The frequency of different frames varied in the sample newspapers. The proportion of articles framing migrants and refugees as villains and a threat was highest in The Telegraph (59.32%, n=175/295), while The Times of India (49.36%, n= 317/642) and The Pioneer (50.56%, n=88/174) used the victim frame more frequently than the threat frame. Figure 1 shows this variation across the three newspapers.

The difference in framing is distinct between the sample newspapers. For example, when a legislator called for Rohingya refugees to be shot and “eliminated” if they didn’t leave India, the differences in coverage by The Pioneer and The Times of India were apparent, as evident in the opening paragraphs of the stories:

‘Bangladeshi infiltrators and Rohingya refugees should be shot dead if they refuse to leave the country honourably,’ said a BJP MLA here (36) (The Pioneer, July 31, 2018). The story had no quote countering him.

‘Rohingya Muslims and Bangladeshis living illegally in India should be shot dead if they do not leave the country,’ said BJP MLA from Goshamahal Assembly constituency Raja Singh. He is known for his hardcore Hindutva remarks (37) (The Times of India, July 31, 2018).

The article went on to present voices from human rights organisations and other political outfits criticising the leader for his inflammatory remarks. While the former article framed migrants as a “threat” with the legislator being the only source of information, the latter takes a more humanitarian view of the subject, portraying refugees as victims of a hostile system.

This section presents how the three frames were utilised in coverage of migrants.

Articles that framed migrants as villain portrayed them as being dangerous, violent, criminals, and as a threat to public order and safety of native people. They are also shown as competitors for scarce jobs and state funds, and the reason for increasing unemployment and insecurity in the destination states. Within this frame, two recurrent narratives were identified:

Vilification of migrants was distinct in The Pioneer which, although it had fewer articles, frequently used inflammatory language, especially while alluding to Muslim migrants from Bangladesh and Rohingyas from Myanmar.

According to sources, these (Bangladeshi) immigrants are notorious in nature. We need to take steps to ensure our people are safe. (38) (The Pioneer, November 28, 2018).

According to reports, about 11 lakh Rohingyas have already reached Bangladesh and may try to infiltrate into India. The Government of India has taken the righteous decision to not bow down to international pressure. (39) (The Pioneer, May 1, 2018)

In both editorials from which the above quotes are excerpted, vague statistics, attributed to even vaguer sources, are used to paint a picture of a crisis, while words like “infiltrate” and “notorious” evoke fear. All three papers frequently used water-related metaphors like “inflow”, “spurt” and “flushing out” while referring to migrants to create a sense of urgency and notion of scale, especially in the context of immigration from Bangladesh.

Frequent mention of migrants in crime reportage was observed in all the newspapers, along with a general tendency to attribute such crime to migrants, even when evidence was lacking. In some instances, migrants were vilified even after their innocence was proved. This was evident in the reporting of a case of child rape in Uttarakhand, where a migrant labourer was suspected by locals of perpetrating the crime. Covering police investigation that found otherwise, The Pioneer concluded:

Though a local may be the accused in this crime, there are various other crimes including murder which have been committed by those who are not natives of the state. (40) (The Pioneer, August 25, 2018)

Similar framing of migrants as potential criminals was found in other newspapers that frequently quoted unnamed sources (eg police officers):

Thefts and dacoity in Panambur, Mulki and Surathkal have been on the rise in the recent past. Police say these areas see a lot of migrant labourers and suspect their hand in the crime. (41) (The Times of India, October 28, 2017)

Associations with crime were also made indirectly through descriptions of neighbourhoods that have high migrant, refugee population as being overcrowded, with poor sanitation and high crime rates.

A large migrant ‘Poorvanchali’ population lives in these rickety buildings and makeshift structures in congested alleyways. The area is infested with criminals and gang wars are frequent. (42) (The Telegraph, July 2, 2018)

Numerous industries and a large migrant population offer the right environment for anti-social elements in Coimbatore and neighbouring Tirupur to camouflage themselves. (43) (The Times of India, July 8, 2017)

Although less common, the victim-framing portrayed migrants as victims of inequality and/or discrimination. Within this frame, a little more than half the articles were on low-income internal migrants. These presented migrants as a homogenous group, or clubbed them together with other marginalised communities, frequently referred to as “rural landless poor”, or the “urban poor”, or “poor settlers”.

