The ‘good story’ and kindness Twitter: Tales of hope and fears of dupery during Covid-19

Abstract

There is a long history of investment in solidaristic stories in the face of social upheaval, threat or conflict, and this was especially evident in relation to Covid-19. This article examines the way that one such narrative – the idea of kindness – was drawn on by Twitter users during the pandemic. Setting it in the context of a wider cultural preoccupation with kindness that both predates and continues beyond the pandemic, we draw on empirical research to develop the concept of the ‘good story’: a compelling collective narrative about goodness that carries with it anxieties about authenticity and dupery. We argue that ambivalences are inherent in the good story but are especially prominent in an age of social upheaval, polarisation, digital interaction and uncertainty. Such ambivalences cannot be wished away but need to be worked with, theoretically and practically. The good story is one tool for beginning to do this.

Introduction

Drawing on a dataset of ‘kindness tweets’, compiled as part of a wider project on stories and storytelling, ¹ this article explores – in the context of a particular online space – how narratives relating to kindness came to occupy a central place during the pandemic and the form that these took. Our primary aim is to show how these illustrate a specific kind of collective narrative – what we call the ‘good story’. We argue these stories about ‘goodness’ are characterised by an underlying ambivalence, provoking hope and anxiety, specifically about authenticity, and that they assume particular resonance in the digital age and at times of heightened uncertainty.

The sharing of the good story on Twitter ² can be seen as part of a longer history of ‘prosocial’ vocabularies being introduced to publics through storytelling. In previous centuries, for example, the theatre was understood as a key space where stories about goodness (such as empathy) were tested out – hence Rosenthal’s (2016) coining of the term ‘theatre of feeling’. Social media can, we suggest, be seen as a new theatre of feeling, one where kindness stories have proliferated.

To develop these arguments about the good story, we explain what we mean by kindness and give some context to how sociology has engaged (or not) with this concept; we consider the role of stories and storytelling and, in particular, the relationship between kindness stories and our concept of the good story, highlighting the significance of authenticity to both; we explore how these stories translate online, and outline how we set about researching Twitter; and, finally, we present data from ‘kindness Twitter’ in the context of Covid-19 to illustrate the good story in action. In doing so, we locate these data within a pre-existing contemporary cultural preoccupation with kindness and, at the same time, highlight how Twitter shaped the nature of the stories told. We conclude by showing how, and why, kindness has been productive in developing the notion of the good story and make the case for the wider analytical purchase of this concept.

Kinds of kindness

The preoccupation with kindness stories during the Covid-19 pandemic did not emerge out of thin air. Different versions of kindness have held sway at different times and in different cultural contexts (Pollock, 2011). In some contemporary English-speaking societies (Dutta & Elers, 2020), recent decades have clearly witnessed a renewed preoccupation with kindness. Indeed, over the last 20 years, what might be termed a ‘kindness industry’ (Brownlie, 2024) has grown up. As part of that, the idea of kindness has become increasingly institutionalised across politics and public policy, charity, commerce and popular culture.

That process of institutionalisation does not mean, however, that kindness is a singular, easily definable phenomenon. As an idea, it sits within a dense landscape of overlapping prosocial concepts and, indeed, is often used in conjunction with, or as a shorthand for, the prosocial in general (Cazenave, 2023). Indeed, most sociologists approaching kindness have tended to use it in this broader way. That is to say, much extant sociological work seeks to identify verifiable categories of kind behaviour (e.g. volunteering) and then seeks to quantify, document or explain this behaviour (Habibis et al., 2016; Hookway & Woodman, 2021). In recent years there has also been a growing sociological interest in small informal ‘kind’ acts that have the potential to counter, materially and symbolically, the brutalising tendencies of neoliberalism (Baccolini, 2017), as well as a body of work challenging what it sees as an uncritical turn towards kindness, raising important questions about who is being asked to be kind and by whom and for what reasons (Asafo, 2021; Canning, 2020).

In previous work, we have suggested that the contemporary idea of kindness has some distinctive features that may help to explain why we often reach for it in place of some of its conceptual competitors (Brownlie & Anderson, 2017). First of all, contemporary narratives of kindness tend to attach to the everyday. Unlike, for example, altruism or compassion, kindness is not generally linked to sacrifice or suffering. Instead, it is used to describe small acts of help and support of a type that occurs in the course of ordinary relationships or encounters. We also tend to see kindness above all in those moments when people appear to act in the interests of others in unforced or apparently unobligated ways. However, this focus on kindness as somehow supererogatory, or beyond duty, produces a particular vigilance around intent and a related concern about authenticity. We cannot erase this uncertainty and still hold on to a sense of what in contemporary times we understand the idea of kindness to be. Sociological research suggests we are often moved by kindness, but, at the same time, can find it discomfiting, anxiety-provoking and even shameful (Cole, 2020). Ambivalence, then, is core to the idea of kindness as it is currently conceptualised and experienced, at least in some English-speaking societies.

