“Identification of election technology that meets the public acceptance is as important as the election itself” Burmester & Magkos, 2003 (Towards Secure and Practical E-Elections in the New Era)
The crux of this article was written a few weeks before the last Presidential elections, and was shared on my social media account. The heated passion at the time meant the message suffered series of refraction and diffraction as it travelled through partisan lenses. The elections were won and lost; most of my fears crystallized. With the card reader as a recurring subject at the election petitions tribunal, I think it is worth revisiting.
First off, by my training as a Computer Engineer, the use of technology in the electioneering process comes as a welcome development. I have always advocated the use of Information Technology in governance and in strengthening democratic processes/institutions. However, there are procedures for successful implementation of a people-centric technology that will radically change an important and sensitive process such as voting.
That said, INEC got it wrong ab initio by refusing to sort out the legal issues before deploying the technology.
You do not rush to force a piece of technology on a process governed by an existing legislation, without amending the relevant laws to reflect this. INEC didn’t do this; they missed the opportunity to do this, and so the card reader lost absolute and exclusive legitimacy without a legal backing. INEC should have known that Nigeria is not a one-man company where the owner can just decide to introduce a piece of technology that drastically alters the workflow process, without the needed stakeholder buy-in. No, in deploying public systems such as this, you must work from first principles – (1) Get the concept right (2) Ensure it is within the legal/regulatory framework.
Where the answer is “no”, then propose an amendment to the laws (3) Get your design right (4) Involve the stakeholders through all the stages (5) Ensure you get your technical specification right, to address critical issues such as security, integrity, confidentiality, etc are addressed (6) Sufficient time for stakeholder testing to build trust and confidence.
E-government must derive legitimacy from an underlying legal framework that recognizes and supports its operations. In this case, the conduct of elections is regulated by the Electoral Act, and which operates in isolation, and in complete ignorance, of a piece of technology called the Card Reader. The only thing it emphasizes is the voter register. Once your name is on the register, and you have a valid proof (Voter Card) that you are the one, then no one should stop you from exercising your civic responsibility. As long as this law did not restrict one to a particular verification process, card reader can only serve as one of the means, but not the only means to verify the identity of someone whose name is on the voter register. The appeal court ruling in the case of APC Vs Agbaje, which the ruling party was a prime beneficiary, has already addressed this. The Judge ruled that the card reader is alien to the electoral act, therefore elections cannot be cancelled on the basis of failure of it. Other judgments have also been in the same direction, except the curious case of the Rivers State elections tribunal that nullified the gubernatorial elections on the grounds of failure to stick to the card reader, yet about a third of House of Assembly elections held same day under same conditions were upheld.
It is noteworthy to mention that INEC in apparent realization of the need for a contingency arrangement, decided to provide incident forms, to allow voters whose names appear on the Voter register, but had challenges with the card reader to vote and log in the incidence. This was to ensure that valid voters are not disenfranchised on the grounds of failure of technology.
In introducing a new technology, it is always important to learn from the experience of others. In 2005, e-voting was deployed in the _Bundestag_ election (Germany). In 2009, Joachim Weisner and his son, Ulrich challenged this in court, arguing that relying on the e-voting machines amounted to “blind trust” in the machines, since they could not verify the inner workings or the integrity of the system. Germany’s federal Constitutional Court agreed with them and ruled that the use of e-Voting as unconstitutional. That is a clear case where failure to get legislative backing for a technology could make it lose legitimacy. Fortunately, they didn’t nullify the elections, but they reverted to paper voting thereafter.
For a public system to be declared fit, it is expected to go through a robust multi-stakeholder testing process to establish the systems integrity, efficiency, accuracy, reliability, and security. I am not quite sure how well the stakeholders (political parties, Civil Society Organizations, etc) were carried along right from the conceptualization, to the alpha design to the beta design, and deployment; apart from the few public demonstrations and the poorly attended mock elections. Also, a robust testing process should have included independent professional organizations that can further add credence to the integrity of the testing process. Did INEC involve the Nigerian Institute of Electrical and Electronics Engineers? Did they involve the Nigeria Computer Society? Did they work with the Ministry of Information and Communication Technology? Did INEC involve the Standards Organization of Nigeria? We also have respected COREN registered Computer Engineers, Electronic Engineers, Quality Engineers, Reliability Engineers, System Testers, etc, who could have been invited to robustly assess the system and processes. Did INEC do any of these? I’m not quite sure they did. If they did, most of the technical challenges they had with the systemcould have been spotted in good time for correction. These professionals being part of the testing process would have deepened the whole process of ensuring the transparency, efficiency, effectiveness and integrity of the system. If this was done, it would have gone a long way in building public trust in the new system. Public trust is crucial to the acceptability of any new system; it is important to carry them along in every single process. In the ruling of the German Constitutional Court cited above, the judge held that:
_“In a republic, elections are a matter for the entire people and a joint concern of all citizens. Consequently, the monitoring of the election procedure must also be a matter for and a task of the citizen. Each citizen must be able to comprehend and verify the central steps in the elections.”_
I find this very instructive. Every process in the entire electoral system must be accepted by the people and be transparent in its operations. If any piece of technology will substitute the role of citizens in the verification stage, then such must be subjected to rigorous and robust scrutiny by all stakeholders. You need time to pull this through; INEC didn’t give the citizens that time. The late demonstrations and last minute voter registrations were grossly insufficient.
