics in interaction.
The role of DMs in hate speech
Turn-taking devices
Ecco
Previous studies identify multiple functions of the DM ‘ecco’ in Italian. In sentence-initial position, it often signals the speaker’s intent to take the floor, introduce a new segment, or present a participant, serving as a turn-taking device. It may also act as a filler, marking hesitation or cognitive effort. As a modulation device, ‘ecco’ can emphasise upcoming content when paired with rising intonation or reduce speaker commitment with falling intonation and pauses. In sentence-final position, it takes on distinct pragmatic roles, such as indicating agreement, confirming receipt of information, self-validating lexical choice, or interrupting the interlocutor (Bazzanella, 1995, pp. 233–234; 2006a; 2006b).
This study reveals an additional, context-specific function of ‘ecco’ in adversarial radio interaction: its use as a rhetorical device for confrontation. In La Zanzara, ‘ecco’ often co-occurs with derogatory or sarcastic content, becoming a marker of mockery and adversarial stance, especially in sequences where speakers discredit or ridicule their interlocutors.
Empirical Examples
This rhetorical refunctioning of ‘ecco’ is particularly evident in the Examples (10) and (11), in which the DM is repeatedly deployed at the start of utterances to preface personal attacks, amplify sarcasm, and assert an adversarial stance. Repetition and prosodic emphasis further enhance its performative impact, transforming ‘ecco’ from a neutral presentational cue into a strategic device for ridicule and confrontation.
This function is clearly illustrated in Example (10):
| (10) | Eccolo qua il genio… ecco il genio del genio… ecco l’ingegnere del giorno. (La Zanzara, 9-03-2020) |
In this instance, Parenzo uses ‘ecco’ iteratively to introduce a series of mocking labels. The repetition lends the utterance a sarcastic rhythm, undermining the caller’s perceived intelligence and recasting ‘ecco’ as a framing device for personal disparagement. A similar rhetorical function is performed by the DM in Example (11):
| (11) | Ecco il falsario! Ecco il falsario! (La Zanzara, 11-03-2020) |
Here, Gottardo deploys ‘ecco’ in a chant-like repetition to brand someone as a forger. The DM prefaces the insult, amplifying its accusatory tone and introducing a confrontational stance from the outset.
Analysis
These examples demonstrate how ‘ecco’, though traditionally associated with turn management and discourse organisation, is refunctionalised in the polarised setting of La Zanzara to perform adversarial and delegitimising functions. Its prosodic repetition and placement before insults reposition it as a strategic cue for hostility, framing the interlocutor not through argument but through ridicule or character attack.
This rhetorical shift exemplifies the recontextualisation of a structurally neutral marker into an aggressive discursive device. When paired with derogatory epithets, ‘ecco’ enables ad hominem fallacies by drawing attention away from the argument and toward personal attributes, often in a mocking tone. In this way, ‘ecco’ becomes part of the ideological style of the programme, helping to sustain its confrontational ethos and its boundary-marking logic of inclusion/exclusion.
While Bazzanella (1995; 2006a; 2006b) emphasises the DM’s metatextual and interactional functions, its strategic use in media discourse reveals a broader rhetorical elasticity. It thus deserves recognition not only as a discourse organiser but also as a vehicle of antagonism and discursive control in contentious public interaction.
Phatic devices
Phatic expressions, as defined by Bazzanella (1995, pp. 236–238; 2006a; 2006b), facilitate social cohesion and interpersonal closeness by establishing rapport and signalling shared affiliation.
In La Zanzara, phatic DMs, especially vocative forms like proper names, serve as rhetorical tools for presenter Cruciani and his audience, albeit in markedly different ways.
While audience members often use proper names to confront or challenge, Cruciani extends this repertoire, incorporating terms such as ‘ragazzi’, ‘amico mio’ and ‘caro amico mio’ to foster a discursive in-group. This deliberate layering of familiar address enhances solidarity among aligned participants while marking boundaries against ideological opponents, reinforcing the role of language in shaping affiliation and othering.5 Phaticity operates alongside confrontation, as Cruciani blends warmth with mockery, transforming these expressions into tools that both include and exclude.
Empirical Examples
The following examples demonstrate how phatic expressions—particularly vocatives and terms of address—function in La Zanzara as tools for shaping social alignment and managing interpersonal dynamics. While some instances foreground confrontation, others foster a sense of group cohesion, revealing the strategic versatility of phatic language in adversarial media discourse.
