key: cord-0986863-x9dg9if7
authors: Martin, Sam; Vanderslott, Samantha
title: “Any idea how fast ‘It’s just a mask!’ can turn into ‘It’s just a vaccine!’”: From mask mandates to vaccine mandates during the COVID-19 pandemic
date: 2021-10-28
journal: Vaccine
DOI: 10.1016/j.vaccine.2021.10.031
sha: f8ca30dc26327dd0e1eb6f23e7aef0838fa886bc
doc_id: 986863
cord_uid: x9dg9if7

Protests starting in the summer of 2020, notedly in the US and UK, have brought together two constituencies: pre-existing anti-vaccine groups and newly formed oppositional COVID-19 groups. The oppositional COVID-19 groups vary in composition and nature, but the central focus is a disagreement about the seriousness and threat of COVID-19 and with the public health measures to control COVID-19. What unites many disparate interests is an aversion to mandates. The compulsion to undertake particular public health activities such as mask-wearing and vaccination is a complex topic of public attitudes and beliefs alongside public health goals and messaging. We aim to analyse social media discussions about facemask wearing and the adoption of potential vaccines for COVID-19. Using media monitoring softwares Meltwater and Talkwalker, we analyse English-language tweets for one year from 1st June 2020 until 1st June 2021. We pay particular attention to connections in conversations between key topics of concern regarding masks and vaccines across social media networks. We track where ideas and activist behaviours towards both health interventions have originated, have similarities, and how they have changed over time. Our aim is to provide an overview of the key trends and themes of discussion concerning attitudes to and adoption of health measures in the control of COVID-19 and how publics react when confronted with mandatory policies. We draw on an already extensive literature about mandatory vaccination policies to inform our assessment, from psychology and behavioural science to ethics, political theory, sociology, and public policy.

Protests against COVID-19 health measures (also termed 'anti-lockdown' protests) beginning in June of 2020, notedly in the US and Europe (including the UK) (Gerbaudo, 2020; Meeker, 2020) 1 brought together two constituencies: preexisting vaccine critical 'anti-vaccine' groups and newly formed oppositional groups to COVID-19 health measures. Sometimes going against government advice and marred by violence, the protests provided a public display of deep dissatisfaction but have rarely been about one topic alone. The result has been a new configuration of public attitudes and beliefs about health, which has been pitted against public health goals and messaging. Both masks and vaccines come from a backdrop of public resistance -although the opposition to vaccines has been a more enduring social trend -mostly because vaccination forms such a large part of public health and the existence of a variety of vaccine mandates across countries (Attwell et al., 2019; Vanderslott & Marks, 2021) . The need for mask-wearing on the other hand has been mostly restricted to healthcare workers in their occupational setting. Aside from facemasks being more habitual and accepted in some countries (such as Japan and Taiwan: see Burgess & Horii, 2012; Chen, 2021 , which we will not address specifically in this paper), they have also played an important role to limit the spread of infection during outbreaks and pandemics.

While opposition to vaccines -especially the formation of organised groups -was an earlier response to public health mandates, opposition to masks similarly came in a reaction to mandates, but these occasions have been limited to outbreak and pandemic contexts. For example, during the Spanish flu pandemic (1918) (1919) (1920) , opposition occurred in San Francisco in 1919 when the authorities attempted to put in place a mandate for mask-wearing in public places, which prompted the formation of the Anti-Mask League (Dolan, 2020). However, this protest movement was only limited to this specific event. Incidentally, the roots for pro-mask-wearing behaviours in Japan can also be traced back to the Spanish flu pandemic, although Burgess & Horii (2012) argue that the practice only became socially embedded in the 1990s through "commercial, corporate and political pressures that responsibilised individual health protection" (p. 1184). In comparison, early 'Anti-Vaccination Leagues' against mandatory smallpox vaccination that began in the 19th Century in Europe and US, have evolved into organised and long-standing anti-vaccine groups today (Willrich, 2011; Rosner, 2017) . Small but vocal, these (sometimes very loose organisations) are especially active on online media and social media (Germani, 2021) . A very recent development has been how well-established public vaccine opposition has mobilised against masks and the connections that have forged between opposing both vaccines and masks.

To inform our assessment of this convergence between interests of groups of publics on health issues, we draw on previous literature about mandatory vaccination policies, from psychology and behavioural science to ethics, political theory, sociology, and public policy. Very recently a literature on attitudes to mask-wearing has developed through psychologists, interested in knowing the factors that drive use (Mallinas et al. 2021) , including by demographic group (men vs women) (Palmer & Peterson, 2020; Mahalik et al, 2021) , as well as political scientists and public health researchers focused on mask messaging (Utych, 2020) and culture change (Mello, 2020) . Through this paper we aim at understanding how social media discussions and debates about mandates from mask-wearing to vaccination were connected through their opposition to COVID-19 public health policies and have evolved over the course of the pandemic.

