This article first appeared on Dhivehi Sitee. Republished with permission.
Just before midnight on 7 October 2013, the Maldives Supreme Court issued a ruling cancelling the first round of the second democratic elections held on 7 September 2013.
Although the election was found to free and fair by over a thousand domestic and international observers, Jumhooree Party candidate Qasim Ibrahim filed a case at the apex court citing fraud and vote rigging during the election.
The 4:3 verdict cited a confidential police report submitted to the court allegedly claiming that 5600 votes were ineligible. The report has not previously been available to the public and was not shown to the Election Commission’s defence lawyers.
On the Supreme Court’s request, a team of ‘forensic experts’ from the Maldives Police Service worked from 26 September to 3 October inside the Supreme Court premises analysing ‘evidence’ of alleged fraud and vote rigging on 7 September.
It was cited in the Supreme Court ruling as the main basis on which the court ruled in Jumhooree Party’s favour. Dhivehi Sitee discovered a copy of the report online. The following is an English translation.
Maldives Police Service, Forensic Service Directorate, Male’, Republic of Maldives
Report No: K/JER/2013/0001
Report on discrepancies found in lists compared in the case submitted on vote rigging in the first round of the presidential election
This is a report on the findings of the investigation into the validity of evidence submitted against Elections Commission by Jumhooree Party in case No. 42/C-SC/2013 being heard at the Supreme Court, conducted upon a request made by the Supreme Court in its letter 197-C1/171/2013/30 dated 26 September 2013.
A large number of Forensic Document Examiners, Digital Examiners and support staff participated in the analysis, gathering and compilation of the information contained in this report. The work was conducted in the Supreme Court of the Maldives, in the presence of Court staff, between 26 September 2013 and 03 October 2013.
2. Material compared
2.1 List of 470 ballot boxes described as containing (in separate boxes) List of People Who Voted in the 2013 Presidential Election (7 September 2013): (Note: Although this is how the box files sent by the Elections Commission were marked, the separate booklets contained in the box files were labelled ‘List of Eligible Voters in 2013 Presidential Election’. From now on, all references to this list or to the booklets in this report will be as ‘List of Eligible Voters in the 2013 Presidential Election’. Thus, in total, the number of booklets or lists were contained in 44 box files, total 796 booklets.)
2.2 The following documents submitted to the Supreme Court by Jumhooree Party
2.2.1 Children under the age of 18 in the Eligible Voters List (41 names included);
2.2.2 List of dead people in the Eligible Voters list (list including 669 names);
2.2.3 Names of people repeated in the Eligible Voters List (list including 204 names);
2.2.4 List of people in the Eligible Voters List not included in the ID card registry of the Department of National Registration (List in which names of 1818 people are included);
2.2.5 List of people in the Eligible Voters List registered at addresses without knowledge of the home owner (Including information on 1187 people).
2.3 Personal information of citizens stored by the Department of National Registration (DNR Database);
2.4 Access logs of the computer software Ballot Progress Reporting System (or B.P.R.S) used in the Elections Commission on polling day;
2.5 Copy of the Eligible Voters List provided to the Jumhooree Party by the Elections Commission;
2.6 Eligible Voters List in the 2013 Presidential Election published in the Government Gazette.
3. Methodology of verifying validity of the lists
To establish the validity of the lists submitted to the Maldives Supreme Court by the Jumhooree Party, they had to be compared against the ‘List of Eligible Voters in the 2013 Presidential Election’ submitted to the court by the Elections Commission. The origins of the on-going case submitted to the Supreme Court by Jumhooree Party lie in doubts raised over the Voters Registry of the Elections Commission. Hence, assessing the validity of the Elections Commission’s ‘List of Eligible Voters in the 2013 Presidential Election’ was also seen as essential and efforts were made to do so. Thus, information contained in the ‘List of Eligible Voters in the 2013 Presidential Election’ was compared with personal information stored by the Department of National Registratin (DNR Databse) and discrepancies found are included in this report.
Further, the ‘Presidential Election 2013 – Eligible Voters List’ published in the Government Gazette Vol.42, No.94 of 30 May 2013; ‘Amendments made to the 2013 Presidential Election Eligible Voters List of 30 May 2013 following complaints received’ published on 29 June 2013; and the ‘List of Eligible Voters in the 2013 Presidential Election’ submitted to the Supreme Court by the Elections Commission were also compared and the findings are included.
To conduct this work, separate lists in the 470 boxes of ‘List of Eligible Voters in the 2013 Presidential Election’ were re-typed into computers and compared with the DNR database. As more than 1 (one) list was included in one ballot box, information in such lists were looked at and analysed individually.
Further, the lists referred to in 2.2 of this report were individually compared with the ‘List of Eligible Voters in the 2013 Presidential Election’ and the DNR Database.
Access logs of the computer software Ballot Progress Reporting System (or B.P.R.S) used by the Elections Commission on polling day were obtained. But, due to the manner in which the system was used with access allowed from many IP addresses and the short time available for research, adequate analysis was not carried out.