Our aim in this article, however, is not to make the case for defining kindness as a specific set of behaviours or values, but rather to consider how it is narrated and what such storytelling achieves (or fails to achieve) in particular settings.

Why stories matter: From kindness stories to the good story

Stories are not just about what we do: stories do things in their own right. ³ For Polletta et al. (2011), a sociological approach to storytelling involves exploring interactional, institutional and political context. Of late, storytelling as method has been inclusive of indigenous research approaches (Fraser & O’Neill, 2021) and as an ‘art form’ facilitating imagined futures (Rieger et al., 2023). Depending on the type of story, ‘truth’ matters to varying degrees and, in practice, the social impact of narratives does not always depend on their veracity. The notion that stories can be transformational (Squires, 2020) is engaged with in wider popular culture, too, where there is also an understanding that the ‘wrong’ stories can be destructive. Toland (2021), for instance, draws on the novelist Ben Okri’s observation that sick stories lead to sick societies, while Solnit (2020), writing during the pandemic, argued that many of the problems we faced were storytelling ones – a claim that is increasingly also being made in the context of climate change (Haraway, 2016; Thaler, 2022).

Stories about goodness or the broadly solidaristic have long been part of understanding this relationship between storytelling and social change. Wuthnow (1991), for instance, argued that in order to act compassionately it is as important for us to have stories to help make sense of our actions as it is to have ‘a free afternoon to do it’ (p. 45). Stories about kindness, too, are part of the genre of prosocial stories with claims increasingly made for kindness as a source of individual and collective wellbeing (Macfarlane, 2020).

Sociological approaches, however, also explore ‘narrative constraint’, highlighting the social inequalities that make some groups’ stories more convincing than others (Plummer, 2019; Polletta et al., 2011). At the same time, sociology (along with other disciplines) has raised concerns about whether stories might lead to a turning away from the social, that the emotion they produce, such as compassion, might end up being in lieu of action (James, 1890); that ‘losing oneself’ in a book, for instance, could lead to inequalities rather than solidarity (Felski, 2020, p. 25).

Polletta (2006) saw such concerns about the effectiveness of stories as part of an ambivalence about stories in general, exactly because they are ‘authentic and easily manipulated, universal and dangerously subjective, normatively powerful and politically unserious, valuably therapeutic and unhelpfully self-indulgent’ (p. 121).

We develop an argument that kindness stories are an example of a particular kind of historically-embedded ambivalent story, one we are calling the good story. Such stories are ‘good’ in three interrelated senses. First, and most obviously, they are broadly about goodness: they carry a message about humans’ capacity to act towards one another in generous, other-oriented, non-instrumental ways. Second, they are good in the sense that they are compelling or convincing: they captivate or hold us. But finally, such stories produce concern because of the possibility that they may turn out to be nothing more than a story (‘that’s a good one’) – at best, a harmless yarn, at worst, an attempt to distract or deceive us.

Concerns about authenticity are built into what such stories are but we argue that the ambivalence these stories provoke is intensified during times of instability, uncertainty and social division as this is when investment in goodness but also anxieties about its authenticity are at their peak. As Elias (1990) argued, our investment in stories at such times help us deal with the ‘chill of life’. When we feel least in control – most obviously, as in the present, in the face of global crises – we are especially likely to invest in optimistic stories such as those about kindness or those that otherwise offer hope.

In the face of existential crises or futurelessness (Tutton, 2023), stories of hope – like stories of kindness – have also become subject to renewed attention, though hope too is ambivalent, not least because it concerns the ‘not-yet’ (Anderson, 2017) and is always prone to disappointment (Bloch, 1998). Wettergren (2024) argues that it is the uncertainty and the limited agency associated with hope that makes fear into hope’s ‘companion’ emotion. Resolving the anxiety-inducing oscillation between hope and fear, she suggests, involves accepting fear or magnifying hope. In what follows, our focus is instead on how both hope and fear are held in tension in stories that emerge during times of crisis and on a particular fear that comes to the fore, especially when the stakes are high: namely fear of dupery, of things being inauthentic, of not being what they seem.

We could see these authenticity anxieties as being about the subjective nature of ‘goodness’. ‘Goodness’, a quality we assign to a relationship or behaviour, is based on our assessment of the intentions of those involved. We assume authenticity in the sense of identifying a ‘genuine’ concern for the other. However, the impossibility of knowing the intent of another, of being completely sure about ‘authenticity’, of how an act will be perceived, renders ‘goodness’ – and the ‘prosocial’ more generally – profoundly precarious, not only because the motivations of others are difficult to access but because they may change (Cotney & Banerjee, 2019).

