Biometrics at work: replacing humans, companies, and places at the heart of technology

Since at least ten years, we have often heard speaking about biometric technologies and their promises in various fields: distributing state finances and public subsidies to citizens, allowing the poor access to financial services and banking as well as in the southern countries the issue of an “identity gap” and questions how to solve electoral disputes and state legitimacy etc. However, even if this technology has been often described as a way of owning democratic attributes, not a lot of people beyond practitioners really understand what “biometric” means and what biometrics actually do. This blog article aims at describing biometric technology for civil identity management in an accessible and simple way by focusing on the production process of the Senegalese identity card.[1] The main goal of this blog article is that citizens – to whom biometrics are mainly addressed and who are the first implicated actors within it – shall be able to identify more clearly what this technology means that is increasingly occupying the public space. The case study example of biometrics at work considers the Senegalese ECOWAS identity and electoral card that is also known as biometric technology for civil and electoral identity management (which is, at the same time, the topic of my thesis). For doing so, I will “humanise” technology by adding geographic places, company names as well as people who work building, entertaining, and sustaining technology. Digital technology is about automatization, but this automatization is inscribed in a larger network of administrative work done by social actors.

So let’s go to the field: Senegal is a West African country and member of the Economic Community of West African States (ECOWAS), which, since 1975, has promoted economic integration of member countries, collective self-sufficiency, and free circulation across borders. In 2016, ECOWAS decided that member states introduce a new biometric identity card, which would facilitate intra-regional mobility of people and commodities. Senegal introduced that card shortly after, using it, in addition, as a voter’s card for elections. Thus, biometrics and elections have become closely interconnected in Senegal.

Since the 1970s and 1980s, Senegalese civil society and opposition parties have been protesting around the credibility of electoral rolls. They protested against the ruling party, at that time, the Parti socialiste, that managed the electoral apparatus of the State such as electoral lists and districts, in order to utilise these for their own benefits, mainly by increasing the numbers of votes for their candidates. Those from the opposition argued that government institutions were added fake identities in electoral rolls, identities through which real persons were then able to vote for candidates of the ruling party. For example, having access to the electoral roll offers knowledge which the ruling party was able to use for reinforcing its action. Accessing information on voters and electoral maps provide power for making projections and statistics and for eventually reorienting electoral campaigns, a knowledge to which opposition did not have access. Debates around reliability, calls for transparent inspection and the refusal of the electoral roll have become until today a repertoire of political protest and action in Senegal as in other African countries.

Image mise en avant: Direction de l’automatisation des fichiers, Dakar, Sénégal, février 2019, photographie © Cecilia Passanti

I still don’t know what came first – the protests or the computerization of electoral registers. Some say that creating in 1978 the Direction of the Automatization of the Files (DAF) as part of the Ministry of the Interior and Public Security was an institutional answer to electoral protests on register reliability: introducing computer science in citizens data rolls aimed at boosting confidence. Others, however, say that introducing computer science and creating the Direction of the Automatization of Files produced electoral protests on the reliability of the electoral registers. People mistrusted the Ministry of Interior and thought it owned two different electoral registers, one to enrol real voters and the other one for cheating elections. The DAF office is today based within the headquarters of the Ministry of Interior and of Public Security in Dakar, Plateau.

Since 2005, the DAF is in charge of biometrics: It used to manage the network for managing citizen identities, and is now designated to develop and manage the network for the biometrisation of these identities (Awenengo Dalberto, Banégas, and Cutolo 2018). Within the DAF offices at the Ministry headquarters, we find today two private service providers that develop and maintain biometrics for identification. These two companies are Synapsys Conseil and Iris Corporation Berhad. Synapis Conseil is a Senegalese company founded in France (Lyon) by a Senegalese manager. In 2000, the founder decided to sell the French headquarters and to focus on its main office in Senegal, some years after they became the main government collaborator. Iris Corporation is a Malaysian company that has worked for the Senegalese government since 2007. At that time, the company won a public tender to provide biometric passports in Senegal.[2] Since the introduction of the new ECOWAS biometric card, the company has expanded its market to identity cards. In this project, it is overseeing the work of Synapsys. Between their offices in the public building of the Ministry, they are working on the digital management of civil identities for producing an object that embodies the whole identification system: the ECOWAS national biometric identity card.

