Partition Revisited: Bridging History and Technology to Trace Refugee Rehabilitation

This is a guest post by Gursimran Kaur Butalia.

The Partition of India in 1947 remains a deeply personal and profoundly transformative chapter in the history of the subcontinent, shaping millions of lives, including my own family’s. It marked one of the largest human migrations in modern history, with over 14 million people being displaced amidst unprecedented violence. Refugees fleeing West Punjab and other parts of the newly formed Pakistan sought shelter in Delhi and Punjab, reshaping these regions’ urban and rural landscapes. Their stories of resilience and adaptation offer valuable insights into state policies, community dynamics, and the long-term impacts on urbanisation and social cohesion.

While the Partition, as a topic, has been extensively studied due to its importance and relevance on the Indian subcontinent, the abundance of data available suggests the need for adopting a distinctive approach when examining such a diverse and complex subject. Given the substantial body of research containing valuable information, the focus has now gradually shifted towards studying and compiling this knowledge in a more nuanced and technologically advanced manner. This article presents various digital tools and methods being employed in the investigation of rehabilitation in post-Partition India, specifically in Punjab and Delhi, and potential future applications to deepen and expand our understanding. 

By leveraging digital tools such as geospatial technologies, automatic transcription, and data analysis, we can explore refugee rehabilitation in a profoundly cathartic way. Through these technologies, we can uncover patterns of displacement and settlement, building a more nuanced narrative of resilience and survival in the face of overwhelming adversity. 

Geospatial Technology 

Geospatial technology, such as Geographic Information System (GIS) mapping, offers transformative tools for studying the rehabilitation process following events like the Partition of the Indian subcontinent. Using GIS, researchers can visualise and analyse migration routes, settlement patterns, and the distribution of relief resources with spatial precision. For instance, a study of Partition migrations demonstrates how GIS tools, such as ArcMap, can be employed to geocode data points from survivor testimonies, enabling the mapping of migratory inflows and outflows at the district level. This research revealed significant regional concentrations, particularly in urban centres like Lahore and Delhi, where rehabilitation efforts were concentrated. 

Using GIS mapping to study Partition migration. Credit: Tiara Bhattacharya.

GIS analysis tools such as Kernel Density mapping (the Kernel Density tool determines the density of features in their immediate vicinity, which can be estimated for both point and line characteristics) can be used to highlight areas with higher refugee camp concentrations, correlating these with the socioeconomic status of displaced populations. These insights can help identify spatial iniquities in resource allocation and can guide more equitable rehabilitation strategies. Additionally, ArcGIS tools like Global Moran’s I (the Spatial Autocorrelation tool calculates spatial autocorrelation by taking into account both feature locations and feature values) and High-Low Clustering analyses can be used for the evaluation of socioeconomic clustering, offering a holistic understanding of how economic vulnerabilities influenced migration and settlement. 

One notable example of such work is a collaboration between GISCorps and The 1947 Partition Archive. This initiative utilised GIS tools to map survivor testimonies, enabling the visualisation of migration routes and settlement patterns during the Partition. Additionally, the Punjab Remote Sensing Centre (PRSC) in Ludhiana, Punjab, though primarily focused on applications like agricultural monitoring, demonstrates the capability of remote sensing technologies to study demographic and landscape changes, potentially applicable to post-Partition studies.

Through utilising mid-20th century archival imagery and aerial photographs, remote sensing can significantly enhance the study of the refugee rehabilitation process. Researchers can map out land-use changes triggered by massive demographic shifts, such as the creation of model villages in rural Punjab designed to integrate displaced populations into agrarian economies following the influx of refugees.¹ By comparing historical imagery with modern data, researchers can identify transformations in agricultural landscapes, highlighting how these regions adapted to demographic pressures and new economic realities.

Despite challenges like limited historical data, geospatial technology provides a powerful framework for reconstructing complex historical events, assessing their socio-economic impacts, and informing the planning of relief and rehabilitation programs in similar contexts.

Other Digital Humanities Tools

A lot of official government documentation has been found from the pre- and post-Partition era, including government records and compensation lists. These provide a rich but challenging resource for understanding the refugee crisis and rehabilitation processes. Digital humanities tools, such as Optical Character Recognition (OCR) and text mining, enable the systematic analysis of this archival material, as demonsrated by the Crowdsourcing Memories of the 1947 Partition of British India project.² By digitising records like land allotment documents and refugee registries, and converting them into searchable formats using OCR, researchers can use Natural Language Processing (NLP) techniques to extract recurring themes and patterns. 

Text mining can also be used to highlight important policies, such as housing and employment for refugees, or reveal the obstacles they faced during rehabilitation through official documentation. This approach not only makes large datasets more accessible but could help uncover critical insights into the state’s administrative response to one of history’s largest human displacements. 

Aside from these, interactive storytelling platforms, such as StoryMapJS, further enhance the understanding of Partition experiences by combining geographic data with personal narratives, as has been done with the 1947 Partition Archive. These tools allow for the mapping of individual journeys, offering a visual and textual representation of migration paths from cities like Lahore to Delhi or Amritsar. Such platforms make it possible to contextualise these journeys within broader historical events, creating a multidimensional narrative that enriches public engagement and academic research. 

