Introduction

The fifth generation of wireless infrastructure (henceforth 5G) is generally perceived as a technological issue in our digital society. Instead, it has also become an internationally controversial theme, fueled by several socio-economic and political factors beyond its technical components. Therefore, unlike previous wireless generations, 5G is more extensively discussed in mainstream media and public forums (Radu and Amon, 2021) where power and values are intricately intertwined. Campbell et al. (2021, p. 559) have noted the significance of investigating “patterns and themes in news articles that reflect and shape social imaginaries of 5G”. Public understanding of science and technology is prone to be shaped by not only science popularization and education activities but also the particular ways it is communicated through the text or talk of institutional actors such as media, policymakers and legislators. Mass media play a crucial role in shaping public perceptions of technology. The past two decades have witnessed increasing attention to media discourses on science and technology such as hackers (Pei et al., 2022), biotechnology (Holmgreen, 2008) and gene editing (Zhang et al., 2021).

Prior works (e.g., Mansell and Plantin, 2020; Ventsel et al., 2021) demonstrate that news media representations of 5G in a specific country such as Britain, Russia or Germany have been touched upon. However, relatively little attention has been attached to a cross-country comparison of how 5G is represented in news media. The present study offers a comparative analysis of the particular ways of portraying 5G in the Chinese and British press. The rationales for choosing China and Britain are as follows: (i) Huawei, which plays a leading role in 5G technology and acts as the world’s largest telecoms equipment producer, is a Chinese company; and (ii) Britain is one of the two prominent western markets of Huawei and the first established European partner of Huawei (Liu, 2020). The Britain-Huawei relationship has witnessed a sequence of twists and turns. In earlier years, Huawei had a close partnership with Britain and became a critical provider of network gear to some British service providers by 2010 (Liu, 2020). Though the US has long perceived Huawei as a security threat and even pushed its allies to ban Huawei’s 5G from their national networks, the UK National Cyber Security Center concluded that Huawei was a manageable risk to 5G in 2019 (Sevastopulo and Bond, 2019). Instead, in July 2020, Britain issued a decision to remove all Huawei equipment from its 5G networks and ban buying new Huawei 5G equipment. Against the backdrop, it is vital to get an overall view of how 5G is envisioned within such a changing and complicated socio-political context.

The upcoming sections will begin with a review of extant research on the media discourse of 5G. Then, the corpus-assisted critical discourse analysis, data collection and analytical procedures will be discussed. Finally, the findings and analysis will be presented, followed by a discussion and final remarks.

The media discourse of 5G

Given the novelty of 5G as a recently emerged technology, previous works on 5G and media discourse have been primarily published over the past three years. Generally, prior studies are overwhelmingly oriented towards two strands of research: (i) the production and transmission of 5G conspiracy theories in digital media and (ii) the media framing of Huawei’s image. Buarque (2022, p. 326), for instance, conducts a multimodal critical affect-discourse analysis of one video surrounding the 5G conspiracy theory, stating that “digital media has facilitated the legitimation of conspiracy theories through novel ways of claiming authority and visibility”. Yang and Kang (2020) utilize text-mining techniques to expound on how Huawei is framed in global media discourses, revealing that Chinese media tend to frame Huawei from the trade war perspective, whereas non-Chinese media concentrate on its security concerns. Gagliardone et al. (2021) analyze the differences in how social media in Nigeria and South Africa represent COVID-19 conspiracies on 5G and on Bill Gates and how the way of representation interacts with particular socio-political contexts.

Despite the preceding two strands, there have been limited investigations into 5G representations in news media, with merely a few exceptions such as representations of 5G in the British, Russian and German press. Mansell and Plantin (2020, p. 29), for instance, illustrate how 5G deployments are represented in the British broadsheet and tabloid newspapers from 1 January 2017 to 7 March 2020, claiming that the British press fails to “provide comprehensive insight for citizen readers about the arrival and implementation of a new 5G network”. Ventsel et al. (2021) compare Russia’s strategic blame narratives in the domestic and international media coverage of 5G. Dong and Gao (2022) unpack the representation of the 5G image in a corpus of German news texts on 5G, arguing that German media stress both a necessity of Germany-Sino 5G cooperation and a hesitancy for that due to the security risks and economic losses. Instead of focusing on news media, Campbell et al. (2021) elaborate on the different ways of portraying 5G in Chinese and United States (henceforth US) advertisements, concluding that 5G is taken as a source of national pride and elevation in China and a ground for domestic competition in the US.

Previous studies and the aforementioned historical trajectory of the Britain-Huawei-5G relationship manifest that 5G is not only a scientific matter but a social and geopolitical issue. 5G has evolved as an issue of security, national interests, geoeconomics and geopolitics and is sometimes linked with negative perceptions, images and narratives (Cheng and Liu, 2022; Robles-Carrillo, 2021). Accordingly, the departure point of this study is that 5G could carry social and political meanings. This renders Critical Discourse Studies (henceforth CDS) a helpful and practical analysis toolkit for this study, since it deals with the investigation of discourse in its historical, socio-political and cultural contexts (Flowerdew and Richardson, 2018). Furthermore, it perceives language as a social activity and especially focuses on how ideologies and power dynamics are conveyed, maintained and challenged through language (Baker, 2015). CDS has been widely used in studies on the media representations of different technological discourses, such as artificial intelligence (Ouchchy et al., 2020) and hackers (Pei et al., 2022). Nevertheless, relatively few studies have examined the press representations of 5G from a CDS perspective. In doing so, this study, drawing upon CDS, addresses this lacuna by undertaking a comparative corpus-assisted analysis of how 5G is envisioned in the Chinese and British press.