In examining the ‘List of Eligible Voters in the 2013 Presidential Election’ efforts were made to establish how [officials] verified the identity of which person from the ‘List of Eligible Voters in the 2013 Presidential Election’ was casting the vote. This was done with reference to Chapter 3 (Taking Votes) of the ‘Officials Handbook – Presidential Election 2013’ compiled by the Elections Commission.
Moreover, efforts were made to assess information around the lists that we believed must be looked at in conducting work of this nature. Points of note from these efforts are also included in this report.
3.1 To establish that persons named in the ‘List submitted to the Supreme Court by Jumhooree Party of 41 children who were under 18 and ineligible to vote but are said to have been allowed to vote’, referred to in 2.2 of this Report, are Maldivian citizens, the list was compared with the DNR Database. To this effect, as those not included in the DNR database are people who have not obtained a national identity card, they have been regarded as people who not eligible to vote. If the people in this List were found in the DNR database, their ages were verified and it was established whether or not they were 18 years of age by 06 September 2013 and therefore eligible to vote.
3.2 Information relating to the 669 people who are said to be dead were compared, as stated in the previous point, to the DNR Database. In addition, it was checked whether those people from the 669 whose information was found to be in the DNR database were included in the Eligible Voters List said to have been used in the polling stations of the presidential election held on 7 September 2013. It was also checked whether such an individual had voted in the presidential election on 7 September 2013.
3.3 The list submitted by the Jumhooree Party of 1187 people from the Eligible Voters List in the Presidential Election on 7 September 2013 said to have been registered at a different ballot box without the knowledge of the voter cannot be verified through data comparison work such as this. Therefore, this work could not get to the bottom of it. But, we state that the issue must be addressed in concluding this case and propose that it be verified by the relevant complaints (elections) office or committee records.
3.4 Information relating to the 102 people noted by Jumhooree Party as repeated in the lists of eligible voters in the presidential election of 7 September 2013 were compared with the ‘List of Eligible Voters in the 2013 Presidential Election’.
3.5 Checked whether any individual’s name was repeated in the ‘List of Eligible Voters in the 2013 Presidential Election’. It was also checked whether such a person voted more than once.
3.6 To verify Jumhooree Party’s claim that 1818 people described by Jumhooree Party were entered into the List of Eligible Voters in the 2013 Presidential Election without valid information, their details were checked against data comparison [sic], DNR database and the ‘List of Eligible Voters in the 2013 Presidential Election’.
3.7 Checked whether any person was included in the lists of eligible voters used in the polling stations in the presidential election of 7 September 2013 that was not in the DNR Database or had invalid information. Clerical evidence available was used to verify whether or not any such individual voted in the presidential election of 7 September 2013.
4.1 Following are the findings of note resulting from examination of the lists submitted as evidence by Jumhooree Party
4.1.1 The ‘List of children under the age of 18 included in the Eligible Voters List’ submitted by the Jumhooree Party did not give the identity card number of the 41 children named in the list. When this list was compared with the DNR Database, 32 of them were found to have been under the age of 18 by 7 September 2013. The remaining 9 people’s information could not be verified because the list did not contain identity cards. That is, their names could not be found by the work conducted using their permanent addresses, names and ballot boxes as the basis. List of People Who Voted in the Presidential Election of 7 September 2013 notes that 12 of those 32 children voted. Information relating to the 32 persons identified are listed in Annex 1 of this report.
4.1.2 When the ‘List of Dead People Included in the Eligible Voters List’ containing names of 669 people submitted as evidence by the Jumhooree Party was compared with the ‘List of Eligible Voters in the 2013 Presidential Election’, information relating to 637 people were found. 14 of them were noted in the List of People Who Voted in the presidential election 2013 (7 September 2013). 2 of the 14 people were recorded as having voted using an identity card that was no issued in their names. Moreover, when the Elections Commission official noted the identity card of one of the two people using a pen, one digit from the identity card number was omitted. DNR database shows that 156 people recorded as deceased died on 01 January 1800. This information is in Annex 2.
4.1.3 When the names of 204 people submitted by the Jumhooree Party as list of people repeated in the Eligible Voters List was compared with the ‘List of Eligible Voters in the 2013 Presidential Election’ 174 matching entries were found. This includes 22 people the DNR Database had noted as repeated and entered “REPEAT / REPEAT.REPEAT” as their permanent address. This information is in Annex 3.
4.1.4 When the list submitted by Jumhooree Party as containing information of 1818 people who are not in the Department of National Registration’s ID Card Registry were compared with the DNR Database it showed that DNR has not issued 1637 of these people with ID cards. Efforts to find the remaining people by matching other information were unsuccessful. This also includes 7 people referred to in Point 4.22. Records show that of the people identified, 207 did vote. 96 of them voted using identity cards that were not the same as listed in the gazetted list. Information relating to this is in Annex 4.