What we uncover is a complex topic of public attitudes and beliefs in reaction to public health goals and messaging. Particular attention was paid to connections in conversations between key topics of concern regarding masks and vaccines across social media networks. To analyse dominant topics of discussion, we concentrated on stance and associated language used on Twitter about public health activities to control outbreaks. We used media monitoring software Meltwater for one year to track social media conversations and make an assessment of key trends and themes of discussion concerning attitudes to and adoption of these health measures to control COVID-19 and track the public reaction when confronted with mandatory policies.

Social media data was collected from Twitter using media monitoring software Meltwater™ (Meltwater, 2021) . Tweets were collected where there were mentions of supportive or oppositional mask and COVID-19 vaccines, between June 2020 and June 2021. We chose to limit our geographic scope to the UK and US as countries where high profile protests had occurred and where political events we strongly intertwined with government and public responses during the pandemic. An initial general Boolean search (Appendix a.) was conducted to get a broad overview of the main pro and anti-mask and COVID-19 vaccine topics. "Pro" in this context means an overall positive attitude towards the wearing of masks and the protection offered by vaccines within the context of the COVID-19 pandemic. "Anti" in this context means an overall negative attitude towards the wearing of masks and vaccines within the context of the COVID-19 pandemic. Clearly the distinction between pro and anti is more nuanced than this binary. However, we found that in the context of protest and public dissatisfaction it was necessary to seek extreme views in order to understand the discourse.

A total of 7.89 million (abbreviated to 'm' from here on) tweets were collected and analysed from Twitter for a broad overview of themes and co-related hashtags occurring over the period between June 2020-2021. We focused on Twitter as a social media platform because it produced the most data that was time-linked, to view responses to events over the course of the year, and also acted as a signpost to relevant content on other social media platforms. Once an overview of the data was established, we conducted a second more focused Boolean search (Appendix b), which included keywords and hashtags found in the initial data sample, where usage referred to people's opinions and experiences of masks and vaccines.

We then conducted a discourse analysis using text network analysis software Infranodus (Paranyushkin, 2019) to measure themes and patterns occurring in discussions around masks and vaccines, as well as the betweenness centrality of subtopics -an analysis of connections between subtopics that link different types of conversation clusters together to identify, including similarities in behaviour. We imported data into Infranodus to highlight key word clusters, and thus influential topics used in discussions on social media. Semantic networks were generated by organising the data by specific topic (masks and vaccines) and analysed each topic separately to find specific subthemes within each one. Word clusters derived from analysis using modularity class were used to highlight the most common topics of conversation within a cluster of tweets. Keywords derived from analysis using betweenness centrality were used to give a more indepth understanding of influential words that link different clusters of conversations together. Word frequency analysis was then performed on the dataset to provide insight into the most frequent and weighted subtopics of conversation. This provided a more detailed content review of topics generated by Infranodus .

We explored discussions of anti-mask views that also mention vaccines, where mentions of vaccines also occurred in conversations regarding confusion about government mask policy. Once the key themes were identified, individual tweets were selected for textual analysis in order to draw out specific issues for a more qualitative interrogation of topics of concern at hand. Our approach to analysis draws on Lewis et al. (2014) where empirical materials are treated not only as data that is freestanding but also as an 'empirical trigger' to describe and analyse research topics. We employed a constructivist grounded theory (Charmaz 2009 ) for our interpretation of the empirical material, following the principles of an iterative and reflexive process.

In using market analytical software (such as Meltwater) to assess sentiment we are aware of the difficulty of accurately classifying sentiment or stance of a tweet or post within the correct context or perspective of the study subject being researched (Wouter van Atteveldt et al., 2021). Thus, drawing on previous research Kummervold et al., 2021) , we aimed to get a better understanding of sentiment by creating a 'manual sentiment framework'. To break out of the grip of marketing bias within the commercial software, this framework used definitions based on our analysis to 're-annotate' a smaller subsample of tweets, and applied a more specific qualitative lens. Under the framework, sentiment was measured in terms of differences in attitudes towards masks and vaccines within the context of the pandemic (Appendix c, d). Tweets were classified as positive towards masks or vaccines if, for example, they were affirming of masks or vaccines or communicated overall trust of masks or vaccines. Tweets were marked as negative if they contained negative attitudes or arguments against masks or vaccines, shared bad experiences or discouraged the following of public health guidelines. Tweets were then marked as neutral if they contained only a general statement, with no expression of sentiment or opinion.

Between June 2020 and June 2021, we used Meltwater TM to perform a global search of Twitter, and we found that across a sample of the UK and US, there were 7.89 million English language tweets mentioned both pro and anti opinions regarding the wearing of masks and COVID-19 vaccines. The most messages about masks and vaccines were from the US (6.98m), followed by the UK (915k ( Figure 1 ). Analysis showed that overall, discussions about masks were most dominant between June 2020 and June 2021, with 5.91m specific mentions of masks, and 1.98m specific mentions of the COVID-19 vaccine ( Figure 2 ). The dominance of discourse around masks may have occurred due to events during the most of the first year of the pandemic, where masks were seen as the main source of protection against COVID-19, and were mandated as necessary in some countries, while work on an effective COVID-19 vaccine was underway (Mitze et. al, 2020) .