We are suggesting that these anxieties intensify during times of social upheaval and find expression in a concern that good stories, including those about kindness, are a cloak or a disguise for something else such as virtue signalling or clicktivism (Wade, 2024), existing in lieu of action. This is more than a response to a current ‘age of creeping artifice’ (Olsen, 2023) – authenticity claims, after all, have long been positioned as a reaction to rapid socio-technological change (Thurnell-Read, 2019). As we will see later in the article, for instance, during the so-called age of sympathy in the eighteenth century there were also concerns that stories might lead to emotions becoming an end in themselves. Good stories, then, have a dual quality of hope and anxiety but their specific character is formed by the wider socio-historical relations in which they are embedded.

Moreover, as we will go on to show, anxieties about good stories are not evenly distributed and are intensified online. Kindness stories on Twitter are, then, a particularly fruitful way of exploring the instabilities of the good story and the tensions between hope and fear of dupery that lie at their heart.

Kindness Twitter – a new theatre of feeling?

In 2023, when Mark Zuckerberg launched an alternative micro-blogging platform, Threads, he identified kindness as one of its distinguishing features: ‘We are definitely focusing on kindness and making this a friendly space’ (quoted in Milmo, 2023). The implication was, of course, that Twitter was the opposite: an unkind space. This is a perception that has only intensified since Elon Musk assumed control of Twitter, now X, in 2023. But regardless of the actual experience of using Twitter, it is clear that talk about kindness is pervasive on the platform.

It is not possible to map in detail the content of kindness tweets here but the word cloud shown in Figure 1, for the four years 2018 to 2021, shows the most common ‘co-occurring’ hashtags with either the word kindness or #kindness. ⁴ Essentially, the larger the text, the more frequent the number of occurrences. It is clear that concepts linked to ideas from positive psychology (inspiration, positivity, happiness, gratitude) and the prosocial (compassion, empathy, humanity, community) feature prominently and that compassion, in particular, remains a close sibling of kindness. This is consistent with what we know about the density of the prosocial conceptual hinterland of which kindness is a part and the dominant influence of positive psychology in shaping contemporary attachment to the idea of kindness at least in English-speaking, predominantly high-income countries (Brownlie, 2024).

Figure 1.

Word cloud of kindness-related hashtags.

The dominant co-occurring hashtag in the word cloud is #bekind. Despite being attached to particular messages and campaigns (particularly around bullying), ⁵ it has no single meaning, nor can it be said to create a coherent digital public in the sense that #MeToo arguably has. Instead, it acts as a readily-available synonym for kindness itself and so can be drawn on by diverse actors with equally diverse concerns.

This offers an indicative sense of at least one aspect of kindness Twitter beyond Covid-19, but what does it mean to think of this space as a theatre of feeling? Papacharissi (2014, p. 2) argues ‘networked publics come together and/or disband around bonds of sentiment’ and this affective orchestration often happens through the sharing of the solidaristic or the broadly empathic. Indeed, sharing such stories has become a part of everyday social media practice with the relationship between ‘feel good citizenship’, moral performances of self, technologies and commercial interests now well established (Nikunen, 2019).

Like traditional theatre, Twitter is a public emotional space, though there are clearly significant differences between the two that we will get to – not least, as Rathnayake and Suthers (2018) point out, because Twitter’s ‘publicness’ involves ‘abstract audiences’ (Litt & Hargittai, 2016). ⁶

The work of Adam Smith offers an interesting lens through which to think about this relationship between spectatorship and sympathy or the broadly solidaristic. For Smith, sympathy was a practice not just performed in the everyday but on stage (Camp, 2020). Kerr et al. (2016, p. 12) suggest Smith’s The Theory of Moral Sentiments can be read as a treatise on ‘the arts of story-telling through the theatre, arts which ideally encouraged sympathetic feeling around a shared and virtuous recognition of the human condition’, hence Rosenthal’s description of Smith’s work as concerned with a ‘theatre of feeling’.

Yet, as in everyday life, Smith recognised that sympathy in the theatre was uneven, encouraged towards kings and great men rather than women and the poor, the vulnerable and ‘inferior’ (Rosenthal, 2016). Expressions of sympathy provoked by theatrical performances also have to be managed, Smith (1982) argued, lest these are read by others ‘as effeminacy and weakness’ (p. 46; I.iii.1.9). ⁷ Parisot (2016) identifies other risks of representations, particularly of the sentimental, in the eighteenth century including the provocation of misery and what Smith referred to as ‘extreme sympathy with misfortunes which we know nothing about’. Such sentiment might be attainable, Smith argued, but is ultimately ‘perfectly useless’ (Smith, 1982, pp. 140; III.3.9, cited in Parisot, 2016, p. 196).

Similar concerns about ineffectiveness, unevenness and inauthenticity can be heard in relation to stories about kindness on social media – specifically, that kindness tweets amount to no more than the commodification of emotions, or digital emodities. ⁸ Such tweets can also, as noted, be dismissed as a form of virtue signalling. But there are other ways of reading empathic digital performances. Ingraham (2020), for example, frames such content as gestures of concern that are both means and ends – their effects exhausted by their expressivity. These spread an ‘affectability’ which, interestingly, he chooses to call ‘kindness’. Such gestures, which are increasingly aesthetic, involving music and visuals, matter, Ingraham suggests, because they contribute to a background of expectations of non-instrumental ways of being together.

