Abstract
This study explores the association between the untouchability experiences in households and the migration status of India’s Scheduled Caste community. Utilizing descriptive statistics and binary logistic regression models, this study analyses migration determinants that are induced by untouchability. The findings indicate that households experiencing untouchability are twice as likely to migrate (uOR = 2.03, CI = 1.71–2.42). Even after adjusting for socio-economic and regional factors in the regression model, the higher odds persist at 1.5 times higher (aOR = 1.51, CI = 1.15–1.98). Uttar Pradesh (21.94%), Rajasthan (12.61%) and Madhya Pradesh (9.81%) exhibit the highest untouchability rates. Madhya Pradesh (33.65%) records the highest migration rate due to untouchability, followed by Rajasthan (18.01%), Uttar Pradesh (9%) and Bihar (9%). This study underscores a noteworthy disparity in migration patterns between those experiencing untouchability and those who do not. Addressing untouchability is essential for mitigating its profound effects on migration, socio-economic mobility and community dynamics, promoting a more inclusive and equitable society.
Keywords
Introduction
Untouchability is the term used to describe some ‘upper caste’ traditions in Hindu society in India, such as avoiding having physical contact or sharing food and drink with those who have been designated as ‘Untouchable or Dalits’ (Borooah, 2017; Deliège, 2002; Sarukkai, 2009). The concepts of ‘Untouchable or Dalits’ have a long history in Indian Hindu society, and these concepts come from the ‘caste stratified varna system’ in the Hindu religion, which is determined by birth (Borooah, 2017; Thorat & Joshi, 2020). According to the ‘varna system’ in Hinduism, the caste is categorized into four harrows: Brahman, Kshatriya or Chhettri, Vaisya and Sudra (Sultana & Subedi, 2016). This hereditary system is explained as Brahmins served as priests and teachers, Kshatriyas or Chhettri served as warriors and rulers, Vaisyas served as moneylenders and traders and Sudras served as menial workers. However, the rest of other castes who do not fall under these four Varna systems are considered as ‘outcaste’, ‘impure’ or ‘Untouchables’; Gandhi used the term ‘Harijan’ (Son of God) instated of Untouchables (Borooah, 2017; Jodhka, 2017; Sooryamoorthy, 2008; Thorat & Joshi, 2020).
After Independence, the constitution of India in 1950 adopted a special provision for the upliftment of the socio-economically weakest peoples in the caste hierarchy or those considered as ‘Untouchables’ in the Varna system. Article 341 of the constitution included Scheduled Castes (SCs) for whom special benefits were given to uplift them from their backwardness in societies because of the caste structure (Chandidas, 1969; Fazal, 2017; Sashittal, 2023; Srinivasan & Kumar, 1999). Later on, under the protections of the Civil Rights Acts of 1955 and 1976, many anti-Dalit actions become offences (Borooah, 2017). According to the Census of India reports of 2011, the total population of SCs was 201,378,372, among them 103,535,314 were male population and 97,843,058 were female population. At the same time, Punjab (31.94%) and Himachal Pradesh (25.19%) have the highest SC population among all states and union territories (India, 2011).