Also, Pilot Testing is an important approach to deploying large systems like this. Rather than deploying it in the fiercely contested nationwide election, it should have been test run in smaller elections, where the environment variables can be controlled, in line with global best practices. Supporting this view, E.A. Alebiosu in the paper, Smart Card Reader and the 2015 Election in Nigeria, argued that: _“Moving from manual process to infusing technology in voting process is a milestone that requires high efficiency. Between 2011 and 2015, INEC conducted Governorship Elections and bye-elections in some states of the federation. Why would INEC wait for 2015 general elections to use the card reader?”_
INEC missed two clear opportunities to do that; the Ondo and Osun states gubernatorial elections were clear opportunities for a pilot test. Rather than do this, it was reported in the media that the INEC chairman said this:
“It will not be wise to use a pilot scheme in an election that will be keenly contested.”_ – Prof. Jega, Leadership Newspaper, May 14, 2014.
If this statement above was correctly reported, one would want to find out from Prof. Jega if he had already foreclosed the main election as one that will not be “keenly contested”? Well, as it turned out, the election was “keenly contested”, and the card reader performed below expectations.
That signs of failure started showing early, but again, maybe because we were too eager to show off our new found technology, we ignored them. You may recall that INEC did a so-called mock election for field testing of the card reader; a process which was not altogether hitch free. Curiously, while they announced a failure rate of 41% (finger print matching), they went on to say they were satisfied. How can you be satisfied with a machine that had a failure rate of about 41% in a critical biometric process? That error margin of 41% is an unacceptable tolerance margin for any sensitive public system such as this. In fact, one could argue that that is even a higher error margin than the manual accreditation process. As it turned out on election day, the card readers did fail en masse. The number one citizen of the country was left standing for over half an hour, as they rallied to replace one faulty card reader after another in the full glare of the viewing public and international reporters. This alone could have constituted sufficient grounds to discredit the elections and give room for the former President to challenge the outcome of the last elections; credit to him, he didn’t, and that laid the foundation for the peaceful transition that ensued. As an aside, what happened to the INEC team that supervised the contracting arrangement for this device? What warranty arrangement was in place? Did they insist on getting free replacement for the 41% that failed the finger print test? When you spend billions for devices like this, you should not accept 41% failure rate in any aspect of it.
Finally, one would have thought we would learn from the mistakes of our African brother, Kenya. The case of Kenya was a classic example of a disastrous outing for technology; and that’s not because technology is bad, but because the authorities never did what they ought to have done. Hastily deploying technology, out of excitement, without doing your home work is a recipe for disaster. Taha Ali in his article, “How (not) to deploy an Electronic Voting System”, summarized the Kenyan e-voting disaster thus:
“In several instances, fingerprint scanners refused to identify legitimate voters. Election staff forgot PIN numbers and passwords to access the EVMs. Several polling stations did not have electric sockets and batteries started to die out. When the technology failed, several polling centres reverted to back-up paper ballots (which ended up delaying the announcement of election results by a week). A single server, meant to collect results from 33,400 polling stations via SMS, collapsed under the strain.”_
Going forward, looking at things from a broader and professional perspective, it is imperative that the Ministry of Communications Technology should champion a process that will evolve a well articulated e-Government Policy that clearly outlines the guidelines in deploying public-centric systems such as this, to ensure the integrity of the system and protect the interest of the citizenry. This can be mainstreamed and given legislative backing to serve as a standard guide for deploying public systems such as this. An important lesson should be learnt here; no matter how well-intentioned the technology is, with poor planning and implementation, it becomes a recipe for disaster; it was a miracle that total disaster was averted in our case. In deploying public-centric systems and technology, it is important to get it right. Get it right, we must, if we seek to make progress. Let me end with this quote by Dr. Dr. Mourtada Deme, Project Director of the DGD Project of the UNDP:
“For ICT to play its role faithfully and effectively, all technology in elections should serve to enhance the critical principles of accuracy, accessibility, informed electorate, transparency and stakeholder buy-in.”_
That said, the failure of this will not make me give technology a bad name. No, technology, when correctly applied, can go a long way to make our elections more credible and transparent. INEC needs to learn from the flaws of the current system and start preparations immediately for 2019. If you want the best out of technology, then you must be prepared to get it right with the implementation. The case of the 2015 elections was that of good thinking, poor planning, and terrible execution.
Written by Engr. R. Tombari Sibe MNSE, a Computer Engineer and Policy Analyst.
@rsibe