Key points of discussion with regards to masks and vaccines were between 25th June -19th July 2020, 12th August to 17th September 2020, 1st October -25th November, and then January and March 2021. Between 25th June -19th July 2020, a total of 810k tweets in the UK and the US discussed mask mandates, and a series of anti-masks protests and anti-lockdown protests held both in the UK and the US. From 12th August -17th September, the main debate occurring on Twitter centered around initial political discussion from 12th August to official tweets on 16th August by both US President Joe Biden, and Vice President Kamala Harris calling for a federal mask mandate to be issued nationwide across the US, as well as an acceleration of the development of treatments and vaccines.

Between 1st October and 25th November 2020, debate across social media platforms also focused on masks and vaccines most specifically in response to tweets by President Joe Biden about mask mandates and vaccines, as well as discourse in the US about the need to socially distance and wear masks around the US national Thanksgiving holiday (26th November 2020). Discussions around masks also spiked around 21st January 2021, in response to a tweet by President Joe Biden, who declared that wearing masks was a patriotic act, and not a partisan issue, and that he had signed an executive order issuing a mask mandate after months of cross-party debate. On 2nd March 2021, the spike in discourse centered around a tweet by a Texas Governor with regards to him ending the mask mandate across the state of Texas as a direct rebuke of President Joe Biden's executive order -debate around this topic was both positive and negative amongst both anti and pro-mask supporters. The content and discourse analysis of all of these topics is further discussed in the next sections about the support and opposition towards masks and vaccines.

In the UK, the discussion points also involved mask mandates but included a wider commentary about the wearing of masks by high profile figures and government policy around both masks and vaccines. One of the most critically engaged with tweet in the UK was by television presenter Piers Morgan on 10 th July 2020, where he criticised the London Mayor, Sadiq Khan for not wearing a mask while doing an interview focused on mask mandates. In the Interview, Khan said that he found it astounding that people found it a sign of strength to see conservative leaders such as Prime Minister Boris Johnson and former President Donald Trump did not wear masks. The discussion around Piers Morgan criticising the Mayor for not wearing a mask, while he admonished other leaders, drew more engagement than the Mayor's actual post, with 11.8k likes and shares criticising Khan, with a reach of 7.63m, compared to the 521 retweets and shares of the original tweet. Between July 15 th -27 th 2020, there was also further critique of mixed messaging from the UK government, with regards to when and where people were required to wear masks when shopping or eating food in cafes and restaurants, most specifically in relation to the £100 GBP fine, that police were allowed to issue. Between 14 th February and 6 th April 2021, the topic of vaccine passports was prominent in the UK, with a high volume of debate around the use of them for both travel and large events, overall, a favourable discussion centered around travel, whereas there was more mixed debate with regards to access to night clubs and larger venues. Between 22 nd February and 23 rd March, discussion in the UK about vaccines, centered around both access through a staggered rollout via priority group, as well as discussion about the effectiveness of vaccines, and their ability to prevent serious illness or death. There were also questions about the vaccine safety, vaccine hesitancy amongst younger people and black and minority ethnic groups, as well as questions about whether it was safe for younger children to get the vaccine. Discussions around masks in the UK also peaked between 3 rd -14 th March 2021, when parents opposed to a mask mandate for children returning to school tweeted their opposition to schools imposing mask guidelines and pushed back at government responses.

Discourse regarding vaccines in both the UK and US can be seen to have started to rise between 4th -10th December 2020, as different vaccines were certified for public use, and public health programmes got underway. From 14th February to 25th March 2021, there was increased discussion in both countries about the efficacy, safety and supply of different vaccines, as public health programmes rolled out vaccines. However, big peaks occurred between 25th March -1st April and 2nd -8th April, where there was increased discussion about the prospect of vaccine passports being made mandatory for global travel and large events (de Figueiredo, Larson and Reicher, 2021; Osama, Razai and Majeed, 2021). Overall, social media sentiment in terms of anti-mask and anti-vaccine attitudes across both countries was similar, with the UK being slightly more negative towards masks (28% compared to the US at 22% negative), and both having similar rates of negative sentiment towards vaccines with the US at 17% and the UK at 15% (Figure 3 , and Appendix b). 