Migration, a dynamic response to social and economic challenges, serves as a significant mechanism through which affected individuals and households seek to escape the constraints of untouchability. Migrants are people who change their residences permanently or semi-permanently within the country or across borders (Anderson & Blinder, 2011; Lee, 1966) whereas migration aims to achieve geographical equilibrium between the supply and demand of human resources and to maximize labour productivity (Das et al., 2021). The census report of India for 2011 show that the population of SC migrants was 62,971,447, which is 31% of its total population. Among them 16,755,153 male migrants and 46,216,294 female migrants are part of India’s internal migration. The dynamics of migration among the SC population have been explored through various studies. Keshri and Bhagat examined temporary and seasonal migration using data from the 55th round of the National Sample Survey, highlighting the significant contribution of SC migrants at the national level (Keshri & Bhagat, 2010). Other important research, such as ‘Migration and Human Development in India’ by Deshingkar and Akter (2009), reveals that the major sectors employing migrants predominantly consist of those from economically lower strata and lower caste groups (Deshingkar & Akter, 2009). Other related studies done at national and sub-national levels in India includes ‘An analysis of rural-urban migration in India’ (Bhattacharya, 2000), ‘Internal migration in India: are the underclass more mobile? (Bhagat, 2012), ‘Rural to urban migration: a district level analysis for India’ (Mitra & Murayama, 2009) and ‘Intercensal change of SC migrants in India: based on census data, 2001–2011’ (Dakua, 2021). These studies collectively provide a broader understanding of the migration patterns, yet there remains a dearth of research specifically addressing caste-based migration and the impacts of untouchability, discrimination or exploitation. A notable study on SC ‘Dadan’ migrants in Odisha reveals that despite efforts to improve employment conditions and promote caste reforms, these measures have largely failed to safeguard SC migrant workers (Acharya, 2021). So, the primary focus of this study is to explore the association between untouchability experienced by household members and the migratory status of India’s SC community. Additionally, the study investigates whether there is a relationship between various demographics, economic and societal representation, and networks inside and outside the community.
Methods
Data
The IHDS-II, 2011–2012, a jointly conducted household survey by the National Council of Applied Economic Research (NCAER), New Delhi, and the University of Maryland, served as the basis for the study’s data. The survey covers all of India’s states and union territories except for Andaman and Nicobar and Lakshadweep. The 42,152 sample households being surveyed are distributed throughout 384 districts, 1,420 villages, 1,042 urban blocks and 276 towns and cities in 33 states and union territories. Through stratified random sampling, both the rural and urban samples are gathered. 83% of the households surveyed for the IHDS-I in 2004–2005 were re-interviewed for these data (Desai et al., 2018), and a total of 2,134 additional homes were surveyed for the IHDS-II (Desai & Vanneman, 2018).
Study Sample
This study includes members of SC households who experienced untouchability in the last 5 years and their migration status in India. The total sample in this study is about 42,152 households, among which are 728 households with missing information on migration status and 41,424 households with complete information on migration status. There are 33,292 SC households with missing information on members who experienced untouchability in the last 5 years, and 8,132 SC households with complete information on members who experienced untouchability in the last 5 years. Additionally, there are 152 SC households with missing information on any independent covariates. For the final study, the total sample includes 7,980 SC households with complete information on migration status and members who experienced untouchability in the last 5 years. The complete sample selection criteria for this study are shown in Figure 1.
Selection of the Study Sample.
Variables Description
Outcome Variable
Migration status is considered an outcome variable in this study, which is based on the question, ‘Have you or any member of your household left to find seasonal/short-term work during the last 5 years and returned to live here?’ The response was categorized as yes and no.
Key Explanatory Variables
The key explanatory variable in this study is ‘Untouchability’, which is based on the question, ‘In your household have some members experienced untouchability in the last five years?’. The response was categorized as yes and no. This question was only put to SC households during the time of the survey.
Other Covariates
On the basis of the existing extensive literature reviews, this study includes different types of independent covariates. These covariates were classified into three following sub-groups.
Demographics Status
Four subgroups were created to categorize the demographic status: no schooling, primary, secondary, and secondary and above were the categories used to classify the highest households’ adult education. The religion was categorized into ‘Hindu’ and ‘non-Hindu’ categories. The place of residence was categorized into rural and urban categories. The response was classified as either rural or urban, based on the criterion ‘Urban residence from Census 2011’. North, Central, East, West, North East and South were used to categorize the regions (IIPS & ICF, 2021).
Economic Status
The economic status was categorized into two sub-groups: occupation was categorized as a primary sector, secondary sector, service sector and others. The occupation was measured by ‘What was the primary occupation of the household head father/husband father (for most of his life)?’, and there were 116 types of occupation responses. The occupation was categorized on the basis of the national classification of occupations in 1968. The wealth was categorized as poor and non-poor, which were measured by ‘According to you, is your household poor/middle-class/comfortable?’, and the response categories were poor, middle class and comfortable.