Corpus-assisted critical discourse analysis

This study is a corpus-assisted critical discourse analysis combining corpus-assisted discourse studies (henceforth CADS) and Wodak’s discourse-historical approach (henceforth DHA) in CDS. Corpus linguistics has become a widely adopted method of uncovering linguistic patterns across larger datasets by utilizing techniques such as frequency, keyword, collocation and concordance analyses. The keyword investigation could aid in unraveling “important concepts in a text (in relation to other texts) that may help to highlight the existence of types of (embedded) discourse or ideology” (Baker, 2004, p. 347). Collocation and concordance analyses could help to detect more co-text and contextual information of a word. CADS, a term first coined by Partington (2004), is a subset of corpus linguistics. CADS not only concerns quantitative analyses of corpus data but also considers the corpus-external data to interpret and explain the corpus data and identify areas for analysis (Partington et al., 2013). This study conducts a quantitative analysis of recurrent linguistic patterns in the corpora and a qualitative analysis of such patterns in the expanded co-text, while also taking into account the socio-political context of the corpora under study. Besides, it employs an external comparison corpus, the 100-million-word British National Corpus (henceforth BNC), to identify salient themes for analysis in the corpora. BNC is selected not only because it is one of the largest corpora covering a wide variety of texts by genre and subject fields (Kennedy, 1998), but also because the British media examined in this paper are specific to British English.

CDS researchers share interests in examining the relationship between language, power and ideology and unpacking discourse in social, cultural, political and institutional contexts (Flowerdew and Richardson, 2018; Pei et al., 2022). CDS is not a method of discourse studies but an array of approaches (Wodak, 2001), such as van Dijk’s sociocognitive approach, Wodak’s DHA, van Leeuwen’s social actors approach and Fairclough’s dialectical-relational approach. The distinctiveness of DHA lies in its focus on integrating “historical, political and socio-cognitive dimensions into the analysis and interpretation of discursive events” (Anagnostopoulos et al., 2013, p. 165). DHA encompasses three dimensions of analysis: (i) specific topics of a specific discourse, (ii) discursive strategies, and (iii) linguistic means and specific, context-dependent linguistic realizations (Wodak, 2001, p. 72).

Accordingly, this study, defining discourse as principally topic-related (Baker et al., 2008), takes thematic or topical analysis as a departure. When dealing with larger corpora, thematic analysis can offer a first and overall idea of what a discourse or corpus is all about (van Dijk, 2001). Then, this study delves into the discursive strategies utilized by the Chinese and British press and the socio-political contexts in which their discursive practices concerning 5G are encoded. Five discursive strategies of self-other presentation are identified under DHA. These strategies encompass referential or nomination strategies realized through membership categorization devices; predication strategy through which social actors are labeled positively or negatively; argumentation strategy that concerns the justification of positive or negative attributions; perspectivation strategy that concerns the positioning of the speaker’s point of view; and intensification or mitigation strategies that refer to the modification of the epistemic status of a proposition (Wodak, 2001).

The corpus-assisted critical discourse analysis has become a popular and fruitful avenue for linguistic research (Baker, 2015; Samaie and Malmir, 2017). Accordingly, this study not only exploits CADS to conduct keyword, collocate and concordance analyses of the news texts, but also takes a critical discourse analysis to elucidate the underlying ideologies behind media’s particular ways of representing 5G. More specifically, the present study seeks to capture and demystify the frequent themes of, or issues discussed in, news articles relating to 5G in China and Britain, and interpret how and why the themes or issues are anchored in broader socio-political contexts.

Methodology

Data

To build two corpora of Chinese and British press articles on 5G, we extracted all news texts containing the lexical item 5G or fifth generation of wireless infrastructure in their headlines from the Lexis-Nexis database. The Chinese corpus (henceforth CC) includes news from China Daily, its Africa Weekly, its US Edition, its Hong Kong Edition and its European Edition. The British corpus (henceforth BC) includes news from The Guardian (with its Sunday edition The Observer), The Times (with The Sunday Times) and The Daily Telegraph (with The Sunday Telegraph). China Daily, one of the major English-language media in China with wide circulation, has been popularly utilized as the data source for corpus studies about Chinese media (Pei et al., 2022). It tends to convey the official Chinese government’s positions and values (Liu, 2009). The three British broadsheet newspapers enjoy high circulation and prominent online presence, exemplify the British quality press and concern both the left (The Guardian) and right (The Times and The Daily Telegraph) of the political spectrum (Walsh, 2020). The time span set for data collection is from 13 May 2013 through 13 June 2022: the former is the date on which Samsung announced it had created a 5G network and since which companies such as Huawei and Google also announced plans to invest in 5G; and the latter is the date on which the data collection was accomplished. After removing duplicated news, 657 news articles were acquired for CC and 577 for BC. Table 1 offers detailed information about the two corpora.