Through a discourse analysis of the conversations about masks and vaccines, we can organise the social network discussions across four themes. Subthemes that ran throughout these were about social norms, beliefs, and discomfort or inconvenience. The four key themes of discussion were:

1. Support of masks and vaccines (Pro); 2. Opposition to mask and vaccines (Anti); 3. Conspiracy theories; 4. Public health messaging and scientific uncertainty. We discuss these themes next. In identifying reasons in the anti-mask discourse, we found that one of the main topics of conversation was around mask mandates and mixed government messaging about the effectiveness of masks. In the pro-mask discourse ( Figure  4) , while people supported the wearing of masks as protection against COVID-19, there was also confusion about mixed, changeable and unclear public health messaging -especially the official advice regarding what types of masks were appropriate to wear and when (Kottasová, 2020) . Early on in the pandemic, the World Health Organization (WHO) had recommended that masks not be worn at all by the public, as they did not see their effectiveness for general use and wanted to reserve masks for healthcare workers and people known to be infected with COVID-19, but as evidence mounted, they were forced to reverse their position (Feng et al., 2020) . The messaging then evolved to encourage and mandate mask use in public places. The social media conversations in response largely related to school and business contexts (including wearing masks in lifts), and the differing mandates across geographical areas (in state jurisdictions for the US and in the different nations of the UK), as well as rationale for central government advice and decisions (e.g. the Centers for Disease Control and Prevention -CDC -lifting of the mask mandate for vaccinated individuals).

The reasons for supporting the wearing of masks were broadly split with those who referred to a "mask vaccine" emphasised in the first six months (June -Dec 2021) -that using a mask had a similar effect to being vaccinated and should be used as much as possible before vaccines were widely available. Others viewed the good behaviour of wearing a mask as needing to be rewarded and deemed that those who did not wear masks, should be punished by not being allowed to get a vaccine, see box below.

Pro-mask support: Calls for people anti-mask wearers to be punished by not getting vaccine "No Covid denying politician should get a vaccine before every Dr, nurse, hospital worker, EMT, cop, fire fighter, teacher, supermarket worker, pharmacy worker, foodchain worker, book store worker who has to deal with abusive no mask a-holes & many others @user "Young, healthy Senator, who spoke at rallies packed w/thousands w/o masks, who supports Trump -who's down-played COVID & mocked those who wear masks, is 1 st to get vaccine while most medical workers, elderly & infirm Americans, wait. Congratulations on ur privilege, @user".

While the wearing of masks in schools was also discussed in both the UK and US, many saw mask wearing as a good option for preventing the spread of the virus from schools and out into the wider community as lockdown was lifted and children also accompanied adults to shops and other businesses. Discussion of the comfort or discomfort that children experienced while wearing masks was present, while others also worried that authorities were not taking mask wearing seriously enough, especially when children were not able to be vaccinated, and were deemed to have less protection (despite official status quo that children were less likely to be badly affected by . There was also discussion of anti-mask aggression, where anti-maskers in Europe were reported as ripping off the masks of those policing the protests (Guardian, 2020).

With regards to COVID-19 vaccines, masks remained a key theme in discussion, with people discussing anti-mask protests, also including anti-vaccine protests and the danger of going unmasked in large crowds and on public transport, before a high proportion of the public were vaccinated. Most of the discussion by pro-maskers about anti-maskers also conflated their actions with those who were also opposed to the COVID-19 vaccine, with the view that voiced opposition to masks and vaccines suggested that they were putting others (and their own) lives at risk. The top 10 pro-mask co-occurring hashtags used alongside these discussions were specific to the UK and US with: #maskupuk and #maskupamerica, along with: #nomasknoentry, #mandatemasksnow, #MasksSavelives, #masks4all, #nomasknopass, #maskisamust, #maskswork, #maskup. Pro-vaccine discourse focused on discussions of hopes and the belief that the vaccine would work to protect against COVID-19, as well as discourse around the vaccination process itself. What entailed was a discussion about the staged rollout of vaccines across the UK and US, with different age-groups being eligible to receive the vaccine at different points in time from December 2020 onwards. Most tweets were shared alerts about what age groups could access vaccines, which type of vaccine was available, and where in each country or region these could be accessed. There was also discussion about the safety of different types of vaccines, and an emphasis that the overall risk of reported vaccine side effects (e.g. Oxford-AstraZeneca and Pfizer-BioNTech) was worth the benefit of protection against COVID-19. The pro-vaccine hashtags #vaccineswork, #VaccinesSaveslives as well as #gotvaxxed, #1stjab and #2ndjab were also shared when people received their doses of vaccines, along with the type of vaccine that was received (#Pfizer, #AstraZeneca, or #Moderna). In relation to this were discussions of access to vaccines, worry about the amount of vaccines available, and whether vaccine passports would be needed to travel and access social events going forwards. In terms of vaccine passports, the hashtag #vaccinepassports was shared in relation to discussion of government messaging around the future issuing of this, whether it would be enshrined in the country law or just be part of international travel regulations.