Consider Example (12):
| (12) | Voice message: Parenzo, ma non ti senti un coglione ad avere preso per il culo Meluzzi ed è successo tutto quello che diceva lui e niente di quello che dicevi tu, Parenzo, non ti senti un coglione? Non ti senti un coglione? Come Gottardo, del resto, ma da quello non ci si aspetta niente. Tu sei Parenzo… Parenzo, non ti senti un coglione? (La Zanzara, 9-03-2020) |
In this example, the repetition of ‘Parenzo’ targets the interlocutor, while ‘coglione’ underscores the message’s hostility. The vocative form lacks affiliative function, signalling antagonism and positioning the speaker outside the programme’s discursive community.
Example (13) presents a markedly different use of phatic DMs by Cruciani:
| (13) | Cruciani: Amici miei, è arrivato il momento di chiamare il genio del giorno, il sindaco di Boscoreale, ragazzi. Caro amico mio David, come non risponderà? Sì, il migliore… Il signor Diplomatico. Tra l’altro si chiama Diplomatico. Pronto? Sindaco? (La Zanzara, 12-03-2020) |
Cruciani’s use of ‘amici miei’, ‘ragazzi’ and ‘caro amico mio’ constructs camaraderie and frames the interaction as familiar and unified. These phatic expressions simultaneously address his co-host and the audience, constructing an inclusive discursive space and reinforcing an imagined sense of collective participation.
A similar affiliative function is observed in Example (14):
| (14) | Cruciani: Ragazzi, mi sono dimenticato di dirvi che il coglione del giorno, l’inglese che ha detto che per gli italiani questa epidemia è come una siesta, questo dottor Jensen, dottore, questo fa anche là il dottore, questo qua, capisci? È la star del programma. Parenzo: Sì, quello di Malattie Imbarazzanti. Cruciani: Esatto, bravo, è la star del programma TV inglese Malattie Imbarazzanti, che peraltro è abbastanza attinente col personaggio. Ma ragazzi, è imbarazzante lui, e dico di più, ragazzi, non viene. (La Zanzara, 13-03-2020) |
In this instance, the repetition of ‘ragazzi’ functions to frame the critique of Dr Jensen as a collective judgement. By repeatedly addressing the audience in familiar terms, Cruciani presents his mockery as shared and consensual, reinforcing group identity while softening the aggressiveness of the content. The phatic DM thus operates as both an inclusionary tool and a rhetorical buffer for confrontation.
Analysis
The examples above reveal how phatic markers, typically associated with relational maintenance, are strategically employed to perform both inclusion and exclusion. Cruciani’s use of terms like ‘ragazzi’ and ‘caro amico mio’ blends familiarity with critique, turning phaticity into a mechanism for reinforcing ideological alignment while softening or masking rhetorical aggression.
While Cruciani uses phatic markers to foster a sense of in-group cohesion, his audience often strips these expressions of affiliative value, repurposing vocatives like ‘Parenzo’ as weapons of confrontation. This contrast illustrates the dual nature of phatic DMS: they can enhance interpersonal connection, but they can just as easily delineate and intensify oppositional roles.
Ultimately, phatic DMs in La Zanzara reveal themselves to be more than socially lubricating formulas: they are ideologically charged resources that help construct the show’s polarised and performative style. By framing criticism as communal rather than personal, Cruciani increases its persuasive force, while audience vocatives accentuate outsider status. In this way, phaticity plays a central role in shaping participation, identity, and ideological division in Italian media discourse.
Dai
Previous scholarship identifies a wide range of functions for the DM ‘dai’. While it may appear in reactive interventions—such as accepting requests or offers initiated by another speaker—it more typically occurs within directive speech acts (e.g., requests, orders, or suggestions), where it carries injunctive force. Its pragmatic function is closely tied to its prosodic realisation and position within the utterance: depending on these variables, ‘dai’ may serve a perlocutionary function, acting either as a politeness booster or as a mitigating hedge (Bazzanella, 1995, 2006a, 2006b; Ghezzi & Molinelli, 2014; Molinelli, 2018).
In the examined corpus, ‘dai’ is among the most frequently occurring DMs, appearing predominantly at the end of utterances and used almost exclusively by the hosts. In this context, its primary function appears phatic: ‘dai’ signals shared knowledge and reinforces implicit agreement between speaker and audience, fostering a sense of communal understanding.