In the space between these framings of kindness tweets as a commodity that manipulates or as a means of expression that should be celebrated is a more ambivalent reading. Berlant (2011) saw a sentimental account of the social world as creating what they term an ‘intimate public’ rooted in fantasies of the common. They argued that crises such as the pandemic quickened the turn to ‘nostalgic ‘we’s’, but these imagined affective collectives have always existed (Berlant, 2022). Berlant did not dismiss outright such investments, recognising that for many they allow a sense of being transported somewhere else, albeit briefly. This position, however, is not easy to reconcile with their own critique of attachment to the sentimental as harmful and ultimately apolitical (Anderson et al., 2023). The irreconcilable nature of such ambivalences, especially in relation to authenticity, is core to good stories.

Itself an unstable ‘social construct with moral overtones’ (Grazian, 2019, p. 169), it is unsurprising that authenticity is open to contestation (Williams & Vannini, 2009), particularly online (Audrezet et al., 2020). Though we are constantly in thrall to the digital, in an era of fake news and bad faith, we are also deeply suspicious of what we find there (MacKenzie et al., 2021). Being online can rob us of the ability to gauge trustworthiness through the usual cues of in-person interaction. But as we will see in the analysis to follow, this is not simply a case of ‘is this person who they seem or claim to be?’: in an age of AI and automation, the question ‘is this a person at all?’ might be equally relevant

In practice, of course, anxieties about the authenticity of kindness on-and offline are intertwined. This is a point best illustrated by a story. In June 2022, an Australian TikTok influencer named Harrison Pawluk posted a 19-second video of him giving a bunch of flowers to a stranger – a woman drinking a cup of coffee in a shopping mall – which he captioned with the words, ‘I hope this made her day better’. The video was subsequently viewed more than 60 million times. Three weeks later, the woman – identified only as Maree – told journalists that she felt ‘dehumanised’ by the experience: ‘He interrupted my quiet time, filmed and uploaded a video without my consent, turning it into something it wasn’t, and I feel like he is making quite a lot of money through it.’ ⁹ This episode illustrates the commercial power of the idea of kindness – ‘kindness for clicks’ (Wade, 2024) – and its amplification through the internet; ¹⁰ but it is also revealing of the relationship between authenticity and kindness and how we feel about the inauthentic or, in Maree’s words, about things being turned into something they are not.

Pawluk’s story of his gesture was that it was driven by humane intent and potentially transformative, but Maree felt dehumanised, experiencing his actions as inauthentic. This kind of inauthentic kindness has been linked to ideas of the sentimental or an over-investment in the emotional. Phillips and Taylor (2009), for instance, sought to distinguish between true and ‘magical kindness’ – seeing the latter as sentimental and ineffectual. The authentic here is associated with significance, including as Maree hinted, the very nature of humanity, while the inauthentic is dismissed as trite.

Pawluk and Maree’s stories about kindness involve the same act interpreted very differently, and were circulated to greater and lesser degrees online. Overall, however, they perfectly illustrate the profoundly ambivalent nature of the good story, its compelling and anxiety-provoking nature. The point here is not that there is actually a hard line between the authentic and the inauthentic but that, when the issue at hand is ‘goodness’, there is a strong emotional investment in believing such a line exists. We will go on to illustrate this ambivalence through data drawn from kindness Twitter during Covid-19, but first we sketch out the project’s methodology.

Researching Twitter stories

There are well-recognised complexities in researching Twitter – including its distinctive user population; ¹¹ its restricted linguistic form; the absence of (reliable) information about individual user characteristics; and, as noted, the absence of direct accounts of intentions and of how content is understood. In other words, when we research Twitter, it is difficult to be confident about what it is we have a view of.

Bearing this complexity in mind – and building on previous mixed-method approaches to the analysis of Twitter data (Karamshuk et al., 2017) – we used the Twitter API to create an initial dataset of some 13 million tweets across four calendar years (2018 to 2021) that included either the word kindness or a related hashtag (e.g. #bekind). ¹² While there are disadvantages in using hashtags in this way, not least that they are only present in a minority of all tweets, we turned to these as a form of user-generated tagging that enables cross-referencing of content by topic and at scale.

Alongside quantitative visualisation of the distribution of tweets over time and large-scale quantitative temporal analysis of tweets relating to kindness (Wilson, 2020), we employed qualitative approaches: a thematic, abductive manual coding of a subsample of tweets, drawn at random from across the four years of the dataset, as well as thematic – and temporal – analysis of a further subsample of tweets concerning Covid-19 and kindness, generated by running a list of Covid-related hashtags against the full kindness dataset. ¹³

The analysis for this article involves both the full kindness Twitter dataset and the Covid-19 kindness subsample. After a process of random sampling, the Covid-19 dataset was reduced from 50,000 to 3,000 tweets and coded to saturation – i.e. to the point when no major additional issues were identified. To ensure inter-coder reliability, these were read and coded by two different members of the research team. As a precaution, following Markham (2012), we paraphrased tweets from individual Twitter users rather than quoting directly, in recognition of the fact that individuals may tweet without imagining subsequent re-use. With this in mind, the only information attached to such tweets below is their approximate date. Tweets from organisational accounts are, however, reproduced in full.