(2). Opposition to mask and vaccines (Anti) Figure 6 . All Anti-mask discourse: Most influential topics (June 2020-2021) -(Link to hi-res)

The main themes that arose and were connected to each other in anti-mask discourse against mask mandates were discussions about individual liberty vs collective culture and solidarity -especially for mandatory health measures ( Figure 6 ). The literature on vaccine mandates has similarly emphasised these distinctions between individually and collectively orientated values (Giubilini et al, 2019 , Brennan, 2018 . These values related to reconciliation of what could be considered the social norm for certain conduct approved socially, when there had not been previous widespread experiences of wearing masks in some countries. The main topics discussed were about the restriction of people's liberty and right to choose whether they wanted to wear masks or the belief that masks were necessary as a protection against COVID-19. People also focused on the feeling of being uncomfortable about upsetting social norms, in discussing the challenges of having no previous experience of wearing masks, how comfortable masks were, and in some cases not witnessing others wearing masks around them, and not wanting to upset specific no-mask status quos. News articles such as an article in Scientific American (See Denworth, 2020) speculated that masks could become status quo. A UK public survey also showed that people had adapted to the need to wear masks and the majority intended to carry on wearing their masks inside shops and on public transport even once there was no longer a requirement to do so (Office for National Statistics, 2021). After the legal COVID-19 restrictions were lifted in England in 2021, English people stated they wanted the rule for masks on public transport (79%) and in shops (76%) to be reimposed (YouGov, 2021).

One of the oppositions to mask wearing, was a dislike of the feeling of discomfort or inconvenience wearing a mask and being unused to wearing one in public settings. Another worry was expressed about problems with communicating clearly, harming the social development of children (the argument being for the importance of cues in facial expression in learning how to socialise appropriately).

Lastly was a worry about medical conditions, both respiratory, and in terms of those with learning disabilities (i.e. Down's Syndrome and other conditions, who might find it distressing to wear a mask). The worry about mandates related especially to the idea about being forced to wear masks when entering stores, going to a public event, or travelling, as well as questions about the robustness of masks, whether they offered any 'real' protection, and mask slippage or loosening if worn continuously.

"#MaskVaccine! Face diapers are gross -it was always obvious they collect (and spread) pathogens.

Interestingly, where are all the allegedly blocked Covid droplets ?"

The strongest opposition to masks was linked to conspiracy theories, with beliefs that the COVID-19 pandemic was a hoax (#plandemic), that COVID-19 was not a big health threat, and no worse than contracting mild flu, and that it was better to catch COVID-19 and build up a natural immunity, which wearing a mask was thought to impede. Anti-maskers also shared views that people who wore masks without question were blind followers of government control and regulations.

Anti-mask tweet referencing conspiracy theories:

"#MaskVaccine! #MaskCovidians #MasksDontWork #MasksHaveConsequences #MasksOff Quote Tweet: · Mar 27 What's great about masks is that they never have to actually work. Like the Czech Republic, which @DrX said "conquered" Covid with masks, seeing results dramatically worse than countries with much, much lower mask usage.

But none of that matters! Cause experts."

Mask wearers were accused of acting like sheep who were being directed down a hopeless path and protecting against a survivable virus -using the nickname and hashtag #Sheeple to describe them. The hashtags #Wakeup and #Fightback were also used in relation to discourse around this topic, with masks argued to be a symbol of showing how 'scared', 'compliant' people were, as opposed to what was seen as fake liberal propaganda messaging that mask wearing was 'brave' or 'caring'. Véronique Steyer (2020) has highlighted the strong symbolic nature of mask-wearing and Martinelli et al (2021) have argued that a multitude of meanings are conveyed -sociocultural, ethical, and political.

Symbolism: Tweet from Anti-masker, conveying that belief in effectiveness of masks is "tribalism" "#MaskVaccine ! QT: Tribalism > science: "I'm going to wear a mask for a long time to come," said XX, X, who was grocery shopping in San Antonio on Friday morning while wearing a black cloth mask. "I trust the mask more than the vaccine."

It was also argued that mask wearing could have psychological impacts or claustrophobia, raised anxiety and other related issues (Goh et al, 2020) . This view was also linked with suspicion about perceived government control around vaccines, and speculation that a compulsory vaccine mandate would be imposed on people at the expense of choice and liberty. Surveys in US told a similar story and equated the "aversion to being forced to wear masks" as "psychological resistance" (Taylor & Asmundson, 2021, p.1).

In the US in particular, the anti-mask and anti-vaccine discourse was discussed along party political lines, as we have outlined in the peaks of discourse. Some Republican and Trump supporters argued that any mask mandates were Democrat attempts to curb their political and human rights freedoms to choose to not wear a mask. Such views have been reflected in public opinion polls (Pew Research Center, 2020), suggesting that people with conservative political affiliations (such as Republicans) are less likely to wear masks compared with people who have liberal affiliations (Democrats). Understanding the reasons for this difference may not be so straightforward. On the one hand, as people may have been following the example of Republican political leaders who were initially reluctant to wear masks, and some went further to mock those who wore masks. On the other hand, more generally, people with politically conservative ideologies have tended to resist government regulatory efforts, 'intrusion' into the personal sphere, and 'big government.

Anti-mask tweets example, linked to mask mandates and politics: "QT @thehill: Trump could fart and the dems would scream Nazi so forgive me if I dont give a shit ; Greene has sparked new backlash in recent days over comments she made equating mask mandates to the treatment of Jews during the Holocaust."