However, its use is far from neutral. Indeed, the data from La Zanzara reveal a notable nuance that extends previous observations. Unlike other affiliative phatic markers, ‘dai’ often conveys a dismissive or condescending tone, particularly in adversarial exchanges. In these instances, it functions as a rhetorical device that facilitates ad hominem attacks. By framing personal insults or discriminatory remarks as self-evident or widely accepted truths, ‘dai’ allows speakers to mask prejudice as shared common sense. This strategy not only diminishes the perceived severity of the attacks but also reduces the likelihood of immediate resistance or challenge.
Empirical Examples
The following examples illustrate how ‘dai’ functions as a rhetorical device for dismissal and mockery, diverging from its traditionally affiliative uses. In the corpus, ‘dai’ is frequently employed utterance-initially to frame disagreement or ridicule as self-evident, thereby reducing the face-threatening nature of the statement and discouraging resistance from interlocutors or listeners.
Consider Example (15):
| (15) | Parenzo: Sei un poveretto, un mitomane. Sei un mitomane, un matto, uno spostato, un poveretto, un senza nulla, un cretino. Cosa vuoi che ti dica di più? Caller: Ma tu sei senza nulla. Coglione! Parenzo: Ma è un poverino, dai su! (La Zanzara, 9-03-2020) |
Here, ‘dai’ deflates the caller’s response and reaffirms Parenzo’s derogatory framing. The marker trivialises the insult sequence, suggesting that the judgment is a self-evident truth, implicitly shared with the audience.
A similar framing strategy appears in Example (16), where ‘dai’ is employed not only to reinforce a shared conclusion but also to embed a cultural stereotype within a familiar, humorous register:
| (16) | Cruciani: E però, dove capita una cosa di questo tipo? Dove è capitata? Sarà un caso? Certamente sarà un caso, non c’è dubbio. Non c’è dubbio. Sarà un caso, ma dove è capitata? Nell’ospedale di quale città è capitata questa roba? Dai su. (Starts playing Tarantella Napoletana) E dai su. Qual è l’ospedale ragazzi? In quale città? Bolzano? Bologna? No, un tizio ha sputato addosso a un medico e a un infermiere. Tra l’altro un responsabile sanitario di Napoli, che cos’è che ha detto? Ha detto che praticamente è come avere sparato a qualcuno, adesso magari esagerando, no? (La Zanzara, 11-03-2020) |
In this instance, ‘dai’ frames Cruciani’s insinuation, that such behaviour is predictably Neapolitan, as a self-evident conclusion. The ironic tone is reinforced by the insertion of the Tarantella Napoletana, which embeds the stereotype within a cultural script. This multimodal strategy masks discrimination with humour and familiarity, trivialising the incident while constructing symbolic boundaries between an in-group and a culturally marked other.
Analysis
In both examples, ‘dai’ functions as a discursive shield, veiling ad hominem attacks beneath presumed shared cultural assumptions. In Example (15), the DM trivialises a direct insult exchange, reaffirming a derogatory stance while presenting it as self-evident and unworthy of further contestation. Example (16) builds on this function, pairing ‘dai’ with the Tarantella Napoletana to activate a regional stereotype through multimodal irony. In both instances, ‘dai’ functions as a device of rhetorical closure, recasting offensive or prejudicial content as unproblematic and at times even humorous.When accompanied by condescension, cultural cues, or ironic performance, the DM invites audience complicity and renders resistance socially less viable.
These findings align with Molinelli’s (2018) observation that ‘dai’ may carry epistemic value, addressing not only interpersonal alignment but also the evaluative framing of content. Crucially, in these contexts ‘dai’ operates not as a mere softener or politeness marker, but as a mechanism for normalising ideologically charged content. It reframes insults and discriminatory insinuations as collective insight, reinforcing in-group solidarity while marginalising dissent. This rhetorical strategy transforms individual bias into symbolic consensus, legitimising exclusionary attitudes in the guise of shared truth.
In sum, ‘dai’ plays a pivotal role in legitimising discriminatory speech within La Zanzara’s discursive repertoire. Its sentence-final placement enables it to signal closure and consensus while framing offensive remarks as universally intelligible. When paired with humour, cultural references, or ironic tone, this function is amplified—further obscuring the ideological force of the message. Through this subtle yet powerful mechanism, ‘dai’ exemplifies how minor pragmatic elements can facilitate ideological entrenchment and social exclusion in public discourse.
