Complexity and lack of adequate technical and operational awareness on the use of biometric system for the October 20th.2018 Election
Kabul: Independent Election Commission, after a sever insistence drawn by political parties and coalitions, in contrast to their previous claims which were based on time inadequacies, have pledged to political parties and Afghan citizens on using biometric on the day of elections, but to what extent this scheme of IEC is deemed practical on the day of elections which is only 24 days far from now, and that to what extent use of biometric can actually assist in mitigating or eradicating election fraud from grass-root level still remains a questions for many, and till now no satisfactory response has been provided by the election management bodies in this regards.
Transparent Election Foundation of Afghanistan call for IEC’s attention to clarify the following questions with vivid and rational answers and to elaborate on the plan for tackling such speculated problems:
1. Every single electoral activity’s procedure must be coordinated with the concerned election organizations and should be made accessible to them prior to execution, but till now, IEC has not shared procedure of any sort with any of the electoral organizations.
2. Election observer organizations, candidates or parties have no way to verify and audit the authenticity of the results of the biometric system. If the Statistic Office or IEC would declare for example that only 5000 voters voted twice, there must be a bulletproof process for parties and candidates to audit if this is indeed the result.
3. The Software source code needs be reviewed by representatives of candidates and parties prior to election. Devices will need to be randomly checked upon dispatch and return by the stakeholders to audit that indeed the reviewed software is installed and it has not been manipulated in order to create fake results and fake audit logs.
4. Data security in the biometric system is another major issue for ensuring observer foundations, political parties, and candidates’ trust.
5. Looking at the ground realities, it is impossible to capture with a single fingerprint reader ten fingerprints, even four will be challenging. This is due to the maximum time that can be spent per voter without creating growing lines and an emotionally loaded atmosphere with increasing pressure on officials, as a result making multiple voting possible on E-day, and what makes it surprising is that this problem has nothing to do with fraud and sabotaging the software.
6. Even if picture would be added, which is questionable to succeed with all female voters, capturing a quality picture suitable for biometric de-duplication is practically impossible in the polling station reality with the proposed equipment. Lighting, expression etc. need to be all perfect and even if the equipment would be suitable, it would slow down the process further.
7. If groups willing to stop the process, they can easily stall the biometric system either by damaging the hardware, or even an IEC employee may claim of lacking requisite skills to operate the system properly. Moreover, chances of deceiving voters on disclosing their identities or that the system is out of order also exist.
8. There are many ways to trick the system. Capturing the fingerprint or picture with insufficient quality for biometric de-duplication is the easiest way. This can be both facilitated by the voter and the operator. Even if quality filters were applied, a voter would need to be rejected if biometrics can’t be captured. This is in reality impossible, as some people have bad fingerprints unsuitable for biometric capturing. Besides being illegal to reject a voter on this basis it would create potentially violence in the polling station. These filters would furthermore slowdown the process and make an already difficult process more impossible to succeed.
9. TEFA’s research shows that the operational roll out of such project has never been achieved in such a short roll out period worldwide. For a system to work hard and software need to be in perfect synchronization with the human factor (operators and voters) and best practice, processes and procedures need to be in place. Neither the IEC/Statistics Office/UN nor the company that is supposed to provide the system has experience in such a roll out in polling stations on Election Day.
10. Use of technology can deliver transparency only if it entails details and that such details are shared.
11. Use of technology can mitigate electoral fraud in a real senesce, only if it is used prior to the day of elections (in voter registration process), and that there should be a pre-defined relationship between the voter database and Election Day biometric. Suppose if a person votes 10 times, and casts 10 ballots in the ballots, then how would the extra 9 ballots be detected and distinguished? How significant the biometric system would be in elections when there is no proper de-duplicated biometric voter database available for matching in such conditions of Afghanistan.
12. On E-day only the combination of biometric authentication and Electronic Voting will be effective as voting can only proceed when the biometric process is applied. Sabotage can only stop the process, but fake results can’t be created.
13. How much time is left for preparations, audit process and help through organizations with real know how in this field for implementing and assessing the biometric system.
14. IEC has already put together a comprehensive public awareness package for raising people’s awareness but did not include awareness on the biometric system, in which one of the steps is to run an TV infomercial explaining the voting procedure, with such tight deadline, what are the possibilities for these messages to be rectified?
TEFA requests all the election stakeholders, especially IEC, political parties and coalitions, and electoral organizations to consider time, resources, need, and conditions into account before embarking on any decision so that it does not fail and does not secure a bad image.