Cities in urban India are flooded with the poor. Over the years, due to high migration, poverty in the country is transiting from rural to urban largely. These groups, who live in clusters in big metros, are often victims of harassment and crime. (44) (The Times of India, December 8, 2017)

All the articles that framed select refugees and cross-border migrants as victims were accounts of their struggle for citizenship, documentation and their constant fear of deportation.

‘Where will we go? We can’t go to Myanmar. We’ll all be dead,’ said Ashadullah, a Rohingya, who moved to the shelter (in Chennai) from a camp in Jammu and Kashmir. (45) (The Times of India, December 13, 2018)

Migrants and refugees represented as contributing to society, and having agency – to live, work, succeed and question – featured the least among other frames in all three papers. A quarter of the articles that used this frame were reports and editorials from Gujarat following violence against migrant workers in October 2018. When locals attacked migrant workers from the Hindi-speaking belt over allegations that a migrant worker was involved in raping a child, local businesses and the media rallied behind the workers. They maintained the state would be crippled economically if the workers left and appealed for peace.

Migrant workers are indispensable to the Indian economy. The mobility of labour pushes the wheels of the economy forward. (46) (The Telegraph, October 9, 2018)

A majority of the reports presented migrants as passive. This is evident from fewer voices being recorded from these communities: 74.5% (n=757/1,854) of the quotes in these sample articles were attributed to the police, government officials and unnamed sources. 99.04% of the articles that quoted migrants and refugees (n=310/313) were those that framed them as victims (eg: victims of crime and discrimination, lack of access to essential services) and heroes (eg: contribution by migrant entrepreneurs and workers to the host economy). Only three articles that framed migrants as a threat included voices of migrants. Figure 2 shows the sources of information in articles on migrants and refugees.

On examining the categories of migrants and the different aspects of their social positions (class, ethnicity, religion etc) that newspapers reported on, we found across the three newspapers, the highest media coverage was for internal (inter-state and intra-state) migrants and internally displaced persons (46.80%, n=520/1,111), mostly low-income migrants from the Hindi-speaking belt, West Bengal and Assam; followed by Bangladeshi immigrants (20.16%, n=224/1,111) and Rohingyas (14.67%, n=163/1,111).

Significant variation was found in application of frames to migrant groups based on their social position. From the data collected, religious identity and ethnicity and nationality appeared as the most prominent markers determining whether they were viewed as heroes, victims or threats. Figure 3 provides a breakdown of the three frames based on type of migration, migrant ethnicity/nationality.

The Muslim majority Rohingyas and Bangladeshi immigrants were portrayed more frequently as a threat compared to other migrants and refugees (for example, Sri Lankan and Tibetan refugees), by all three newspapers. Most of the articles on cross-border migrants (from Pakistan, Bangladesh and Myanmar) made a clear distinction between those who identify themselves as Hindus and Muslims. The former, if they were identified by their religion, were framed only as victims of hostility in source states and heroes for the extent of assimilation in the destination states. No article vilified them. This extends to the media’s portrayal of internal migrants too. For example, Kashmiri Pandits, an upper-caste Hindu Brahmin community who fled the insurgency in Muslim-dominated Kashmir Valley in the 1990s, were represented only as victims of the “Muslim insurgency” (47) and heroes for their contribution to the economy in their destination states/cities (mostly in the national capital region).

This excerpt of an editorial in The Pioneer (25 September, 2017) shows how the intersection of religion and migration influences framing:

The festering Rohingya crisis has caught international attention with Rohingyas having earned the epithet of being ‘one of the most persecuted minorities in the world’ by the UN, even as the body never bothered to show scant sympathy for the persecuted Hindus in Bangladesh (39).

The editorial argues that Bangladeshi Hindus should be granted refugee status or citizenship in India, while filtering out Muslims, as “more than 90% of the crimes in border states are committed by Muslim migrants from Bangladesh”. No source for this statistic was provided.

It is to be noted that reference to “illegal Bangladeshi immigrant” can take different meanings based on the source. When representatives of the ruling right-wing party refer to Bangladeshi immigrants, their primary targets are Muslims, but to political outfits and unions in Assam (which is divided more on ethnic lines), Bangladeshi immigrants could be Hindus and Muslims from either Bangladesh or Bengalis from the neighbouring state of West Bengal (48). A majority of the articles on internal migrants (framed them as victims (n=296/520) of crime, with poor working and living conditions, and lacking access to basic facilities); 33.26% (n=173/520) portrayed them as a threat, and 9.8% (n=51/520) of the articles framed them as heroes highlighting their entrepreneurship and contribution to the economy. 9% (n=100/1,111) of the articles did not provide details of the type of migrants being referred to. More than half (n=52/100) of these articles used the term “migrants” and “immigrants” more to raise alarm than to inform, while 42.2% (n=38/90) portrayed them as victims.