In cases, in both the UK and US, the hashtags #MaskRevolt and #MillionUnmaskedMarch were used by political groups to organise and share the location of anti-mask and anti-lockdown protests, while in the US, the specific hashtag #MillionUnmaskedPatriots was used in association with protests to mask mandates in Idaho and Texas, as well as states and regions where governors had imposed mask mandates. Other co-related anti-mask hashtags used included #masktyranny, #Covid-1984, #COVID1984, #constitutionovercoronavirus, #Covid-mask-vaccine, #nomask, #masksoff, #masksoffamerica, #MasksoffUK, #maskvaccine, #plandemic, #scamdemic, #notmynewnormal, #masksdontwork, and #nomask_campaign. Therefore, the symbolic and very visible nature of maskwearing and parallels made with mouth-gags and the suppression of speech, was used to organise protests against COVID-19 health measures.

Public Health messaging: Tweet example of mixed government guidance re. Masks: "QT:"Fauci has been trustworthy for the whole pandemic, explaining distinctions between developing theories & verified data, making it easier to understand & trust when recommended practices evolved. Good reminder that for public policy and public health, you have to get the social science right and not just the science ; The CDC mask guidance switched too fast without enough explanation and overlooks key sociological factors for indoor mask mandates-especially to protect workers and the immunocompromised. Better to have announced benchmarks-and kept it up just a bit more." ; The CDC mask guidance switched too fast without enough explanation and overlooks key sociological factors for indoor mask mandates-especially to protect workers and the immunocompromised. Better to have announced benchmarks-and kept it up just a bit more."

Scientific uncertainty was also a theme running through both anti-mask ( Figure  6 ) and anti-vaccine discourse (Figure 7) . While emerging science about the effectiveness of mask-wearing in preventing the spread of COVID-19 has been supportive overall and the evidence is growing that masks limit the spread of infection (Peeples, 2020), it has been complicated to be definitive about how well masks work or when to use them. The sharing of emerging research studies and the impact of academic preprints on global discourse in both the media and online -as people tried to make sense of the COVID-19 pandemic -also led to considerable debate and worry about the reliance of policymakers on scientific research that has not had time to be peer reviewed (Majumder & Mandl, 2020) . Partly due to this kind of uncertainty, and the differing views in emerging literature, we found a number of anti-mask posts sharing examples on Twitter and links to TikTok (see Box 6) of assumed scientific evidence questioning the effectiveness of masks, and in some cases amplifying the discussion of known risks of mask-wearing to certain groups with breathing difficulties (e.g. severe Cryptogenic Organising Pneumonia or asthma) as evidence that mask-wearing was not merely ineffective but actually dangerous. These findings match surveys conducted in the US, that a key reason for not wearing a mask was the view of them not being effective in preventing COVID-19 (Taylor & Asmundson, 2021) . Early in the pandemic, surveys cited reasons for not wearing a facemask in the UK being because they did not feel they needed to if they avoided riskier situations and a sizeable proportion (22%) stated it was because masks are not mandatory Vaccine opposition discourse centered around mistrust of the speed and process of creating the COVID-19 vaccine, mistrust of the contents and microconstituents of vaccine ingredients, including conspiracy theories about whether COVID-19 vaccines contained pork, aluminum, DNA-altering properties (mRNA vaccines such as Pfizer and Moderna). Some also implied that governments had bad intentions to use COVID-19 to control citizens: 'Bill Gates' vaccine or COVID-1984 vaccine with microchip -hashtags such as #covid1984, #Covid-1984 and #VaccineChip were also used with these messages. Such discourse often also contained vaccine misinformation regarding vaccine side effects or vaccine harm, such as reduced fertility, chronic illnesses or birth defects or harm to unborn children.

Many of these tweets also utilised pre-COVID-19 hashtags Muric et. al, 2021) , such as #LearnTheRisk #VaccineKill, #BodilyAutonomy and #InformedConsent. These hashtags have previously been linked to more general campaigns against influenza, measles, and pertussis vaccination by large and organised anti-vaccine groups. They were found here to be used with similar sociolinguistic variations, focusing on perceived vaccine hesitancy and anxieties around the safety of new COVID-19 vaccines and using messages with hashtags like #InformedConsent and "#learntherisk", which appear to communicate concerns about the safety of the vaccines and the need for choice, however are also used with text sharing vaccine misinformation and conspiracy theories about the contents and effects of the COVID-19 vaccine (Memon et. al, 2020) .

The discourse about masks and vaccines in both the UK and US, followed similar themes, which could be attributed to how the flow public dialogue between the two countries (particularly the US to UK) and their similar public health programmes in terms of introducing mask mandates, vaccinating the populations over the months; facing similar public debates about COVID-19 health policies and measures involving masks and vaccines. Through our analysis we were able to identify reasons for opposition of facemasks and vaccines during the COVID-19 pandemic. The main reasons given for support of masks and vaccines (Pro) were protection against contracting COVID-19 and stopping the spread of new variants. The discomfort or inconvenience was discussed even by those who were not completely opposed to mask-wearing. Views were also connected with social norms and beliefs. The main reasons in opposition to masks and vaccines (Anti) were, impositions on freedom, independence and body autonomy, beliefs in conspiracy theories, and worry about side effects of vaccines. Opposition to both masks and vaccines were connected to conspiracy theories. Furthermore, public health messaging and the UK and US government advice were referred to by both pro and anti-groups, using different sentiments. The public health messaging and government advice was set against a backdrop of confusion especially with the official advice (Kottasová, 2020) . Mheidly, and Fares (2020) argue that the confused messaging has not only by government, but across different levels public health communication -from how the media shares scientific information to the overuse of pre-prints as fact, and this then fuels misinformation. Lastly, scientific uncertainty featured and to try to address what was unknown, emerging research studies were shared and explored (often in specific contexts) how effective maskwearing is.