Table 1 The detailed information of the two corpora.
Full size table

Methods

As noted, this study combines CADS and CDS. With the help of keyness, the main themes concerning 5G in both corpora surface. One primary objective of this study is to employ the corpus approach to extract keywords from the two corpora and then compare the themes emerging from the two keyword lists. Of note, the resulting themes emerge from the two corpora rather than from any predetermined categories, which helps generate a more abundant classification. In doing so, the obtained categories can be tailored to more accurately reflect the lexical-semantic features and themes about 5G. Moreover, collocation and concordance analyses of the keywords under each theme are further undertaken to provide more useful contextual clues and facilitate an in-depth understanding and comparison of the obtained themes.

This study proceeds in three steps by using WordSmith 8.0 (Scott, 2022). First, it conducted a keyword analysis by comparing each corpus against BNC to retrieve keywords. Keywords were identified by integrating log-likelihood and log ratio statistics. To be a keyword, a word must occur in a minimum of 10 percent of the texts in each corpus, thus with a minimum frequency of 66 instances in CC and 58 instances in BC. The log-likelihood (henceforth LL) test was conducted with a p-value set at 0.000001 to give 99.9999 percent confidence that an identified keyword does not occur by chance. As an effect size measure, the log ratio (Hardie, 2014) was employed to refine the keyword calculation and obtain the words where the difference in frequency is greatest. To render the two corpora comparable in terms of strength, we compared keywords with a log ratio score of at least 7. After removing keywords denoting Arabic numbers, 69 keywords were identified for CC and 50 keywords for BC (see Table 2).

Table 2 The keywords identified in the two corpora (ordered by log ratio).
Full size table

The categorization process was supported by a closer concordance analysis of the keywords to obtain their contextual cues presented in the expanded linguistic co-text and then to ascertain their exact roles in representing 5G. Eventually, four general themes were obtained (see Table 3): (i) problem definition: keywords concerning the nature of 5G; (ii) social actors: keywords denoting potential social actors involved in 5G discourse; (iii) actions: keywords indicating possible actions performed by the social actors; and (iv) timing: keywords referencing years and the era of COVID-19. It bears mentioning that this categorization is based on a thorough examination of the concordance lines of keywords. For instance, keywords Twitter and Facebook are assigned under the category Timing (the COVID-19 era) due to their more frequent occurrence as the two major social media platforms where misinformation about 5G and COVID-19 is transmitted. Keywords trillion and Yuan are allocated to the category Actions (5G implementation) since they commonly allude to the government investments in 5G and the total economic output created by 5G. The next section dwells on the four thematic categories contributing to the representation of 5G in each corpus.

Table 3 The thematic categorization of 5G keywords.
Full size table

Thirdly, keywords under each theme were further selected for collocation and concordance analyses. The space constraints prevent us from providing a detailed examination of all keywords in each theme. The keywords were not subjectively selected, but based on two criteria: first is to select words based on several statistical criteria, such as words with high log ratio or frequency; and second is to select words showing competing representations of specific themes relating to 5G. Of note, all collocation analyses in this study focus on the non-function collocates to mine more meaningful collocation patterns. The whole study is mobilized by analyzing keywords and scrutinizing their collocates and concordance lines to obtain their contributions to the portrayal of 5G. We also used the online UCREL log-likelihood and effect size calculator (https://ucrel.lancs.ac.uk/llwizard.html) to calculate the magnitude of the statistical difference of certain words’ frequency in the two corpora. We reported the LL value representing the likelihood of the difference and the log ratio score representing the size of the difference.

Findings and analysis

This section is a collocate and concordance analysis of the keywords under the four themes that play a role in representing 5G. The aim of the analysis is two-fold: to compare how 5G discourse is represented in the two corpora and to unravel the socio-political motivations behind their particular ways of representation. It is noted that the word 5G with digits in it is unrecognizable in WordSmith’s collocate listing. Hence, we used FiveG to replace 5G in the original text file to make it recognizable in Tables 5–7 and Figs. 6, 7 below. The following analyses tackle each theme in turn.

The nature of 5G

In Table 3, it is unsurprising to find keywords straightly signifying the mobile wireless technologies, including 5G, 4G and 6G in CC and 5G, 4G and 3G in BC. Both corpora share keywords underlining the typically technical natures of 5G, including Internet, broadband, download, telecommunication, tech(nologies), telecoms, wireless and ecosystem in CC, and cyber, broadband, telecoms, Internet, airwaves, mobile and tech in BC. 5G, by definition, is a technology featuring extended bandwidth, high speed, high capacity, low power consumption, low latency, faster download and large-scale connectivity (Liu et al., 2020). These inherent properties of 5G are addressed in both corpora, as revealed by superfast, connectivity and latency in CC, and download and connectivity in BC.

5G is the most key and frequent word denoting the nature of 5G in both corpora. To unfold how 5G is evaluated by each corpus, we investigated the 10 most frequent evaluative adjectival, nominal and verbal collocates of 5G at the L1 position (see Table 4). This method can help to explore how the predication strategies are realized by identifying the collocates that “appear in stereotypical evaluative attribution of positive or negative traits and implicit or explicit predicates” (Meyer, 2001, p. 27).