We also noted the absence of reportage of certain migrants: skilled internal migrants and migrants from north-east India (who are vulnerable to discrimination in other parts of India because of their ethnicity) and migrants from Nepal.

Of the 1,111 articles on migrants and refugees in India in the sample newspapers, 8.01% (n=89/1,111) linked migration with health. Out of these, only 20.2% (n=18/89) focused primarily on migrants and migration. In all the other articles, migrants were clubbed with other marginalised communities like the urban poor, and groups categorised as high-risk for communicable diseases like HIV/AIDS and TB (truck drivers, men who have sex with men, female sex workers, etc).

Notably, 42.69% (n=38/89) of the health focused articles reported on the health status or issues affecting migrants (and other marginalised groups) and the wider determinants of their poor health; whereas the majority (57.3%, n=51/89) reported on the impact of migration on the health of the local population.

We describe below specific issues and topics covered under each of the two categories: migrants’ health, and the impact of migration on population health. In all three newspapers, there was no distinction between different migrant groups in articles on health. They were collectively referred to as “migrants” or the “migrant population”; while no specific reference was made to the health of refugees.

A majority of the 38 articles (84.2%, n= 32/38) in this category portrayed migrants as in need of state assistance (victim/humanitarian frame), though often blamed them for their status. These articles were short, less than 200 words, and mostly announcements of state programmes.

Preventive healthcare was the most widely covered theme under migrants’ health (Figure 4), but mostly included vaccination drives for children, and awareness programmes for HIV/AIDS and TB among migrants. There was no reporting on the lack of basic amenities such as housing and drinking water, or the factors underlying migrants’ limited uptake of aanganwadi and other facilities.

Access to healthcare, the second most widely reported issue under migrants’ health, focused on healthcare provision through government-led camps; only two of these reported on the barriers faced by migrants in accessing healthcare.

Six of the eight articles on sanitation blamed migrants for their squalid living and work conditions and cited this as reason for their poor health outcomes, especially in relation to infectious diseases. Two articles presented migrants in the victim and humanitarian frame, highlighting their vulnerability to diseases – exacerbated by their migrant status – and delved into the lack of sanitary facilities and provision of healthcare for them by the state. This difference in framing is discernable in this reportage of open-defecation by two of the sample newspapers:

The headline of an article in The Telegraph on civic officials in Jamshedpur penalising those defecating in the open read: “Poop cuffs fails to raise stink” (49). The report went on to say: Thousands of urban slum residents, migrant labourers and the homeless – many being women and children sully these spaces.

A subsequent TOI report (50) framed the issue differently: “Toilets few, threat of fines fails to stop open defecation,” read the headline, pointing to the lack of facilities for these communities. The article by The Telegraph, on the other hand, had placed the blame squarely on migrants with no mention of the state’s role in providing facilities.

Articles that reported on population health implications of migration tended to be longer, between 300-570 words, and mostly (92.15 %; n= 47/51) represented migrants as a threat to population health.

Articles focused on infectious diseases (86.27%, n=44/51) and sanitation (13.72%, n=7/51). Increased prevalence of communicable diseases like HIV and TB in destination states was attributed to a rise in migration in 38.29% (n=18/47) of these articles. In 46.8 % (n=22/47) of the articles, migrants are framed as a threat by being either carriers or potential carriers of diseases; being blamed for poor sanitation in cities and the source of outbreaks (14.89% n= 7/47). Only four articles identified migrants as being victims of a disease/outbreak along with other members of the local community.

Frame analysis of articles on migration in the sample newspapers revealed three main frames, applied to a range of issues (crime, health, economy, poverty alleviation, security and border control, urban development, among others) and involving different migrant and refugee groups. Overall, migrants and refugees were most frequently framed as a threat, while the second most recurrent frame presented them as victims. The hero frame was used the least. Applying these frames to health, the study found that the threat/villain frame was most frequently applied, pathologising migrants and projecting them as carriers of infections and diseases; a view that appears to have intensified in the Covid-19 crisis (12).

We discuss our findings under three main issues with regard to media representations and their implications for policy prioritisation of migrants’ health and well-being.