Throughout the analysis we moved between findings emerging from the computational and qualitative approaches. This proved particularly helpful in making sense of temporals shifts in talk about kindness, for instance, how reaction to Covid-19 interacted with annual fluctuations in expressions of kindness on the platform shaped, for instance, by events such as World Kindness Day. ¹⁴

The size of the overall Twitter dataset allows for further computational analysis but, in the absence of previous research on kindness Twitter that could be used as a benchmark, our aim here is to provide an overview of the content of kindness Twitter during the pandemic to illuminate further what we mean by the ‘good story’, while being cognisant of how the platform itself shaped this storytelling.

Viral kindness: The good story of kindness during Covid-19

The vocabulary of kindness, as we have seen, was already part of Twitter prior to the pandemic. There were, for instance, notable spikes in the volume of kindness tweets towards the end of each of the four calendar years. From what we know about the patterning of Google searches over time, it is likely that these were linked to the timing of World Kindness Day and the broader institutionalisation of kindness through other events and organisations. But Covid-19 changed kindness Twitter in specific ways. Most obviously, in terms of volume: there was a much larger volume of tweets about kindness in general in 2020 and 2021 than in the two previous years. ¹⁵ In terms of specifically Covid-related tweets about kindness, there was a pronounced spike around the beginning of the pandemic, especially in the period between February and May 2020, when anxiety about Covid-19 was at its peak. Thereafter, the volume of Covid-19 kindness tweets stabilised at a lower level for the rest of that year before tailing off further during 2021 (see Figure 2).

Figure 2.

Covid-related kindness tweets 2020–2021.

From this data visualisation, we see the compelling character of the good story – the way in which such narratives engage and travel. It points to a wave, or a protective wall, of kindness tweets and, as we will go on to show, qualitative analysis of tweets during this period supports the argument that the idea of kindness was being powerfully invested in as a protection against the (physical and social) dangers posed by the virus. Other stories of kindness also emerged, however – of kindness as, at best, ineffectual and, at worst, a form of dupery or a cover for the absence of humanity.

The first tweets referencing kindness in the context of Covid-19 were in late January 2020 and tended to involve expressions of international solidarity and concern. For example, there were tweets from Filipino users thanking ‘the Chinese people’ for providing free masks; stories about a Jewish-Israeli man donating masks to China (to ‘repay the kindness’ shown to Jewish refugees in Shanghai during World War II); and from Canadians acknowledging the kindness of Cambodia in allowing a cruise ship to dock, despite an outbreak of coronavirus on board. These tweets were part of an early trend that saw kindness positioned as part of an expression of shared humanity and global solidarity in the face of an apparently universal threat. ‘We are all human – let’s treat each other with kindness’ was typical of such sentiment. Unsurprisingly given the fears of what lay ahead, this positioning was key to the cultural power or resonance (McDonnell, 2023) of kindness in the early part of 2020.

Even in the face of COVID, we need to hold onto our humanity – hope, compassion and kindness even in these dark moments. (March 2020)

The embodiment of humanity is kindness. (April 2020)

Be Human and Be Kind! (July 2020)

This focus on kindness, hope and solidarity in the face of crisis is familiar from work on disasters (Solnit, 2009) and is consistent with the ‘goodness’ element of the ‘good’ story. The referencing of our better selves is partly an injunction, but also offers grounds for hope in the face of uncertainty and fear.

Around the globe, there’s a wealth of kindness, compassion and solidarity. People support one another. Amid chaos, darkness and uncertainty, our fundamental human qualities can rescue all of us. (March 2020)

The early weeks of the pandemic also saw the emergence of a specific narrative of kindness as mirroring and potentially countering the spread of Covid-19 through small acts of help and support. So, for example, Twitter users called for an epidemic of kindness, referred to its viral quality, or represented it as inoculating, an ‘antivirus’. In short, the idea of kindness assumed an antonymic, curative quality, and a claimed importance inversely related to the scale of the acts being encouraged. All of this was evident in Twitter discourse almost before the direct consequences of Covid-19 began to be experienced on a widespread scale. In other words, appeals to the idea of kindness were part of how, individually and collectively, we braced for the impact of the pandemic and tried to ward off danger. These tweets, in other words, are what culminated in the kindness ‘peak’ illustrated in Figure 2.