Table 4 The top 10 evaluative adjectival, nominal and verbal collocates of 5G.
Full size table

Due to the influential role of Huawei in 5G development and the journalistic value of geographical proximity to the targeted audiences, it is common to find collocates China’s and Huawei’s in CC and Britain’s, UK’s and Huawei(’s) in BC. Moreover, both corpora cover a string of adjectives and nouns denoting degree and magnitude, demonstrating their differing attitudes towards 5G. CC endorses completely positive attitudes towards 5G, as revealed by collocates first, million, largest and superfast. A collocation analysis of first 5G at the R2 position finds that the most frequent collocates of this cluster are smartphone (25), network (12), phone (12), remote (surgery) (8) and chipset (7). It suggests that CC underlines the breakthroughs created by 5G and its wide applications in smartphones and healthcare. A collocation analysis of million 5G at the R2 position shows that it most frequently collocates with users (34), subscribers (27), mobile (23), smartphones (21) and connections (17), evincing CC’s great attention to the rapid popularization of 5G. A collocation analysis of largest 5G at the R2 position shows that it mostly collocates with world’s (68), market (39), network (14), field (14) and test (13), showing CC’s emphasis on China’s leadership role in 5G. The collocates commercial, commercialize and superfast in Table 4 suggest CC’s emphasis on 5G commercial use and its high-speed feature.

In BC, similarly, the most frequent collocates of first 5G at the R2 position are network (6), iPhone (6), UK’s (4), handsets (3) and available (2), indicating 5G’s broad applications in the smartphone industry. However, the collocates anti and linking, which rank fourth and tenth respectively in Table 4, uncover the negative perceptions of 5G. The most frequent collocates of anti 5G at the R2 position are groups (8), Facebook (4), protesters (3), group (3) and protests (2). The most frequent collocates of linking 5G at the R2 position are COVID (3), coronavirus (2), technology (2) and health (2). As seen from the further concordance analyses of anti 5G and linking 5G in Figs. 1 and 2, reports about anti-5G protesters and conspiracy theories linking 5G to coronavirus have received lots of attention in BC.

Fig. 1
figure 1

The sample concordance lines of anti 5G in BC.

Full size image
Fig. 2
figure 2

The sample concordance lines of linking 5G in BC.

Full size image

Thus, the image of 5G is constructed differently in the two corpora. CC constructs an utterly positive image of 5G, devoting more attention to its largescale, influential and trustworthy role, its commercialization and its widespread applications. In contrast, despite the reference to the fast feature of 5G, BC also focuses on the misinformation and conspiracy theories linking 5G to COVID-19. In doing so, scientific uncertainties and controversies concerning 5G are induced in BC, which is shaped by and may in turn impact public perceptions of 5G. Such uncertainties and controversies arise from a conflict of opinions or contradictory media reports (Retzbach and Maier, 2015) around the benefits and risks of 5G, such as the extensive description of anti-5G activities (lines 1–9 in Fig. 1), social media commentary linking 5G to health and coronavirus (lines 1, 4 and 7 in Fig. 2) and the rebuttals of the false claims (line 10 in Fig. 1 and lines 2–3, 5–6 and 8–10 in Fig. 2).

Of the 18 instances of linking 5G, 16 concordance lines point to the negation of the false claims or conspiracy narratives, suggesting the perception of countering disinformation around 5G is dominant in BC and the critical role of the government during this process (lines 2 and 7–8 in Fig. 2). Nonetheless, the warning narratives concerning countering misinformation and conspiracy theories may be ineffective or even counterproductive, especially those with bland rebuttals and inexplicit explanations (Ross et al., 2018). We further inspected the 16 instances of linking 5G denoting the negation or warning narratives about conspiracy claims linking 5G and coronavirus. Results show that all 16 instances are just simple warnings or negations without specific advice to audiences on how to act. Considering the Streisand effect, the publicity of these narratives may lead to growing attention to such content as a result of endeavoring to curb it (Chipidza and Yan, 2022). Put differently, the frequent mention of false claims about 5G in BC could render such false claims newsworthy, make individuals more convinced of their beliefs and further escalate uncertainty rather than prove the veracity of a claim. After all, 5G conspiracy theories are newsworthy in themselves, since “people enjoy conspiracy theories, preferring them to bland reports about science and mundane politics” (Machin, 2022, p. 22).