Society Representation and Network Status
The status of society representation and network was categorized into three sub-groups: engagement in social and political activity as yes and no. The engagement in social and political activity was measured by ‘Now, I would like to know about the groups or organizations that you and others in the household belong to’, and the response was categorized as yes and no. The social network within the community was categorized as yes and no. These were measured by ‘Do you or any members of your household have a personal acquaintance in someone who works in any of the following occupations…. Among your relatives/ caste/ community’, and the response was categorized as yes and no. The social network outside the community was categorized as yes and no. These were measured by ‘Do you or any members of your household have a personal acquaintance in someone who works in any of the following occupations…. Outside the community/caste?’, and the response was categorized as yes and no (Thorat & Joshi, 2020).
Statistical Analysis
Descriptive statistics were used to analyze the socio-demographic characteristics of the study population (migrants and non-migrants). The significance level of experiencing untouchability in the last 5 years across socio-demographic variables of migrant and non-migrant households’ members was examined through the chi-square test. The spatial analysis was examined through ArcMap 10.7. The determinants of experiences of untouchability among migrant households’ members were examined by binary logistic regression. The equation of the logistic regression is as follows:
The regression models, both unadjusted and adjusted, have been developed to examine the relationships between explanatory, covariate and outcome variables. The result was presented in the form of odds ratio (OR) and with a 95% confidence interval (CI). All the statistical analyses were computed using StataMP 14 (StataCorp., 2021).
Results
The socio-demographic profile of the migrant and non-migrant households has been shown in Table 1. A total of 21% of the households had only secondary and above education (non-migrants 22.06% vs. migrants 12.29%). Almost 71% of the migrant households were associated with the occupation in the primary, sector which is higher than the non-migrant households (62.67%). In terms of wealth distribution among households, the data reveals that a greater percentage of migrant households were classified as poor (76.98%) compared to non-migrant households (59.34%). Social networks within the community were higher among the non-migrant (44.02%) households as compared to the migrant households (36.55%). The majority of migrant households resided in rural areas (92.62%), compared to non-migrant households (71.73%).
Socio-demographics Characteristics of Study Population.
Figure 2 presents the proportion of migration experiencing untouchability of any member from the household. Overall, around 16% of any member from the households migrate due to experiencing untouchability, which is much higher than those who did not experience untouchability (8.75%).
Prevalence of Migration Due to Experiencing Untouchability.
The prevalence of migration due to experiencing untouchability across various socio-demographic characteristics is shown in Table 2. Households with a poor wealth status accounted for a significantly higher proportion of migration (79.62%) due to experiencing untouchability, compared to those having a non-poor wealth status (20.38%). A higher percentage of individuals residing in rural areas (98.10%) were prone to migration due to untouchability compared to their urban counterparts (1.90%). A significant regional difference was also observed in Table 2 where migration among the households from the central region (44.08%) were much higher to experiencing untouchability compared to other regions such as North region (21.33%), East region (26.07%), West region (2.37%) and South region (6.16%) in India.
Prevalence of Migration Due to Experiencing Untouchability Across Socio-Demographic Characteristics.
Figure 3 revealed the state-wise differences of migration due to experiencing untouchability. A positive relation has been observed between migration and the untouchability of any member from households among all the states. Among all the states, Uttar Pradesh (21.94%) experienced the highest level of untouchability, followed by Rajasthan (12.61%), Madhya Pradesh (9.81%) and West Bengal (9.38%). At the same time, migration due to experiencing untouchability was also high among the states. Among all the states, Migration was highest in Madhya Pradesh (33.65%), followed by Rajasthan (18.01%), Uttar Pradesh (9%), Bihar (9%) and West Bengal (8.06%).
State-wise Prevalence of Migration from Households Experiencing Untouchability.
Unadjusted and adjusted odds of migration by key explanatory variables (i.e., experiencing untouchability) and other covariates have been presented in Table 3. Five models have been used to perform the association between untouchability and the migration of any member from households. Model 1 indicates the unadjusted effect of untouchability on migration. Model 2 includes the socio-demographic adjusted effect of untouchability on migration. Model 3 shows the adjusted economic and occupational effect of untouchability on migration. Model 4 includes the social network adjusted effect on migration and Model 5 shows the adjusted effect of all the covariates of untouchability on migration.