The subthemes that ran throughout were about discomfort or inconvenience (including issues of communication and problems for those with medical conditions or disabilities), and beliefs and social norms. The discomfort and inconvenience of wearing face masks here referred to a dislike of the experience and feeling of wearing a mask. Other points raised were that it caused problems with communicating (e.g. surveys about adherence to public health measures) and medical conditions or disabilities that made mask wearing difficult. Masks are also symbolic of underlying beliefs in their value -being 'scared', 'compliant' or 'brave', 'caring' and so make a statement. Beliefs that have run through include, COVID-19 not being a big health threat and political beliefs, where being anti-mask is aligned with libertarian views. This also spills over into conspiracy theories about mask-wearing for government control. We have outlined how social norms are being reconfigured, with limited previous experience with mask wearing. What has emerged is a conflict in values on individual liberty vs collective culture -especially for mandatory health measures. The learnings from social media analysis of the connection between anti-mask and anti-vaccine views are that while some discussions and subthemes overlap, there is also a connection back to pre-COVID-19 terminologies and hashtags (e.g. #learntherisk and #informedconsent) from previous anti-vaccine campaigns. Future analysis of these connections in terms of both patterns of discourse and group organisational psychology may be of benefit to establishing a more robust public health communications campaign to enable stronger future vaccine uptake.

Mandatory vaccination and no fault vaccine injury compensation schemes: An identification of countrylevel policies

A libertarian case for mandatory vaccination

Risk, ritual and health responsibilisation: Japan's 'safety blanket' of surgical face mask-wearing

Dis/Avowing Masks: Culture, Race, and Public Health between the United States and Taiwan

Appropriate attitude promotes mask wearing in spite of a significant experience of varying discomfort

Shifting the grounds: Constructivist grounded theory methods

Masks Reveal New Social Norms: What a Difference a Plague Makes

The potential impact of vaccine passports on inclination to accept COVID-19 vaccinations in the United Kingdom: evidence from a large cross-sectional survey and modelling study

Rational use of face masks in the COVID-19 pandemic

The Pandemic Crowd

The anti-vaccination infodemic on social media: A behavioral analysis

Nudging Immunity: The Case for Vaccinating Children in School and Day Care by Default

The Face Mask How a Real Protection becomes a Psychological Symbol during Covid-19

Belarus: Women tear balaclavas off security officers amid mass arrests in Belarus -video

The muddled public message on coronavirus isn't just confusing. It's harmful, CNN

Street protests in times of COVID-19: adjusting tactics and marching 'as usual

Categorising Vaccine Confidence with Transformer-Based Machine Learning Model: The Nuances of Vaccine Sentiment within Twitter Discourse

Relocation, Realignment and Standardisation: Circuits of Translation in Huntington's Disease

Men's attitudes toward mask-wearing during COVID-19: Understanding the complexities of mask-ulinity

Early in the epidemic: impact of preprints on global discourse about COVID-19 transmissibility. The Lancet Global Health

What factors underlie attitudes regarding protective mask use during the COVID-19 pandemic

Face masks during the COVID-19 pandemic: A simple protection tool with many meanings. Frontiers in Public Health

Science, society, and policy in the face of uncertainty: reflections on the debate around face coverings for the public during COVID-19

Vaccines for pregnant women…?! Absurd"-Mapping maternal vaccination discourse and stance on social media over six months

The political nightmare of the plague: The ironic resistance of anti-quarantine protesters

Innovation diffusion, social capital, and mask mobilization: Culture change during the COVID-19 pandemic

Social Media Monitoring

Characterizing sociolinguistic variation in the competing vaccination communities

Leveraging media and health communication strategies to overcome the COVID-19 infodemic

Face masks considerably reduce COVID-19 cases in Germany

COVID-19 Vaccine Hesitancy on Social Media: Building a Public Twitter Dataset of Anti-vaccine Content, Vaccine Misinformation and Conspiracies

Covid-19 vaccine passports: access, equity, and ethics

Toxic Mask-ulinity: The link between masculine toughness and affective reactions to mask wearing in the COVID-19 era

InfraNodus: Generating insight using text network analysis

Face masks: what the data say The science supports that face coverings are saving lives during the coronavirus pandemic, and yet the debate trundles on. How much evidence is enough? Nature

More Americans say they are regularly wearing masks in stores and other businesses