The social actor Huawei

A constellation of social actors is involved in developing, testing and deploying 5G, including operators, equipment vendors, government and other related stakeholders. As Table 3 shows, the subsequent three sub-categories are classified under the category social actors, including telecom operators and equipment suppliers, countries, cities, institutions, government officials and corporate executives. The technical nature of 5G contributes to the significant role of telecom operators and equipment suppliers among all social actors. In Table 3, both corpora pay attention to respective telecom operators in their countries, such as (China) Unicom, (China) Mobile and (China) Telecom in CC, and O2, Vodafone and BT in BC. Meanwhile, they focus on a range of worldwide equipment suppliers, such as Huawei, ZTE, Qualcomm, Nokia, Ericsson, Samsung and OPPO in CC, and Oneplus, Google, Qualcomm, Nokia, Ericsson and Samsung in BC. Huawei ranks second in the keyword list of each corpus in Table 2 and first among all keywords denoting social actors in Table 3, demonstrating that Huawei forms a staple part of media coverage of 5G in either corpus. Moreover, the frequent occurrence of 5G’s two collocates Huawei and Huawei’s in Table 4 suggests that 5G is typically represented by Huawei in either corpus. Indeed, Huawei functions as the sole entity capable of manufacturing “at scale and cost” all components of a 5G network (Kaska et al., 2019). Thus, we chose Huawei as one main research object for further scrutiny, with a particular focus on the two corpora’s competing ways of portraying it.

Table 5 outlines the 15 most frequent collocates of Huawei* at the L5-R5 position in both corpora and the six categories these collocates are classified into. It is unsurprising to find collocates bound up with Huawei-related actors and 5G’s technical nature in both corpora, occupying 86.25% (in CC) and 69.88% (in BC) of all occurrences of the 15 most frequent collocates. The collocate giant, which points to the leading and technologically advanced role of Huawei in the telecom industry (3.09%), is unique to CC, casting a positive tone in Huawei’s image. In contrast, the category related to the British decision to allow Huawei a role in building part of its 5G network (22.12%) and the category about 5G security (8.00%) are unique to BC.

Table 5 The 15 most frequent collocates of Huawei* grouped into six categories.
Full size table

A concordance analysis of Huawei-decision (see Fig. 3) points to Britain’s changing attitudes towards Huawei, from a decision to allow Huawei’s involvement in its 5G network building (lines 1–2, 8–9, 12 and 14) to the reverse of that decision (lines 3–6, 10 and 15). Besides, the frequent occurrence of the US and its government officials and companies in Fig. 3 (lines 2, 5–6, 9 and 13) demonstrates the role of the US in influencing Britain’s changing decisions, such as Britain’s consideration for the US-Britain relations (line 2), the US ban on Huawei (lines 5 and 13) and the US’ pressure to ban Huawei from Britain (lines 6 and 9).

Fig. 3
figure 3

The sample concordance lines of Huawei-decision in BC.

Full size image

The role of the US in impacting Britain’s decisions and the triangular nature of the US-Britain-China relationship concerning 5G are further accentuated through the concordance analysis of Huawei-security (see Fig. 4), such as the US antagonism towards Huawei (line 5) and the labeling of Huawei as a security danger or risk by the US (lines 1–2, 4, 6–9, 12 and 15). Figure 4 also indicates that the discursive strategy of negative out-group representation is used in BC to construct Huawei as a security risk, whether by Britain (lines 1, 10–11 and 13) or the US (lines 2, 4 and 6–9).

Fig. 4
figure 4

The sample concordance lines of Huawei-security in BC.

Full size image

The concordance analyses above show that BC attempts to ascribe the British decision to the pressure from the US, through which Britain is set out as one playing field in the US-China dispute over technology companies and 5G. It was alleged that the British government privately told Huawei that geopolitical factors were instrumental in their decision on banning Huawei’s 5G, particularly the intense pressure from President Donald Trump (Helm, 2020). This consistency can be perceived as the ideological alignment of the British press with the government. Likewise, this underpins the finding that Huawei in BC is depicted more as an incendiary geopolitical topic than a technical or security concern. Political inclusion is implicitly attributed to the US legitimized as an in-group, and political exclusion to China justified as an out-group. As a result, the image of Huawei should be understood in a geopolitical sense, equating Huawei as a symbol for its home country of China and its 5G as a geopolitical battleground. This finding resonates with Mansell and Plantin’s (2020) argument that the British press constructs Huawei as a problem with very few positive mentions of Huawei and 5G. A comparison of the frequency of Huawei (1884 instances for CC and 2901 instances for BC) reveals that BC significantly overuses Huawei compared to CC (LL = 26.54, p < 0.0001, log ratio = 0.22). LL value over 15.13 indicates that the difference is statistically significant at the level of p < 0.0001 (see the scale of significance level at the above UCREL website). This finding suggests that compared to CC, BC prefers to center its 5G-related narratives on Huawei which is presented as a negative label. This stands in notable contrast with CC where Huawei is portrayed more as a positive and trustworthy company by using the discursive strategy of positive self-representation.

Furthermore, the frequent mentioning of the pressure from the US and its sanctions on Huawei, in a sense, has some meaning entailments that are not explicitly expressed by the British press. That is, the British press does not want to assume full responsibility for explicitly asserting that Huawei’s 5G poses security risks on account of communicative reasons, for instance, because they do not want to make enemies of Chinese recipients or telecom operators. These not explicitly spelled out meaning implications can be taken as “a form of self-protection or positive self-presentation” (van Dijk, 2008, p. 184). Here, argumentation strategies, particularly “shifting the blame” and “scapegoating” strategies, are used to shift the responsibility associated with Britain to the US. These strategies can help legitimize and justify the British government’s behavior of removing Huawei from its 5G networks.