Together, OCR, text mining, and interactive platforms bridge the gap between archival data and contemporary analysis, providing innovative ways to study and represent the complexities of the post-Partition rehabilitation process.

Socioeconomic Data Analysis Tools

Statistical tools like SPSS, and programming languages such as R and Python, which are widely used for statistical computing, along with machine learning algorithms and data visualisation software, can play a crucial role in analysing the refugee rehabilitation process. These software programs can examine census data and compensation schemes, which can uncover trends in resource allocation, such as land distribution patterns in Punjab and Delhi, and help researchers evaluate the success of rehabilitation policies in both urban and rural settings. 

Machine learning models can help process extensive datasets, such as compensation records or demographic surveys, identifying correlations and anomalies,³ which can also help highlight issues like caste or gender-based disparities in understanding the rehabilitation process. Tools like Tableau and Power BI can further enhance these insights by creating clear, visual representations of refugee rehabilitation patterns, such as mapping the timeline of camp closures and permanent housing developments. 

Environmental and Urban Perspectives

Technology can also be used to study the long-term impact of migration on both the environment and urban infrastructure. Environmental monitoring tools, such as sensors and historical climate data, can track how refugee settlements influenced local ecosystems. For example, the creation of agricultural colonies in Punjab required vast land and water resources, significantly altering the natural landscape, which merits further investigation.

Urban planning software, like ArcGIS Urban, can help analyse how refugee colonies were integrated into cityscapes, offering insights into how these settlements influenced rapid urbanisation in regions like Delhi and Punjab. Furthermore, 3D modeling and reconstruction technologies allow researchers to preserve and study the original structures of refugee settlements, many of which have been lost to redevelopment over time. These technological tools provide a comprehensive view of the environmental and urban transformation caused by the Partition and its long-term effects.

Conclusion 

While technology shows great potential for studying the Partition, it also comes with challenges, such as restricted access to certain archival data and the resource-intensive digitisation process. Moreover, ethical considerations are paramount when working with such sensitive and personal data, ensuring refugees’ stories are handled with the respect and sensitivity that they deserve.

Studying refugee rehabilitation goes beyond academics; its ultimate goal is to honour and give tribute to the resilience of those who rebuilt their lives amid upheaval. By combining these tools, we can create a comprehensive narrative of displacement and resettlement. These insights are not only meant to preserve Partition memories but also provide valuable lessons on resilience, adaptation, and the search for stability, which is still very relevant as the world continues to face displacement and migration challenges.

Projects employing technology to study Partition and refugee rehabilitation are steadily growing, focusing on different aspects such as migration mapping, oral histories, and resource allocation. For instance, GISCorps and The 1947 Partition Archive highlight the use of mapping technologies and survivor testimonies, while organisations like PRSC showcase potential applications in monitoring demographic changes. However, much of the focus has been on data collection and visualisation, with less emphasis on integrating these insights into policymaking or community development. Despite this, there is hope that future projects will expand into more diverse applications, combining interdisciplinary approaches to address gaps in historical research and modern refugee crises. As technological accessibility increases, these initiatives are likely to grow, offering new ways to engage with history and inspire solutions for contemporary displacement challenges.

---

1. Amrita, Paul and Prithvish Nag. “Pattern of Post 1947 Refugee Resettlement in India” in International Journal of Geology, Agriculture and Environmental Sciences, 3.1 (2015): 68-74 ↩︎
2. This study conducted by the Lakshmi Mittal and Family South Asia Institute and Harvard’s Laboratory for Innovation Science, collected over 2300 oral histories (2017-2019) from survivors of the 1947 Partition. It focused on underrepresented voices, including women, Muslims in India, Hindus and Sikhs in Pakistan and Bangladesh, Parsis, Dalits, and Christians. Using the Ambassador Model and partnerships with groups like the Citizens Archive of Pakistan, the project employed geospatial mapping, text mining, and quantitative analysis to examine migration routes, refugee camps, violence sites, and broader socio-economic and emotional experiences across India, Pakistan, Bangladesh, and the diaspora.
   ↩︎
3. For example, see: Rashidi, Atina Najahan Binti Mohd et al. “Computational analysis of dystopian elements in the partition fiction: A machine learning approach to the indian English novels.” (2024). This research focuses on literature based on the Partition, but the idea can be further developed and applied to socioeconomic data on the post partition rehabilitation process. ↩︎

References

Amrita, Paul and Prithvish Nag. “Pattern of Post 1947 Refugee Resettlement in India” in International Journal of Geology, Agriculture and Environmental Sciences, 3.1 (2015): 68-74.

Bhattacharya, Tiara. “Migratory Flows & the Partition of British India.” May 7, 2019.

Rashidi, Atina Najahan Binti Mohd et al. “Computational analysis of dystopian elements in the partition fiction: A machine learning approach to the indian English novels.” (2024). https://doi.org/10.1016/j.ssaho.2024.100897