First, the media homogenises migrants’ experiences and erases their unique contexts of structural vulnerabilities and agencies. While the sample articles make a distinction between different migrant groups in their reportage of crime and national security, while covering health, we observed that not only do the media fail to make this distinction, but they also ignore the distinct problems migrants face compared to other marginalised groups. Migrants’ lived realities and needs are invisibilised in the media narratives by failure to report on their health, livelihoods and other needs and lived realities, as well as relying primarily on the views and accounts of others (eg State representatives). Only 18 (20%) of the 89 health-focused articles were primarily about migrants, while the rest clubbed them with other marginalised groups. McKay et al (51) argue that migrants’ health issues and their context are distinct from those of other marginalised communities and the effect of migration itself on their health is complex and varies across migrant groups. Studies have shown that migrant workers in India face multiple challenges like occupational hazards because of unsafe work conditions (52), higher prevalence of chronic diseases compared to those with no history of migration (53) and insufficient nutritional intake (54), subjects that continue to be ignored by the media. The media’s narrow focus on infectious diseases outbreaks, preventive care and sanitation issues while covering migration and health could also be because almost all the stories linked to health were filed from the destination states/cities, and newspapers tend to give salience to those aspects of stories that were deemed of interest to their readers (55). We also noted the omission of gender and an absence of women migrants in migration narratives. Despite women forming a major part of the migrant workforce in India (5) and studies showing significant gaps in migrant women’s access to healthcare (56), no article specifically addresses the health and experiences of women migrants.

Second, despite the homogenous representation of migrants in the newspapers, our study, by utilising an intersectional lens, found variation in frames based on the social position and identity of migrants. Most prominent axes along which their positionality was defined were: religion, nationality, ethnicity and socio-economic status. While all the migrants in the sample newspapers were mostly projected as a threat (47.07%) or as victims (46.35%), there are indications that the frames (and the language used) varied based on how close their identities and positionalities were to those of the majority (dominant) community in the region served by the paper. The closer, the less likely they were to be framed as a threat. For example, all the articles on Kashmiri Pandits, most of which originated from NCR and Jammu or featured in editions here, framed them only as heroes for their contribution to the economy and to the arts and as victims of militancy and displacement (47, 57, 58). Most reports on internal migrants coming from Odisha, Bihar, Jharkhand, West Bengal and Assam were from southern states (Tamil Nadu and Kerala) and framed them as victims of discrimination or as a threat (crime), with their contribution to the economy constituting less than 10% of the coverage on them. Studies, on the other hand, have shown that this silent workforce contributes more than 2% of the country’s Gross Domestic Product and is critical for building cities, maintaining household economies, and keeping consumption growth in the places of their origin through remittances (12, 59). Bangladeshi immigrants, on the other hand, are almost universally framed as a security threat and described as “illegal migrants”, “infiltrators”, “Muslim terrorists” and “Bangladeshi terrorists”. The term “illegal” is often criticised by scholars, not only because it “stresses criminality”, but it is also misleading and demeaning (60). It also does not address the multiple “in-between” situations (61) as evident in the changing policy climate in India and the ambiguous definition of citizenship. Studies have shown that “illegality” leads to social and biopolitical exclusion, which often translates into adverse living and working conditions, poverty, the perpetual fear of arrest and deportation, chronic stress, and other factors that interact to heighten vulnerability to illness and injury (62, 63, 64). This raises the concept of what French anthropologist and sociologist Didier Fassin calls “biolegitimacy”, questions of morality and judgement that lead to the unequal evaluation of human suffering and human lives (63). In this study, the limited corpus on health, for example, mostly pertained to internal migrants. Refugees and undocumented immigrants are completely absent in this discourse, but are prominent in themes related to crime and security. There is clearly an unequal evaluation of human health and suffering here by the media.

Third, through frames — and the use of certain terminologies and references in constructing these frames — the media plays a crucial role in maintaining, if not widening, social gaps and inequalities. It enables “othering”, the dual process of creation of an imagined included “us” through the construction of an excluded “other” (65), by the dominant/majority group, deepening social divisions and hierarchy in the process. These divisions are, however, not naturally occurring, but socially constructed (66). The differentiation is created by assigning opposing value-laded binaries (virtuous/corrupt, normal/abnormal) to both groups. Words like “our” and “us” furthers the “othering”. The process is also used as a means to maintain inequality so power can be exerted. As Cottle argues (67), through these value-laden binaries the media audience are invited to construct a sense of “who ‘we’ are in relation to who ‘we’ are not”. In our study, this process is discernible more prominently in negative frames, where quotes and descriptions are used to reinforce stereotypes and delineate group boundaries. Most of the articles, through attribution or through comments and editorials, consistently put an emphasis on a collective “us” needing protection from “them”. The fear of the other is used to solidify the biases and prejudices that exist along class, caste, religious and ethnic divides. The division can be explicit or implicit. An example of the latter is description of neighbourhoods where poor migrants stay as opposed to where local residents stay, and frequent references to their hygiene and sanitation practices. Studies have shown that such “othering” could cause anxiety and poor mental and physical health among these communities (68), and an erosion of trust in public services (including health) resulting in lower uptake of preventive interventions such as immunisation, and avoidance of healthcare (69).