Vaccination and Its Critics: A Documentary and Reference Guide

Developing Grounded Theory: The Second Generation

Negative attitudes about facemasks during the COVID-19 pandemic: The dual importance of perceived ineffectiveness and psychological reactance

Messaging mask wearing during the COVID-19 crisis: Ideological differences

Charting mandatory childhood vaccination policies worldwide. Vaccine

van der Velden & Mark Boukes (2021) The Validity of Sentiment Analysis: Comparing Manual Annotation, Crowd-Coding, Dictionary Approaches, and Machine Learning Algorithms

General scoping boolean search for Pro + Anti Mask and Vaccine content

Focused boolean (including co-related hashtags and anti/pro mask/vaccine wordings

OR "constitutionovercoronavirus" OR "constitution over coronavirus" OR "covid-mask-vaccine" OR "covid mask vaccine" OR "maskvaccine*" OR "mask vaccine" OR "nomask*" OR "no mask*" OR "idonotconsent" OR "i do not consent" OR "masksoff*" OR "masks off*" OR "masksoffamerica" OR "masks off america" OR "masksoffuk" OR "masks off uk" OR "plandemic" OR "scamdemic" OR "notmynewnormal" OR "not my new normal" OR "sheeple" OR "1984vaccine" OR "1984 vaccine" OR "vaccinechip" OR "vaccine chip" OR "learntherisk*" OR "learn the risk*" OR "maskrevolt*" OR "mask revolt*" OR "mask mandate*" OR "maskmandate*" OR "compulsorymask*" OR "compulsory mask*" OR "vaccinepassport*" OR "vaccine passport*" OR "vaccinesideeffect*" OR "vaccine side effect*" OR "vaccineswork*" OR "vaccines work*" OR "masktyranny*" OR "mask tyranny*" OR "vaccineskilllives*" OR "vaccines kill lives*" OR "mandatemasksnow*" OR "mandate masks now*" OR

• Post discourages the following of recommended guidelines/support related to wearing masks (for personal

• Post shares bad experiences of wearing masks, and discourages others from wearing masks

Neutral (NT) • Post contains no elements of uncertainty, positive or negative content

• Post contains general statement(s) or link(s) to item(s) (e.g. news articles/papers) with no expression of sentiment. • Post includes factual statements/recommendations about mask wearing in the context of the COVID-19 pandemic

Sentiment analysis of attitudes towards COVID-19 vaccine(s) Positive (P) • Post communicating overall trust and satisfaction with public health guidelines and support for COVID-19 vaccine(s)

• Posts are affirming of getting COVID-19 vaccine(s) to protect against COVID-19

• Post describes the importance of COVID-19 vaccine(s)

Negative (N)

• Post contains negative attitudes/arguments against public health guidelines and support for COVID-19 vaccine(s)

• Post discourages the following of recommended guidelines/support related to COVID-19 vaccine(s) -(for personal

• Post shares bad experiences of COVID-19 vaccine(s), and discourages others from getting COVID-19 vaccine(s)

Neutral (NT) • Post contains no elements of uncertainty, positive or negative content

• Post contains general statement(s) or link(s) to item(s) (e.g. news articles/papers) with no expression of sentiment. • Post includes factual statements/recommendations related to COVID-19 vaccine(s), but no sentiment

We are appreciative of ongoing collaborations at UCL, Oxford, and LSHTM, which have enabled this social media research on COVID-19 to continue. This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

No competing interests to declare.

Both authors developed the paper concept, reviewed the relevant literature, analysed and interpreted the social media data, and wrote and approved the final manuscript.

Protests starting in the summer of 2020, notedly in the US and UK, have brought together two constituencies: pre-existing anti-vaccine groups and newly formed oppositional COVID-19 groups. The oppositional COVID-19 groups vary in composition and nature, but the central focus is a disagreement about the seriousness and threat of COVID-19 and with the public health measures to control COVID-19. What unites many disparate interests is an aversion to mandates. The compulsion to undertake particular public health activities such as mask-wearing and vaccination is a complex topic of public attitudes and beliefs alongside public health goals and messaging. We aim to analyse social media discussions about facemask wearing and the adoption of potential vaccines for COVID-19.Using media monitoring softwares Meltwater and Talkwalker, we analyse English-language tweets for one year from 1st June 2020 until 1st June 2021. We pay particular attention to connections in conversations between key topics of concern regarding masks and vaccines across social media networks. We track where ideas and activist behaviours towards both health interventions have originated, have similarities, and how they have changed over time.Our aim is to provide an overview of the key trends and themes of discussion concerning attitudes to and adoption of health measures in the control of COVID-19 and how publics react when confronted with mandatory policies. We draw on an already extensive literature about mandatory vaccination policies to inform our assessment, from psychology and behavioural science to ethics, political theory, sociology, and public policy.Both authors developed the paper concept, reviewed the relevant literature, analysed and interpreted the social media data, and wrote and approved the final manuscript.We have made data available in the Appendix.No competing interests to declare.No conflict of interests to declare.