Unadjusted and Adjusted Odds of Migration by Key Explanatory Variables and Other Covariates.
*p ≤ .05; **p ≤ .005; ***p ≤ .001.
The results from Model 1 revealed that the unadjusted odds of migration were estimated to be 2.03 times (95% CI: 1.71, 2.42) experiencing untouchability. After controlling socio-demographic factors such as level of education, religion, place of residence and regions, the odds of migration were estimated to be 1.56 times (95% CI: 1.18, 2.07) (Model 2) experiencing untouchability. Results from Model 3 showed that the odds of migration were estimated to be 1.74 times (95% CI: 1.34, 2.26) while adjusting to economic factors such as occupation and wealth. At the same time, Model 4 has been adjusted to engagement in any social and political activity, social network within the community and social network outside the community, and the results showed that the odds of migration were estimated to be 1.92 times (95% CI: 1.49, 2.47). Model 5 has been adjusted to all the covariates and the results showed that the odds of migration among those who experienced untouchability were 1.51 times (95% CI: 1.15, 1.98).
Discussion
The current study investigates the migration status of the SC community as a result of experiencing untouchability. An evident difference in the status of migration has been found between those in this study who experienced untouchability and those who did not. Deeply rooted in the caste system, untouchability has led to systematic marginalization, discrimination and oppression of the Dalits (formerly known as ‘untouchables’). Many Dalits are caught in a cycle of poverty and exclusion because they are denied access to education, jobs and basic facilities. As a result, migration becomes an attractive option for Dalits seeking to escape the oppressive conditions and pursue a better life. The migration patterns of Dalits are driven by a desire for improved socio-economic prospects, increased social mobility and the hope of breaking free from the discriminatory norms and practices they face in their hometowns or villages.
When the untouchability induced migration among different socio-economic groups was explored, it was been found that poorer households from rural setups faced greater untouchability and hence migrated from their villages. As previously stated, individuals belonging to the Dalit community, commonly referred to as untouchables, frequently encounter discrimination when it comes to accessing economic opportunities. Dalit day labourers have a hard time finding work in both the agricultural and non-agricultural industries, especially in the picking of fruits and vegetables. There is a lack of demand for the food items that Dalit vendors are selling (milk, fruit, vegetables and prepared foods) since Dalit women are not allowed to work in households and, more specifically, not allowed to cook or clean food grains. Both Shah et al. (2006) and Thorat and Neuman (2010) provide extensive documentation of these and other discriminatory outcomes of untouchability (Shah et al., 2006; Thorat & Neuman, 2012). This exclusion from vital resources and opportunities perpetuates their marginalized status and inhibits their ability to progress economically. Consequently, these social barriers become a significant factor in perpetuating poverty among the Dalits, ultimately leading them to remain trapped in a cycle of untouchability, where they find it challenging to break free and pursue meaningful development.
When exploring the regional variation in untouchability induced migration, this study shows that North Central and Central Plain have witnessed a higher prevalence of untouchability induced migration compared to other parts of the country. The possible reason could be these regions’ traditional characteristics of a rigid social structure and deeply entrenched caste-based divisions. The hierarchical nature of the caste system in these regions has led to a more severe form of untouchability, where Dalits face extreme discrimination and social exclusion. Alongside this, these regions are home to communities that have been more resistant to social reforms and progressive changes. Traditional customs, rituals and beliefs often reinforce the discriminatory treatment of Dalits, perpetuating the cycle of untouchability.
Since the association between experiencing untouchability could be modified by the socio-demographic and other covariates associated with societal norms, we further tried to adjust their effect in the regression model. The findings were still significant, and the selected covariates rarely modify the association between experiencing untouchability and migration behaviour. Among the socio-demographic covariates adjusted in the model, illiteracy and rural residency show a higher level of untouchability induced migration. Education appears to have a direct and detrimental effect on the household members’ propensity to engage in the practice of untouchability (Wodtke, 2012). Families with at least one educated member are less likely to practice the custom (Borooah et al., 2015; Thorat & Joshi, 2020), which may further reduce the migration lead by experiencing untouchability.
