The decision of the Supreme Court (SC) on Tuesday, March 8, requiring the Commission on Elections (Comelec) to print voting receipts was unanticipated.
When it unanimously ordered the Comelec “to enable the vote verification feature of the vote counting machines,” the SC basically reversed its own rulings in 2010 and in 2013 that the precinct count optical scan machine (now VCMs) satisfied the Minimum Systems Capability (MCS) required by Republic Act Number 9369.
The March 8 decision was made by the SC without hearing Comelec’s side. The High Court perfunctorily denied the poll body’s motion for extension of time to submit comment. If this SC order stays or no compromises are forged – I will explain this later – the Philippines should not expect any election on May 9, 2016, at least as planned.
Despite my reservations on the ruling, I will no longer delve into its substance, but will only discuss the practical implications and legal ramifications of the ruling. The Supreme Court might have overlooked these considerations, to the peril of the election preparations.
At the outset, we need to make it clear that the VCMs that will be used in the May 9 elections have a built-in physical capability of issuing voting receipts, only that the Comelec deliberately disabled the feature for reasons I explained in my last article. (READ: EXPLAINER: Why it's alright not to have voting receipts)
This means that the Comelec can actually comply with the Supreme Court’s order with the same VCMs. The big “but,” however, is that there are only two months left until the scheduled elections, and such period is extremely short for the Comelec to actually comply with the order.
The primordial question that needs to be answered is: how deep will the modifications go to enable the VCMs to issue voting receipts? In inserting an additional command and new instructions into the VCM, will the software modification touch on the very source code of the VCMs?
RA 9369 defined “source code” as “human readable instructions that define what the computer equipment will do.” The Supreme Court in CENPEG vs Comelec (GR Number 189546, September 21, 2010) came out with a simple yet exhaustive explanation of the nature of source codes:
Source code is the human readable representation of the instructions that control the operation of a computer. Computers are composed of hardware (the physical devices themselves) and software (which controls the operation of the hardware). The software instructs the computer how to operate; without software, the computer is useless.… [T]he source code is the master blueprint that reveals and determines how the machine will behave.
Source code could be compared to a recipe: just as a cook follows the instructions in a recipe step-by-step, so a computer executes the sequence of instructions found in the software source code.…
The source code in voting machines is in some ways analogous to the procedures provided to election workers. Procedures are instructions that are provided to people; for instance, the procedures provided to poll workers list a sequence of steps that poll workers should follow to open the polls on election morning. Source code contains instructions, not for people, but for the computers running the election; for instance, the source code for a voting machine determines the steps the machine will take when the polls are opened on election morning.
All the 92,000 VCMs that will be employed in the coming elections were formatted using a single “source code,” which already passed all the required tests and evaluations, clearing the same from software errors and malicious codes. This source code contains all commands and instructions as to what the VCMs will do and not do on May 9. This includes the command of not issuing voting receipts.
To give an opposite instruction – for the VCMs to instead issue voting receipts – the very important consideration is whether this would be tantamount to modifying the very “source code” or the “recipe” of the VCMs as contemplated by RA 9369.
The Comelec has explained that the voting receipt “switch” feature of the VCMs’ software has only been “soft coded,” meaning it was left configurable or customizable at Comelec’s option, without having to edit the program’s source code. This is, however, something that can potentially become controversial: it will force the legal question as to whether the source code review as contemplated by RA 9369 is limited to only the hardcoded part. Or should the review rather cover all the final instructions that control the operation of the VCMs – whether softcoded or hardcoded – as explained in the CENPEG case?
In the latter case, this would entail the Comelec resubmitting the modified source code to a new array of tests required by RA 9369. The bad news is that the review by the international certification entity alone takes 4 to 6 months. The result then has to be re-evaluated by the Technical Evaluation Committee, as required by Section 11 of RA 8636, as amended by RA 9369. Then the modified source code also needs to be opened for review by political parties and candidates or their representatives, and by citizens’ arm or their representatives, pursuant to Section 12 of RA 9369.
I am certain that many people, especially Comelec’s staunch critics, will be very uncomfortable with the idea of the poll body inserting new commands into the VCMs’ source code at the last minute, without the modified version undergoing a new source code review and without it undergoing a new array of tests and evaluation.
Skipping all tests and letting the public use VCMs running on modified source code, untested and un-reviewed, might also undermine public confidence in the integrity and accuracy of the voting system.
Considering all these, here are Comelec’s very limited options:
- Proceed with the modifications of the VCM software, dispense the testing and reviews, and push for the May 9, 2016, elections as scheduled.
- Revert to manual elections to make it to the May 9, 2016, schedule of the elections and without contravening Supreme Court’s ruling. We should note that 2 months would also be inhumanly tight even for a manual election and this would also mean throwing away the billions already spent for the automation.
- Move the date of the elections to much later, to have time to comply with the technical changes and the required testing requirements. It must be noted, however, that postponing a national election would need a legislation, following Section 4 Article VII and Section 8 Article VI of the 1987 Constitution.
To my mind, however, the most workable scenario is out of Comelec’s hand. It is up to the Supreme Court to strike a win-win compromise.
While the prospect of the SC decision being reversed is not strong, considering the vote was unanimous, the Supreme Court can at least hold that the ruling would be prospective in application – it will apply to subsequent elections but not to the upcoming May 9, 2016, polls due to its proximity.
Prospective decisions are not alien to the Supreme Court. This approach would only be fair to Comelec, especially when we consider that the SC actually flip-flopped in this instance, perfunctorily abandoning its previous decisions affirming voting machines’ compliance with RA9369’s “Minimum Systems Capability,” including the “Voter Verified Paper Audit Trail” or VVPAT requirement.
In fact, this was the same approach implemented by the Indian Supreme Court in resolving its own VVPAT case in the case of Dr Subramanian Swamy vs Election Commission of India (Civil Appeal No. 9093 of 2013). Their High Court effectively permitting “gradual” compliance with its ruling on practical and pragmatic grounds.
The Philippines’ voting receipt case is yet another illustration of the long tension between the Supreme Court and the Comelec. The poll body may have been negligent in filing a comment with the SC, but I find the High Court equally careless in giving out a ruling without actually hearing the side of the government office that will administer the elections. It could have let the Comelec explain the dire consequences of tinkering with system this late.
While the Supreme Court is admittedly supreme in matters of law, it cannot however decide cases in a vacuum, totally unmindful of its consequences. Even if we have to concede that it has the brightest legal minds, there are matters of electoral technicalities and operations which its justices may simply know nothing about. In this case, it should have trodden that potentially intrusive path with prudence and circumspect, and with due deference to Comelec’s more superior expertise.
Having worked in Comelec, I know that they are in a very deep trouble right now. Not having an election this May 9, 2016, is not just a threat, but an eventuality that is clear and present. With that, I have only one thing to say: May God help us all! – Rappler.com
Emil Marañon III is an election lawyer who served as chief of staff of recently retired Comelec Chairman Sixto Brillantes Jr. He is currently studying Human Rights, Conflict and Justice at SOAS, University of London, as a Chevening scholar.