Actions concerning 5G

The keywords denoting actions about 5G in Table 3 reveal the development, application and deployment of 5G. The shared keyword rollout indicates the emphasis on the active 5G rollout and deployment in both corpora. Actions concerning 5G industries commonly happen in three segments: the upstream referring to research and development, investment and production of components like 5G and chips; the midstream referring to infrastructure construction like 5G networks and supercomputing centers; and the downstream referring to applications and services (Zhang, 2023, p. 820). Keywords denoting 5G upstream merely occur in CC, such as trillion and yuan. Instead, both corpora encompass keywords denoting 5G midstream (networks and infrastructure in CC and masts in BC) and downstream (smartphone(s), commercialize(-ation), gaming, healthcare and licenses in CC and smartphone(s) and driverless in BC).

All keywords denoting 5G downstream in both corpora indicate their emphasis on 5G applications in civilian/commercial domains, such as in fields of smartphones, gaming, healthcare and driverless cars, as revealed by smartphone(s), gaming and healthcare in CC and smartphone(s) and driverless in BC. Keywords commercialize(-ation) and licenses, though not literally denoting concrete application fields, reflect CC’s focus on 5G commercialization and granting of 5G licenses for commercial applications.

To examine the two corpora’s differences in representing 5G actions in the upstream and midstream, we retrieved the 10 most frequent collocates at the L5-R5 position of networks, infrastructure, trillion and yuan in CC and masts in BC (see Table 6). The following findings were observed by analyzing the collocates denoting actions in Table 6. First, the two collocates investment and invest of keywords trillion and yuan suggest CC’s particular attention to 5G upstream investments. Second, networks and infrastructure in CC frequently collocate with words reflecting positive actions concerning 5G midstream infrastructure construction, such as build, construction and building. Quite the contrary, masts in BC tends to collocate with words indicating negative actions, namely attacks on 5G masts, such as attacks, arson and attacked.

Table 6 The 10 most frequent collocates of networks, infrastructure, trillion, yuan and masts.
Full size table

In comparison, there exist similarities and differences between the two corpora concerning the representation of actions about 5G. Some macro-level socio-political, economic, and ideological factors could be interweaved to explain such similarities and differences reflected at the micro-level language analysis. Both corpora accentuate 5G downstream applications in civil/commercial domains, which is unsurprising since 5G commercialization applications are instrumental in boosting economic growth (Zhang, 2023). Instead, CC puts more weight on comprehensive actions in the 5G upstream investments, midstream infrastructure construction and downstream applications, whereas BC focuses solely on the latter two aspects. This finding can be explained by the 5G development landscape and practices in China and Britain. China, as a global leader in 5G, has been seeking a balanced and comprehensive 5G development covering upstream, midstream and downstream segments (Zhang, 2023). Britain, despite being one of the first countries to begin implementing 5G commercially, is in danger of falling behind due to limited investments and its early dependence and later ban on Huawei (Sugiura et al., 2023).

Moreover, although both corpora mention the action concerning 5G midstream, their representations vary. CC focuses on positive-tone 5G infrastructure construction, whereas BC is more concerned with the arson attacks on 5G masts driven by coronavirus conspiracies. China Daily, as one important window for the world to understand China, tends to put a positive spin on the country (Wang, 2018). BC’s emphasis on reports on 5G arson attacks on 5G masts owed more to the lack of social acceptance of 5G in Britain during the pandemic (Meese et al., 2020) and Britain’s decision to ban 5G under the political pressure from the US (Parker et al., 2020).

Timing: years and the COVID-19 era

Unsurprisingly, keywords denoting the years 2018–2022 occur in either corpus, since 2018 marks the year when countries like the US, Canada and Australia banned Huawei from their 5G mobile networks. In contrast, the unexpected occurrence of the future year 2025 is unique to CC in Table 3. A concordance analysis of 2025 (see Fig. 5) provides a strong indication of the optimistic forecasts concerning 5G, such as the huge economic and employment rewards created by 5G between 2021 and 2025 (line 26), China’s becoming the world’s largest 5G market before 2025 (lines 27, 33 and 35), China’s significant 5G capital expenditure and investments before 2025 (lines 28, 32 and 34), and the great number of 5G users and connections by 2025 (lines 29–31). Results show that CC characterizes 5G as an economically beneficial technology receiving proactive support from the Chinese government and takes 5G as a symbol of its rise in the global ICT sectors and a source of China’s national pride. From a CDS perspective, the way of representing the year 2025 in CC is positioned as an expression of social practice concerning China and its 5G development. Following one Deloitte (2018) report, China has made more tremendous strides in 5G investment and capacity development than other countries.

Fig. 5
figure 5

The sample concordance lines of 2025 in CC.

Full size image

Apart from the years mentioned above, both corpora embrace some keywords denoting the era of COVID-19. To explore the 5G/COVID-19 relationship at greater depth, we further analyzed the collocates of COVID at the L5-R5 position (Table 7). The collocate 5G in Table 7 shows that both corpora establish an intimate relationship between 5G and COVID-19. Instead, a close concordance analysis of COVID and its collocate 5G (see Figs. 6 and 7) shows that they differ in construing the 5G/COVID-19 relationship.

Table 7 The top 10 collocates of COVID in both corpora.
Full size table
Fig. 6
figure 6

The concordance lines of COVID-5G in CC.