As the pandemic spread, and in particular as the first lockdowns began to bite in English-speaking countries, Twitter became a place to record, publicise, celebrate or offer thanks for actual acts of kindness, help and support – it became, in other words, a site of ‘care gratitude’ (Wood & Skeggs, 2020). Some of these tweets described users’ own experiences; others related to ‘good news’ stories in the media. Such accounts largely related to the kindness of others and there was little sense of individuals using the platform to self-aggrandise or document their own acts of good citizenship. Overall, what was striking was the way in which the pandemic rendered these accounts ‘Twitter-worthy’. In other words, the significance of small acts and gestures was hugely amplified by the challenges of Covid-19.

This woman helped an elderly couple get food when they were too scared to go into a crowded grocery store during the #coronavirus outbreak proving that even in dark times, kindness prevails. (March 2020)

In my home, everyone’s in quarantine due to #COVID. A dear friend left a care package at our door today, sharing the kindness received during her own illness. It goes a long way – feel so grateful. (November 2021)

Within a few weeks of the pandemic being declared (on 11 March 2020), the language of kindness also began to be attached to the work of those on ‘the frontline’ in the UK and other English-speaking contexts – not only medical staff but, for example, those providing care, driving trucks or working on checkouts.

Very appreciative to all the hard work and generosity from @harvestforheroes delivering fresh food to all the fabulous nursing staff [. . .] #HeroesOfCovid19 #kindness #NHSVolunteerResponders (April 2020)

While this framing could be seen as a celebration of the ordinary – people continuing to do their jobs, albeit in extreme circumstances – the pandemic gave such endeavours a heroic quality (McCormick, 2020) which meant that kindness itself was also elevated beyond the everyday.

As the pandemic wore on, however, this celebratory tone was increasingly accompanied, and sometimes diluted or suppressed, by a more regulatory one. Proving itself to be a remarkable labile concept, the idea of kindness came to be attached – especially by organisational actors – to a range of new objects to justify key health protection measures, including social distancing, isolation and mask wearing during the early stages of the pandemic, followed by testing and, latterly, being vaccinated. In other words, the appeal to kindness became part of a responsibilisation strategy which effectively framed these behaviours as a gesture of concern for the wellbeing of others.

If you’ve yet to take a lateral flow test this week, make a point to do one this evening.[. . .]. #lateralflow #Covidtest #kindness (July 2021)

Universal mask use can significantly reduce the transmission of #Covid19 by preventing anyone, including those who are ‘carriers’ from transmitting it to others. Lets kill #Covid with kindness (September 2021)

We are all equally equipped to be kind to each other. One way to show kindness is to get your #Covid_19 vaccination today. (October 2021)

Such regulatory appeals to kindness were, however, not limited to specific health protection measures. Echoing associations between the idea of kindness and courtesy that were common during the early modern period (Pollock, 2011), from summer 2020 onwards, the idea of kindness was also being regularly deployed in calls for greater civility – particularly in how members of the public interacted with service providers and, indeed, with each other.

Patience, kindness and respect. . .You expect it so please give it to others. Thank you to the @NorHillsHosp #Covid19 Assessment Centre staff who continue to be at the Frontline, providing information, booking appointments and helping keep the community healthy. (December 2021)

This more explicitly regulatory version of kindness was also evident in increasing numbers of tweets ‘calling out’ others for behaviours that were seen to lack kindness or be actively unkind.

I have friends in the beauty industry feeling super anxious and stressed at the moment because of the demands people are making of them. We must prioritize kindness during this Level 2 period. (November 2021)

As the pandemic wore on, some Twitter users conveyed an air of disillusionment, highlighting what they saw as increasing selfishness or intolerance (‘how quickly the milk of human kindness goes off’, as one put it). Reflecting on evidence of indifference, injustice and unkindness, some came to see their earlier belief in the protective effects of kindness as naïve.

Recently revisited an essay I penned early in the pandemic. ‘Across the globe, even here, people are looking out for each other. How generous and kind our fellow citizens can be.’ Looking back, those were indeed innocent times. #Covid19 (December 2021)

The questioning of the nature of humanity (and relatedly of kindness as a true expression of it) intensified, with increasing polarisation around issues such as mask mandates and vaccination. In this context, questions arose about whether people were capable of kindness or whether kindness could form the basis of any kind of effective response to Covid-19. Persistent inequalities and divisions meant that, for some, being asked to do things in the name of kindness was now seen as hypocritical and inauthentic. In other words, we see here the third and final element of the good story – the suspicion that all is not as it seems and that the idea of kindness is being used to deceive or distract. It is the authenticity of calls for kindness, as much as the authenticity of specific kind acts themselves, that is often being questioned.

Medical professionals are realizing some people are idiots. They’re no longer wanting to risk their own lives to save people who contracted Delta due to their stupid choice not to get vaccinated. Good! I’m done with the whole kindness thing. (June 2021)

Antivaxxers who ask for kindness are hypocrites, especially when they show disrespect to healthcare workers and people who’ve lost loved ones to COVID. (September 2021)

By 2021, particular ire was being directed against politicians and others who deployed stories about kindness but whose own actions were felt to lack it, or were seen to be using it as cover for a lack of concrete, just and effective public health measures.