Conclusions
This study demonstrates that DMs in La Zanzara operate along two interrelated axes: the structuring of argumentation and stance, and the facilitation of discrediting strategies and ideological polarisation. These overlapping functions highlight how linguistic choices mediate both interactional control and broader ideological agendas within political-entertainment media discourse.
The first axis includes DMs that organise and modulate argumentative content, such as ‘cioè’, ‘ma’, ‘allora’, and modal expressions like ‘praticamente’, ‘giustamente’, and ‘evidentemente’. Markers like ‘cioè’ contribute to reformulation and coherence, particularly in asynchronous formats, where emotional tone and cognitive processing must be managed. In turn, ‘allora’ regulates speaker transitions and authority claims, while ‘ma’ introduces contrast and counterarguments, signalling shifts in alignment or disagreement. Modal DMs help shape epistemic stance, allowing speakers to hedge, soften, or reinforce their assertions. This enables participation in contentious debate without full propositional commitment. These patterns, as discussed in the analytical sections, align with prior research on how DMs function pragmatically to structure discourse, manage stance, and guide interaction.
The second axis centres on DMs that underpin discrediting strategies and ideological positioning, particularly through ad hominem attacks and exclusionary rhetoric. Markers such as ‘ecco’ and ‘dai’ play key roles here. The former, adapted from a turn-taking cue, often signals sarcasm, and precedes personal attacks, redirecting attention from the argument to the individual. The latter, frequently occurring at the end of utterances, frames discriminatory remarks as self-evident truths, concealing bias within presumed cultural consensus and reinforcing in-group cohesion. Modal expressions intersect with this function by softening prejudicial claims or diffusing accountability, thereby blurring the boundary between reasoned disagreement and ideological distancing. These rhetorical shifts illustrate how certain DMs extend beyond metatextual or interactional roles to act as instruments of exclusion and confrontation, a dimension largely underexamined in existing scholarship.
Further ideological boundaries are marked through phatic vocatives such as ‘ragazzi’ and ‘caro amico mio’. These familiar forms, when used by Cruciani, blend mockery with camaraderie, fostering solidarity among aligned listeners while tempering overt conflict. Audience members, by contrast, employ vocatives stripped of affiliative function to challenge the hosts and assert ideological opposition.
Collectively, these DMs allow La Zanzara to merge argumentation with fallacy, sustaining a discursive style that normalises offensive speech while maintaining the illusion of open debate. This study contributes to discourse analysis and media pragmatics by demonstrating how DMs mediate power and legitimise exclusion. Although situated in a specific media context, the findings open avenues for further research into the ideological work of DMs across diverse communicative settings.
Notes
[1] Prior to 2021, Alberto Gottardo contributed regularly to the radio show La Zanzara. However, following his conviction for incitement to crime, he is no longer associated with the programme.
[2] In the section dedicated to the description of the programme, available on the various digital platforms of the Radio24 radio station, the programme is described as: ‘Giuseppe Cruciani con David Parenzo conduce “La Zanzara”, l’attualità senza tabù, senza censure, senza tagli alle vostre opinioni. Una zona franca per gli ascoltatori, uno spazio nemico della banalità e del politicamente corretto, l’arena dove il primo comandamento è parlare chiaro’ (La Zanzara, Radio24).
[3] This analytical orientation draws on the broader framework of CDA, particularly the work of Van Dijk (1993, 2006, 2017), Fairclough (2013, 2015), and Wodak (2015a, 2015b), who conceptualise language as a central medium for enacting and legitimising power relations. These perspectives deepen our understanding of hate speech as part of wider discursive and social structures, rather than as isolated verbal aggression.
[4] Tables 1 and 2 present an annotation scheme adapted from established classifications in Italian linguistics, particularly drawing on the works of Bazzanella (2006b), Ghezzi and Molinelli (2018), and Sansò (2020). The categories have been adjusted to reflect the specific pragmatic functions relevant to the present study’s focus on DMs in Italian radio discourse.
Competing Interests
This article stems from my doctoral research on Italian politicians’ use of DMs in the early 21st century. A preliminary version was presented at the conference Quelles places pour l’échange démocratique dans les espaces médiatiques numériques? (“What Place for Democratic Exchange in Digital Media Spaces?”), held June 18–19, 2024, at Stockholm University, and was refined following feedback from that event. I declare no conflicts of interest regarding this publication. The article is part of a special issue published with the support of the “Language and Power” network at Stockholm University and the Åke Wiberg Foundation (project H23-0123).