Full size image
Fig. 7
figure 7

The sample concordance lines of COVID-5G in BC.

Full size image

As Fig. 6 shows, CC devotes more care to how to deploy 5G to fight against COVID-19 (lines 2 and 4) and China’s leading role in 5G development despite the COVID-19 pandemic (lines 3 and 5–8), which can be reinforced by the collocate fight in Table 7. Nevertheless, regarding BC, the keyword misinformation in Table 3 and the collocates conspiracy, claims, theories, linked and link in Table 7 demonstrate that the COVID-19 theme in BC is more concerned with the conspiracy theories or claims linking 5G technology and coronavirus, which can also be evidenced by the concordance lines of COVID-5G in BC (Fig. 7, lines 1–10).

In order to probe into how the 5G conspiracy claims pertain to COVID-19 in BC, we thoroughly surveyed the concordance lines of 5G-COVID, 5G-coronavirus and 5G-virus. The following five extracts from the aforementioned concordance analyses are representatives of 5G-conspiracy claims in BC. The 5G-health conspiracy is that 5G causes the symptoms of COVID-19 (Extracts (1)-(2)); the 5G-vaccine conspiracy is that 5G technology is used to track the population through vaccines containing computer chips (Extract (3)); and the 5G-Wuhan-coronavirus conspiracy is expressed by blaming Wuhan’s 5G construction for the COVID-19 outbreak (Extracts (4)-(5)). With this, it is found that the 5G conspiracy claims about COVID-19, especially about its origin and spread, are mainly made with two strains, namely constructing 5G as an issue causing adverse health effects and as an intentional mind-control tool.

  1. (1)

    “… the symptoms of coronavirus are caused by residing near a 5G mast.

    (The Guardian, 2020/04/04)

  2. (2)

    When it comes to blaming 5G for the Covid-19 pandemic, a conspiracy …: 5G can lead to “flu-like symptoms”, …

    (The Daily Telegraph, 2021/08/16)

  3. (3)

    coronavirus pandemic is cover for a global plot to install 5G mobile phone masts, track the world’s population through vaccines …

    (The Guardian, 2021/06/28)

  4. (4)

    … the Chinese city of Wuhan, where the virus is thought to have originated, was one of the first places to receive 5G masts.

    (The Times, 2020/04/07)

  5. (5)

    … there could be a connection between Wuhan having 5G and the coronavirus outbreak, …, 5G is a tool for mind control.

(The Times, 2021/05/14)”

Despite the identification of COVID-19 discourse in either corpus, the 5G/COVID-19 relationship is constructed differently in the two corpora. CC predominantly revolves around the technical nature of 5G in the COVID-19 context and the benefits it creates for COVID-19 containment. In contrast, BC prevalently reports the negative aspects of 5G, particularly the 5G/COVID-19 conspiracies. Looking beyond the conspiracy claims to a set of larger, underlying and motivating geopolitical concerns, these conspiracies make 5G negatively connotated as a dangerous or untrustworthy issue in the COVID-19 context. Moreover, they have provoked increasing scepticism over public health institutions and measures, as reflected by anti (107, 0.26‰), the fourth collocate of 5G in BC. The most frequent collocates of anti at the R1 position are 5G (37), Huawei (10), government (7), lockdown (7) and vaccine (6), unveiling the intricately intertwined environment of anti-5G, anti-Huawei, anti-government, anti-lockdown and anti-vaccine. Contrastingly, anti (8, 0.03‰) rarely occurs in CC. A comparison of the frequency of anti in the two corpora yielded a log ratio of 3.34 (LL = 75.88, p < 0.0001), indicating that the occurrence of likelihood is more than 8 times more common in BC compared to CC. BC’ significant overuse substantiates its focus on scepticism towards 5G.

Such differences can be attributed to various socio-economic and political contexts in which the Chinese and British press operate. Understandably, the positive self-representation strategy is employed in CC to foreground the promising and socio-economically profitable nature of 5G since Huawei, a global leader in providing 5G technology, is a China-based company (Cheng and Liu, 2022). The emergence and prevalence of the 5G/COVID-19 conspiracy in BC need to be interpreted within a more complicated and wider domestic and international context. As Bruns et al. (2022, p. 26) argue, the 5G/COVID-19 conspiracy itself is “a product of the collision of long-standing conspiracist beliefs about the supposed health dangers of 5G, as well as about vaccines, global elites, China and other well-established targets of suspicion”. It is also germane to Britain’s policies and decisions about 5G, which drew considerable public attention towards 5G and amplified the circulation of 5G/COVID-19 conspiracy (Bruns et al., 2022, p. 18). For instance, the articles on Facebook surrounding the 5G/COVID-19 conspiracy emerged when the British government made a major policy decision (Kleinman, 2020), allowing Huawei to participate in developing Britain’s 5G network.