Premier Kenney calling for kindness and unity? Fuck that. You were missing in action while Albertans suffered, then made us all suffer the consequences. You don’t deserve kindness or unity. You deserve the sack. (September 2021)

Please stop talking about kindness. If you’re a politician, medical expert or government official enacting or defending plans that have resulted in thousands of people getting unnecessarily sick and dying, you have no right to talk about it. Amen. (September 2021)

This reflects anxieties about political investment in the notion of kindness that existed in some countries before the pandemic. The ‘politics of kindness’ promoted by New Zealand’s then Prime Minister, Jacinda Ardern, for instance, was seen by some as a form of ‘political capital’ which ‘camouflaged control’. In particular, during the pandemic kindness discourses became closely linked to nationalist framings, some of which were also read as sentimental and inauthentic (Craig, 2021).

While much of this criticism was from those who felt that a ‘kind’ approach would have involved more active or effective management of the pandemic, similar contrast rhetoric (with its implicit sense of the inauthenticity of the claims being made) was also used by those who were critical of lockdowns and the vaccine and who felt that individual liberties were being infringed in unkind ways. In both cases, there was suspicion that the language – and emotional power – of kindness was being used to disguise or mask something else. Wilkinson-Ryan (2023) has explored the power differentials involved in promoting fears of being duped or ‘suckered’. She writes that telling stories about others’ planned duplicity – for instance, through bogus claims of vulnerability – is one way of subordinating the weak, playing to the ‘status fears’ of those who have power. To let those with less power dupe us is more humiliating than to be duped by those in power: the former, she suggests, exposes us as ‘fools’, while to be hoodwinked by the powerful cannot be helped. Policing authenticity in this way, then, heightens anxiety about dupery by the weak even though, as the analysis here suggests, it is in fact the adoption of the language of kindness by the powerful that might have more severe consequences.

Within the scope of this article, it is not possible to explore systematically responses to kindness tweets, but it is worth noting that kindness tweets typically generate little reaction from other Twitter users – often barely registering in terms of likes or retweets. The fact that kindness tweets are often formulaic (re-using existing content), are highly numerous but exhibit little evidence of direct engagement raises the question of whether they are, in fact, produced by bots, and involve automatically generated content. In order to estimate the prevalence of automated content relating to kindness, we used ‘Botometer’ – a machine learning algorithm which checks the activity of Twitter accounts and gives them a score based on how likely they are to be bots (Yang et al., 2022). Though this approach is far from an exact science, the results suggested that automated content is a persistent but not especially common or disproportionate feature of kindness Twitter. A conservative estimate would be that only around 1 in 20 of all accounts tweeting about kindness strongly exhibited ‘bot-like’ behaviour.

In other words, most of this material still seems to involve authentically human users. Nevertheless, the fact that it remains unengaged with may tell us something important about the difficulty of distinguishing human- from non-human-generated content (in other words, much of this content may look like and be responded to as if it were automated, and hence appear inauthentic, even if it is not). Even if not engaged with directly, however, such content, as Ingraham (2020) suggests, may still contribute to an affective atmosphere on Twitter. If so, then this is a different way of telling and listening to stories about the good than that found in traditional theatre.

At story end

Like other forms of ‘goodness’, kindness is subjective, but implicit in contemporary versions of kindness is the notion that such behaviour is unobligated. This produces a particular vigilance around people’s intent and, hence, authenticity; and because – as we have seen – kindness is often framed as the epitome of what it means to be human, the stakes are experienced as especially high. In other words, the anxieties that attach to stories about kindness are proportionate to the extent to which we invest in it. For obvious reasons, such investment was very evident during the pandemic. Overall, then, to engage with the idea of kindness, means – as with other good stories – to experience uncertainty. For this reason, kindness comes in many ways pre-wired for the digital age and its own perennial contestations over authenticity and what can be known for sure.

Theoretical work on storytelling and social change often looks to either the potential or the limits of stories, but the concept of the good story allows us to sit with the ambivalence of stories about the ‘prosocial’ rather than dismiss them outright as the manipulations of governments or markets or overinvest in them as a panacea for complex social problems. By looking at kindness stories told on Twitter during the pandemic we showed how, at different times and/or for different groups, kindness was positioned as an authentic expression of ‘the best of us’ and as inauthentic, a form of manipulation. Although the analysis had a temporal element, we are not suggesting that one kind of story replaced the other. It is likely that fear of potential dupery was always there but that at least in the initial crisis phase of the pandemic this was less obvious, particularly in the shadow of the wall of kindness tweets that emerged in March 2020. Moreover, while any one individual story/tweet, such as the ones we engaged with above, may not explicitly exhibit ambivalence we can still understand the good story as having collective and ambivalent dimensions and effects.