This card responded to the 2016 ECOWAS directive to introduce a biometric ID card in all its member states as a travel document. The Senegalese government turned this into a multifunctional card that serves both as an identity and a voter card. The card contains an electronic code on which citizens data have been digitally printed: both traditional (name, surname, birthday, sex, size, delivery date, expiring date, address etc.) and digital (a picture of the citizen fingerprint). The mix of these two types of data is also known as “alphanumerical data”, because composed both of letters (“alpha”) and digital features (“numerical”). Biometrics, in regard to this card, practically mean that we have a new form of data on this electronic chip in the card, which before had been just alpha data such as name, surname, birthday, sex, size, delivery date, expiring date, address etc. This new data are digital and make reference to a unique biological feature.

The process to get an identity card is composed by three phases: the enrolment, the data treatment, the card impression. If you need a new identity card, you first have to go to a police station or enrolment office. Here, you will find police officers armed with a computer, a camera and a fingerprint scanner, who will ask you your alpha data (name, date of birth etc.), but also to look in the camera and to put your fingers on the fingerprint scanner. This scanner will take some pictures of your fingerprints and will record them in your file. These pictures are than translated into a unique identification number.

All these data are than stored in a database and afterwards treated. The data treatment is the most meaningful task of biometrics. This is done through a software, which runs on servers owned by the Ministry that are able to compare citizens fingerprint pictures, and which can tell, if they look the same or not[3]. The software can analyse the percentage of resemblance of two fingerprints pictures and can indicate, if these two pictures look 30% the same or 99%. If the percentage is very high, it probably means that the same person has been registered twice. So, one of the two recorded files has to get removed. Once the software is launched to search the citizen register, it compares all fingerprints pictures, while looking for similarities. This process is called “deduplication”. The choice between two files is done by police officers who open a little research on the citizen and try to find out what happened in order to keep the “good” file.

Once the file is defined as “unique”, data has to get printed on a card. This is done in a Printing Room where police officers are in charge to transfer citizens’ files to a big biometric laser printer, which is charged with empty blank smart cards. The laser printer personalizes the card with a picture, alpha data, and the numerical date on the integrated microchip. Once the process is over, the card and the data are verified and can then be sent back to the registration centre, where the citizen went for applying for the card, and where the citizen get back for receiving it. The administrative system thus closes the circle back to the citizen. Her or his personal and biological identity has been transformed into an administrative and formalised identity.

[1] Another blog article under the title ‘How biometric works’ explain more generally biometrics without a specific fieldwork https://science.howstuffworks.com/biometrics.htm

[2] The public tender was about 118 072 260 000 FCFA (Rapport publique de la court des comptes sénégal 2009) for a twenty years project. Marché n°F/085/FM pour l’établissement des passeports numérisée du 3 septembre 2007.

[3] The software is named Automated Fingerprint Identification System (AFIS)

Passanti Cecilia

est doctorante à l'Université Paris V Descartes, affilié au laboratorie Ceped et au programme de recherche IHA-CREPOS sur la »bureaucratisation des sociétés africaines«

More Posts

Follow Me:
Twitter


Passanti Cecilia

est doctorante à l'Université Paris V Descartes, affilié au laboratorie Ceped et au programme de recherche IHA-CREPOS sur la »bureaucratisation des sociétés africaines«

Vous aimerez aussi...

Laisser un commentaire

Votre adresse e-mail ne sera pas publiée. Les champs obligatoires sont indiqués avec *

Ce site utilise Akismet pour réduire les indésirables. En savoir plus sur comment les données de vos commentaires sont utilisées.

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search