Discussion and final remarks

To sum up, this comparative study shows that the two corpora have disclosed both similarities and differences in their representations of 5G, specifically in the following four dimensions: the nature of 5G, Huawei as a crucial social actor, actions around 5G and timing (particularly the COVID-19 era) in 5G discourse. Regarding the nature of 5G, the highly frequent positive evaluative collocates of 5G suggest that CC tends to use predication strategies to depict 5G as a trustworthy issue and a revolutionary source of societal change in an utterly positive manner. In contrast, BC prefers to describe 5G with negative evaluative patterns such as anti-5G and linking 5G to COVID-19, suggesting its emphasis on framing 5G as an issue with scientific uncertainties and controversies. Concerning the crucial social actor Huawei, CC is inclined to use words with positive connotations to collocate Huawei* such as giant, indicating its emphasis on profiling Huawei as a leading and technologically advanced company. Nevertheless, BC is prone to collocate Huawei* with words with negative connotations such as security, risk and decision, suggesting its focus on constructing Huawei as a security concern and a geopolitical issue.

The themes and collocation analyses of keywords related to actions taken to 5G reveal that CC attends closely to a comprehensive picture of China’s 5G-related actions, covering 5G upstream investments, midstream infrastructure construction and downstream applications, whilst BC focuses merely on the latter two dimensions. Though the two corpora share keywords relating to 5G midstream actions, CC prefers to collocate these keywords with positive verbal and nominal collocates such as build, construction and building. In contrast, BC tends to collocate them with negative collocates such as attacks, arson and attacked that are related to 5G/COVID-19 conspiracies. As regards timing in 5G discourse, the concordance analysis of 2025 and high-frequency collocates of COVID-19 (such as fight) in CC reveal its closer attention to the benefits that 5G brings to the health sector and COVID-19 containment, whereas the recurrent collocates of COVID-19 (such as conspiracy, claims, link and theories) in BC indicate its emphasis on 5G/COVID-19 conspiracies.

From a CDS perspective, their particular ways of constructing 5G discourses could be ascribed to a set of socio-political and journalistic factors, including the development practices of 5G in China and Britain, Britain’s geopolitical interests and its fierce political pressure from the US, nationalist ideology and journalistic values. China Daily, as an official outbound communication channel, is typical in communicating the Chinese government’s voice to the global community and fostering a positive image of China (Liu and Zhang, 2018). The representations of 5G in CC are in tandem with the Chinese government’s positions and policies that prioritize the depoliticization of 5G and the enormous socio-economic benefits created by 5G (Lee, 2020).

BC’s framing of 5G as a geopolitical crucible and a technology freighted with scientific uncertainties could be constrained by the socio-political contexts where it occurs. Britain and the US, as staunched allies, share common values and maintain highly close technology ties (Narin et al., 1997), which explains why Britain eventually bowed to the pressure from the US, labeled Huawei as a high-risk vendor and banned Huawei’s 5G. Besides, this is also in part due to western governments’ protectionist policies that prevent the entry of Chinese-origin technology into local markets, and their concerns over China’s potential access to sensitive data traveling through 5G networks (Hoffmann et al., 2021). Corpus findings thus show that the three British newspapers, despite their different political leanings, represent government policy positions, defend British interests and construct China as a negative other. It is widely found that Anglo-American media, which have long embedded anti-China ideology, tend to construct a negative image of China (Liu and Zhang, 2018; Stone and Xiao, 2007). Meanwhile, BC attempts to adopt the scapegoating strategy to frame a positive self by repeating patterns related to the pressure from the US and its sanctions on Huawei. Thus, we argue that nationalism is mapped onto 5G media discourses by constructing the “self-other” ideological structure in the two corpora. Besides, BC’s greater attention to news about 5G conspiracy theories potentially indicates that it foregrounds the news values of negativity and uncertainty, which could also be seen as reinforcing the ideology of negative other.

As Samaie and Malmir (2017, p. 1363) posit, CDS should not only be used for “describing, interpreting, and, explaining social problems if it is to fulfill its ultimate goal in denaturalizing the ideologies and changing the society for better”. The overemphasis on 5G conspiracies and the geopolitical component of 5G in media could increase the public’s doubts about 5G and hinder the realization of the full potential of 5G. According to one YouGov survey in April 2022, over 44 percent of British consumers do not believe 5G will benefit their lives (Fernandes, 2022). Hence, it is critical for news outlets that serve as essential sites influencing public perceptions of 5G to depoliticize news coverage surrounding scientific issues including 5G by prioritizing their technical or scientific dimensions.

Methodologically, the corpus-assisted critical discourse analysis not only aids in discovering the most salient statistical language patterns that most clearly disclose the two corpora’s respective preferences for 5G constructions, but also helps to describe, interpret and explain the language patterns in broader socio-political contexts. Focusing on whatever themes that are salient in the corpora themselves beyond any pre-set themes, this study detected the themes of actions and timing concerning 5G, which were overlooked by previous corpus-based research on 5G media discourses.

Our study also provides useful directions for media reporting practices of science and scientific controversies. First, news media should report on science and technology like 5G in a more diverse set of topics and place society’s varied viewpoints in context, presenting both its benefits and risks. Second, when reporting scientific controversies, journalists should provide not only warnings or negations but specific recommendations to readers on how to take action. Third, when selecting and presenting news stories, journalists should raise awareness of the power of framing on news accuracy and fairness and the impact of reports laden with ideological biases on economic growth and public attitudes towards science and technology.