This analysis offers an alternative to a particular narrative arc that has characterised talk about the experience of the pandemic, at least in the UK: that of hope followed by disillusionment, ¹⁶ of investment in the idea of shared humanity being punctured by the reality of persistent inequalities. The notion of the good story helps us move beyond a disillusionment narrative to a dual focus on hope and anxiety.

Anxieties about authenticity took different forms. Sometimes they were focused on hypocrisy – those in power, most notably politicians, seen as deliberately using kindness as a cloak for acting unjustly, for instance. At other times, concerns centred on the ineffectiveness of kindness and of stories about it – either because such stories were seen as a disguise for something else or because they became an end in themselves. In other words, rather than leading to action, they existed in lieu of it.

In relation to kindness stories, if Twitter is a new theatre of feeling its ‘newness’ is not found in the nature of the stories shared there – much of what has been expressed on Twitter is familiar from the offline. In particular we see the same investment in kindness as a reflection of our humanity, fears about inequities of kindness, the impossibility of distinguishing authentic from inauthentic kindness and the ways in which sentimental expressions of kindness are not necessarily the opposite of ‘true kindness’, but rather potentially part of what makes the latter possible – in whatever form that might take. We see, too, continuities in anxieties about the limits of such performative spaces – not least that they are only open to some and that what happens there, however much emotion it provokes, might not effect change beyond it. The theatrical space of Twitter is new, however, insofar as it is shaped by the immediacy of response to events and the scale and speed of circulation, all of which changes how stories catch on – witnessed, most obviously, in the wave of kindness tweets that rose up in the very first months of the pandemic.

But why does it matter if the stories we tell are ultimately ones of hope or anxiety or good stories as we define the term, that is, ambivalent ones of hope and fear of dupery? The answer lies in the fact that stories are key to how ideas and feelings travel, but their role is finely balanced – there is no social change without stories but equally social change cannot only involve narration (Plummer, 2019). Although evidencing these effects has not been the focus of this article, stories matter because they do work, including encouraging (or dampening) the doing of kindness. Leaning into the ambivalences of good stories might not seem like a particularly rousing rallying call, certainly compared to more straightforward uplifting stories of hope but there are, nevertheless, good reasons for it. Garfinkel observed that to try and rid ourselves of ambivalence is to ignore that which needs explanation: ‘like complaining if the walls of a building were only gotten out of the way one could see better what was keeping the roof up’ (Garfinkel, 1967, p. 22). Acknowledging the ambivalence inherent in the good story avoids turning authenticity (both in relation to stories and ‘real life’) into an article of faith, a step that, as Hall (2013) warned, can lead to authoritarianism and to losing the masks (in the Goffmanesque sense of fronts) needed for social spaces to be tolerable. These are risks that have increased in line with the intensification of fears of dupery as a consequence of political polarisation.

Stories of kindness in the Twitter data, as in much recent sociological theory, have tended to be focused on the interpersonal and the mundane and on its potential world-making qualities. On an everyday level, what might thinking about the idea of kindness through the lens of the good story mean for those who wish to work with or invoke the idea of kindness within their communities, workplaces or politics? Above all, it signals the need to carry an awareness of the inherent ambivalence of such stories into how we narrate and practise kindness; to understand that the more full-throated and less nuanced the call for kindness, the greater the likelihood that it will produce a backlash of criticism and concern about authenticity and dupery. This was evident during the pandemic but also now in other contexts where people are struggling with the challenge of being kind in often brutalised conditions in health, social care and education (Burton, 2021).

In short, presenting the idea of kindness as the ‘one thing’ that really matters (as in the ubiquitous slogan, ‘if you can be one thing, be kind’) or as the essence of what it means to be human is asking a lot of it – not least because we often struggle to be sure of what we are dealing with in everyday ‘kind’ interactions. Equally, however, the stories we tell about such interactions can play a part in creating an affective atmosphere which is conducive to small-scale acts and, as such, to broader solidaristic narratives and consequences. For kindness to be a sustainable idea, then, we need to navigate carefully that middle ground, aware of the risk of tipping into pollyannish optimism on one side and paralysis through fear of dupery on the other. This is what James (1897) meant when he argued that the will to believe needs to involve accepting that ‘dupery through hope’ is no worse than fear of such dupery, given the paralysing effects of the latter.

This analysis of tweets about kindness during Covid-19 illustrates the fear that good stories become an end in themselves but also the hope that they facilitate our acting as if we really are all in it together. For Elias, one of the fundamental questions of human existence was whether we wanted to be protected, through fantasies, from cold realities ‘while running the risk that unwanted reality might suddenly break into those comforting dreams, leaving us to live on embittered, disappointed and cynical in the face of our lost dreams?’ (Elias, 2010, p. 79). Through introducing the notion of the good story we are encouraging an awareness of the implications, theoretically and practically, of viewing stories about the good, from the outset, as unavoidably both of these things: a protection and a lost dream.