We observed that stereotypes and prejudices concerning migrants are reinforced through the use of frames as narratives, which are often built on the perceptions of the majority or dominant class (government officials, right-wing organisations, outfits, etc) and are devoid of migrant voices and those of minorities For example, none of the articles on Bangladeshi immigrants where they are framed as a threat had a voice from the community countering this negative perception. The absence of migrant voices as sources in the media can deprive the audience of a complex understanding of migration issues. It can also have negative consequences for migrants’ integration and their sense of belonging (25). By robbing them of their voice, the state and media shape our image of who a migrant is.

We also observed a lack of nuance in articles that frame migrants as victims, the second most recurrent frame in our corpus. While victimhood can be beneficial in invoking empathy and sensitivity to the precarious contexts and lives of migrants, with moving personal stories driving campaigns to end indefinite detention of migrants and asylum seekers (21), it is also limiting. The biggest drawback is it relies primarily on emotional reactions which do not necessarily contribute to a nuanced, well-informed public and political debate on migration. Nor does it contribute to good policy making (21). It also shows migrants as lacking agency, ie, the ability to act in order to advance their own and their families’ interests. The decision to migrate and the journey followed is a complex process that involves decisions and exercise of agency for migrants to leave a familiar setting in the hope of a better life. The question of migrant agency is crucial to how academics, policymakers, and activists frame and understand migration. In our study, agency is one of the variables included in the hero frame – the least utilised frame (6.6% of all articles). Extent of assimilation and contribution to the host society emerged as the cornerstones of the hero frame. For example, three articles within this frame were from Kerala where migrant workers from Hindi-speaking states were rewarded for learning Malayalam, the local language. This corroborates Spoonley and Butcher’s observation that migrants’ contributions and lives are presented as success only if framed within the dominant culture (70). To this end, the narrative of assimilation works to “domesticate” a certain type of migrant within the majority community and serves as a mechanism through which “they” become “us”. In this process, the “difference-oriented” rhetoric gets solidified, and exclusion and segregation along other ethnic, caste and religious lines legitimised by culturally homogeneous majorities.

The current hegemonic discourse in the mainstream Indian print media on migrants and refugees is selective, premised on perceptions of the dominant class (government officials, right-wing organisations, outfits, etc) and advances a poor understanding of migration and migrants’ health. Articles rely heavily on majoritarian views, stereotypes and information handed to them by state authorities. The health of migrants is given the least priority among other themes, and even within the limited corpus there is a tendency to pathologise migrants and treat them as dangerous for the health of local populations, while sparse attention is given to their entitlements, more specifically, their poor health status and rights. These portrayals fail to capture the dynamism and circular movement that characterises migration in India, and risk vilifying or victimising migrants and impeding any potential for their assimilation in the societies to which they contribute. While there has been an evident shift in the Indian media’s reportage of migrants in the ongoing Covid-19 crisis – an increased focus on individual stories, more migrant voices being documented, questions raised over government policies, and recognition of migrants’ contribution in various sectors – it is imperative to note that the pandemic has only exacerbated existing vulnerabilities. As the Indian economy reopens, migrants are now gradually returning to living precariously in cities.

Most studies that highlight the importance of the media in setting agendas and shaping public discourses draw from van Dijk’s theory on the two-way relation between power and discourse in the media (23). On the one hand, power controls these discourses, and on the other, these discourses themselves have the ability to influence power and decision making. Our study found that in the two years leading up to the pandemic, the print media’s coverage on migrants was through a narrow range of issues (crime, national security, disease outbreaks that they are deemed responsible for), that largely relied on stereotypes and information handed out by the State. Bestowed with the power to set the agenda for issues of political and social significance (71), the media, through acts of commission and omission, has contributed to amplifying the health and socio-economic crisis faced by migrants.