The key distinction observed in the research is the contrast between rural and urban areas. Urban spaces promote a more uniform culture by densely accommodating migrants in work and living environments, leading them to embrace modern values of shared public and personal spaces. In cities, newcomers might conceal their caste backgrounds or adopt new ones. Meanwhile, rural regions uphold long-standing customs and traditions over generations. Although villages are becoming more linked to towns and witnessing migration, they are expected to change gradually, though at a slower pace compared to urban areas. Previous studies have also shown similar findings that align with our current study (Thorat & Joshi, 2020).
Community networks play a key role in influencing family behaviours. The likelihood that a household will abstain from the behaviour (untouchability) increases with the extent of its network outside of its immediate neighbourhood (Thorat & Joshi, 2020). At the same time with increasing community involvement the possibility of migration also increases (Munshi, 2020). Similarly, our study also shows that untouchability induced migration increases when the households create their social network outside their own community. The vice-versa situation is found when the people are less interacted with other community and more within their own community.
In comparison to other literature from India, the present study backs different strengths. First, we used the nationally representative dataset that establishes a robust and comprehensive foundation for the study on untouchability induced migration among SC populations. Second, the inclusion of multidimensional variables, such as education, income, employment and social indicators, supports a holistic exploration of the factors influencing migration among SC communities, capturing the complexity of the untouchability induced migration phenomenon. Furthermore, utilizing IHDS data assists in indicative policy implications grounded in evidence, offering valuable perspectives for policymakers addressing concerns related to untouchability and the migration of SC communities. This study also had certain limitations. There is a potential lack of representation, as the dataset may not fully capture the diverse experiences of SC migrants across all Indian regions, possibly resulting in under representation or oversampling of specific groups, and compromising the generalizability of the findings. Additionally, the specific timeframe of the IHDS dataset introduces temporal bias, limiting the study’s ability to account for evolving migration patterns and recent changes in untouchability induced migration. The scope of variables is constrained, with the dataset possibly omitting critical factors influencing migration decisions and outcomes. However, with the help of relevant data, future studies should investigate the potential impact of changes in untouchability practices on migration from various regions in India for a comprehensive analysis.
Conclusion
In conclusion, this study delved into the migration dynamics of the SC community as a response to the experiences of untouchability. The research has revealed a significant divergence in migration patterns between those who have encountered untouchability and those who have not. Untouchability, deeply entrenched in the caste system, has perpetuated systematic marginalization and discrimination against Dalits, leading to their oppression. Denied access to education, jobs and essential facilities, many Dalits face a cycle of poverty and exclusion. Consequently, migration emerges as an appealing avenue for Dalits seeking liberation from oppressive conditions and aspiring for a better future. The investigation of untouchability induced migration across different socio-economic groups has illuminated that impoverished households in rural settings are more susceptible to untouchability and hence are compelled to migrate from their villages. The discriminatory outcomes of untouchability, such as restricted economic opportunities and exclusion from various resources, maintain the marginalized status of Dalits, perpetuating their poverty and hindering economic advancement. This perpetuation of social barriers creates a self-reinforcing cycle of untouchability, making it challenging for Dalits to break free and achieve meaningful development. In essence, this study sheds light on the intricate interplay between untouchability, migration, socio-economic factors and regional influences. It underscores the pressing need to address untouchability and its far-reaching impacts on migration patterns, socio-economic mobility and community dynamics, thereby paving the way for a more inclusive and equitable society.
Footnotes
Data Availability
IHDS 2 data files are publicly available from the Data Sharing for Demographic Research program of ICPSR, the Inter-university Consortium for Political and Social Research, at ICPSR Study 36151. ICPSR requires users to register by creating a user ID. Once registered, users can download all relevant data and documentation for their research needs.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The authors received no financial support for the research, authorship and/or publication of this article.