Can a machine be given the presumption of innocence? Should there be a “priority” rule as between certified election returns and certificates of canvass? These seem to be the main questions that should be addressed by the COMELEC when it promulgates the implementing rules and regulations of RA 9369. Such rules and regulations are mandated by Sec. 30 of RA 9369. It is suggested in this article that machines not be given any presumption of innocence, and that the certified election return be treated as “controlling” when certificates of canvass are questioned.
Are the safeguards enumerated in the automated elections law sufficient? Here, the answer is a qualified “Yes.” Provided that the implementing rules and regulations of RA 9369 are “good enough,” the legally-mandated safeguards appear to go a long way toward ensuring honest and credible elections.
The safeguards are of two kinds. The first ensures that the machine itself is honest in that the “certified election returns” (CERs) that emerge reflect the actual votes of the precincts. The second kind of safeguards ensures that the “consolidation” of election returns is accurate, i.e., not amenable to shaving or addition (“dagdag-bawas.”)
The key to honest automated elections remains the integrity of the “count” as made by the machine to be used. As envisaged in the contract between Comelec and the winning bidder (Smartmatic-TIM), the count is to be done “inside” the machine, which forces the voters and the citizenry to rely on safeguards “outside” the machine. Nonetheless, despite the black-box nature of the machines, their workings must effectively be made open to the public as this is required by state policy enunciated in Sec. 1 of RA 9369. There it is provided that as regards elections, the automated system should provide “that the process shall be transparent and credible and that the results shall be fast, accurate, and reflective of the genuine will of the people.” It is obvious that the law itself imposes an openness standard of transparency and credibility that the Comelec and Smartmatic-TIM must meet. Such openness can be achieved only if there is no presumption that the results of the machine are correct.
As regards the first kind of safeguards, one is a condition precedent. It should be implemented before the election itself. That would be the “pilot testing” required by RA 9369, Sec. 6. Absent such a pilot testing, the elections are, in my opinion, void, as they would have been conducted in violation of the law. Art. 6 of the Civil Code provides: “Rights may be waived, unless the waiver is contrary to law, public order, public policy, morals, or good customs, or prejudicial to a third person with a right recognized by law.” Thus, the right of citizens to a pilot test is not waivable. See my earlier post.
There are other safeguards of the first kind that must be met before Election Day. One is the so-called source code review. The law provides that “once [a] technology is selected for implementation, the Commission [on Elections] shall promptly make the source code … available and open to any interested political party or groups…” (Sec. 12, RA 9369). The same section of RA 9369 provides for pre-election examination and testing of equipment. However, the details of such “opening up” of the machines would only be known after the promulgation of the law’s implementing rules and regulations.
Another important safeguard is the so-called “random manual audit.” RA 9369, Sec. 24, provides that “… there shall be a random manual audit in one precinct per congressional district randomly chosen by the Commission in each province and city. Any difference between the automated and manual count will result in the determination of root cause and initiate a manual count for those precincts affected by the computer or procedural error.” Unfortunately, the law is silent on when and how the random audits will be undertaken, and on how any discrepancy can affect the officially-announced election results. In short, this gap in the legislation can be filled by a proper implementing rule; otherwise, the intent of the law would be undermined.
Safeguards against dagdag-bawas appear to be less problematic, though they still require vigilance by the voting public, the candidates themselves, and the citizens’ arms interested in ensuring an honest count. The most important safeguards here are: (a) the provision of certified election returns to the major citizens’ arms and other non-partisan groups or organizations (Sec. 19, RA 9369); (b) the rule allowing any person to capture an image of the data on the certified election returns and on the certificates of canvass (Secs. 19 and 21); and (c) the rule that allows any “recipient of the print or digital copies of the election return” to consolidate the data from the certified election returns and to announce it to the public (Sec. 33). Although the law limits the official copies of CERs to political parties, media, and certain citizens’ arms and certain groups, the law appears to allow these recipients to allow the public the opportunity to consolidate or aggregate the data. In short, an ordinary citizen who knows how to “capture” the data obtained by media and make his own aggregation can provide an independent check on the accuracy of certificates of canvass (COCs) in relation to CERs.
How can an aggrieved party challenge an erroneous COC, where “erroneous” means that it does not accurately aggregate the data in the election returns? It is suggested that the implementing IRR provide for a “priority rule,” whereby the data in the CERs are presumed correct for purposes of correcting COCs. Otherwise, the old system which potentially allows dagdag-bawas can still rear its ugly head.