Do identity systems work for women? How social relations trump technological systems.

Episode 6 - Article 2

Something we’ve found during our research is that, contrary to what we’d hoped, the implementation of new identity systems often places users in tricky situations which reinforce traditional forms of patriarchy. Achieving the right balance between collecting information that is required for a successful transaction and doing it in a way that satisfies both parties is not always easy.

In Assam, we observed a couple of instances of women being placed in situations that clearly made them uncomfortable. At the Grameen Bank in Bilgaon, we saw a woman who had come to check if her salary had been paid into her bank account. In order to access her account, she needed to provide her thumbprint via biometrics, but it couldn’t be read when she placed her hand on the device.

As a result, the bank agent, Maqbul, a man, intervened by grabbing her hand and placing her thumb in the right position – something she was clearly, and understandably, unhappy with.

On the one hand, this was probably something Maqbul felt he had to do in order to properly execute his duty. On the other, we are left with a situation where the woman was immediately rendered powerless, and it unwittingly reinforced the power play that is typical of male-female relationships in most societies.

One of the most commonly-claimed benefits of identity systems is that they can empower individuals by giving them better access to their data, in what is purported to be a secure manner.


The exchange above showed us that identity systems are simply not immune to gender dynamics. Conversely, were there any other options offered to Maqbul, as the agent who probably performs the same actions on a weekly, if not daily, basis?

We also spoke to Narisa, a woman who had come to the bank to open an account so that she could apply for a loan from the government that she, as a woman, was eligible for. It transpired that although the loan was going to be taken in her name, it was actually for her husband. We asked why he couldn’t apply for the loan himself and she said it was because “it was only for ladies”.

She added: “The [river] Brahmaputra eroded our lands and the government has provided us a plot of land. My husband is in business of hatching [chicks] and wanted a loan to expand.” Being uneducated, she was upfront about this circumvention of the rules of the loan, and possibly unaware that it didn’t show her in a positive light. This incident revealed that identity systems that are put in place to help a particular gender – women, in this case – can easily be used to reinforce the patriarchy by the very people they are meant to help. Maqbul, for his part, wasn’t too concerned by the fact that the loan might be used by her husband. It was more or less accepted that this was a common practice. If he did complain, he’d probably only create more trouble for himself.

After Narisa left, our Sylheti translator stated: “Actually they have some traditional rules, especially in backward societies, that women should be at home only, and she should not work outside the home. This is the main thing.” His biased language reminded us of the clash of contexts for Narisa – she had to navigate her domestic situation, which required her to be submissive and take a less important role, but also the one she was placed in at the bank, where she was required to assert her identity and her role as a potential loan beneficiary. Narisa’s situation is made complex due to the conflict between the two different roles that she was required to perform at the same time – and she was also uncomfortable when we asked to take a photograph of her, out of fear of retribution by her husband.

Identity systems should be objective, and users of a system have a right to expect that they will not be discriminatory. But as we have seen, systems that use technology like biometrics –  and those that have strict processes around identity verification – can’t quite disrupt traditionally held notions of superiority in societies like India. Further, with intermediaries as part of the system, like Maqbul, identity systems are often tinged with human bias – they are not purely machine-driven, and it would be a mistake to consider it as such.

We’re aiming with this research to highlight how users of identity systems exist within the context of relationships, not just the process of creating or using identity credentials.

By doing this, we open up issues such as power and individual agency for wider critique – which otherwise might go unchallenged.