4.1.5 When the information relating to 1187 people listed as registered to certain addresses in the Eligible Voters List without the knowledge of the homeowners was compared with the ‘List of Eligible Voters in the 2013 Presidential Election’ and the DNR database, 1186 records were found. 44 of these people were registered some place other than their home islands. While these 44 people voted, the Registry shows 1115 people in the list voted. It is believed that more information about this list cannot be found by data comparison. Information relating to people with such discrepancies is included in Annex 5.
4.2 It is noted that the ID card numbers of some people in the ‘List of Eligible Voters in the 2013 Presidential Election’ did not match with some of the ID card numbers written in pen on the list as voted. In this regard, 815 individuals had identity card numbers that did not match. Information relating to such discrepancies are given in Annex 6.
4.3 In instances where people’s information contained in the ‘List of Eligible Voters in the 2013 Presidential Election’ did not match their information in the DNT Database, it must be noted that, they were given the opportunity to vote by changing such information with a pen to match that contained in the DNR Database. Information relating to such individuals are contained in Annex 7.
4.4 In sorting the names in the two lists of Ballot Box Number T03.1.4 different methods were used. While one of the lists was sorted according to the addresses on the identity cards the other was sorted according to the order of identity card numbers. Therefore, the names on the two lists were topsy-turvy and the two lists had lost their order.
4.5 Some people’s names were added to the ‘List of People who Voted in the Presidential Election 2013 (7 September 2013)’ with a pen and given the opportunity to vote. It can be noted from the lists that 07 people were given such an opportunity. Information relating to such individuals is contained in Annex 8.
4.6 It was noted that the list used to highlight people who voted in Ballot Box No. Z33.1.1 installed on “Vivanta by Thaj Coral Reef Maldives” was similar to the list gazetted by the Elections Commission. Therefore, the list did not contain important voter information such as Identity Card and Date of Birth. This is a list different from the ‘List of Eligible Voters in the 2013 Presidential Election’ and similar to the list gazetted by the Elections Commission.
4.7 It was noted that the methods used to note those who voted on the Eligible Voters List varied at different polling stations. The following are methods used to note people who voted:
4.7.1 Using highlight markers to note who voted;
4.7.2 A tick against the name of the person who voted in pencil or pen instead of using a highlight marker;
4.7.3 Writing the voter’s identity card number with pen or pencil instead of using a highlight marker;
4.7.4 Putting a cross (x) beside the name with a pen or pencil after using a highlight marker;
4.7.5 Using a highlight marker to denote a person having voted and not at all entering their identity card numbers;
4.7.6 Using a highlight marker to denote that a person had voted, then using a different coloured highlight marker to denote the person had not voted; and
4.7.7 Having noted a person as voted by writing the person’s identity card number in pen, then crossing it out with a pen, then connecting it to an identity card number against someone else’s name with a line.
“Officials Handbook – Presidential Election 2013, Chapter 3 (Taking Votes) states that it should be done thus: ‘After checking the voter’s name in the list, note the voter’s identity card number in the allocated box and, once the person who issues the ballot paper has done so, note the person’s name in the list with a highlighter marker.’
4.8 It was noted that while two or three lists were used with the ballot boxes at some polling stations, others used only one list.
4.9 While it was noted that 690 people recorded as dead on the DNR Database was included in the ‘List of Eligible Voters in the 2013 Presidential Election’, records show that 18 people recorded as dead in the DNR database voted. Information relating to these 18 people are on Annex 9.
4.10 It was noted that the ‘List of Eligible Voters in the 2013 Presidential Election’ included information related to minors under the age of 18. 39 such children were included in the list. 07 of them voted after changing their age. It is evident from the Identity Card numbers entered into the ‘List of Eligible Voters in the 2013 Presidential Election’ with a pen that 4 out of these 7 children voted using another Identity Card belonging to someone over the age of 18. One such person voted at a ballot box abroad, in Malaysia. Information related to persons noted in this point are included in Annex 10.
4.11 It was noted that votes were cast using the same identity card (repeatedly) in the lists of eligible voters in 07 September 2013 presidential election. It was noted that three identity cards (ID Card Number: A047595, A100910, A263120) were used to vote repeatedly.
4.12 ‘List of Eligible Voters in the 2013 Presidential Election’ repeated 07 different identity cards. Information relating to such persons is in Annex 11.
4.13 Of the people included in the ‘List of Eligible Voters in the 2013 Presidential Election’ 2273 people had Dates of Birth different from that listed in the DNR Database. Information relating to people in the eligible voters lists whose dates of birth were listed differently in the DNR Database is included in Annex 12.
4.14 There were 1804 people in the ‘List of Eligible Voters in the 2013 Presidential Election’ whose addresses were typed in the DNR Database as “REPEAT / REPEAT.REPEAT”. The registry of people who voted in the presidential election show that 225 of these people voted in the 07 September 2013 election using this identity. Detailed information relating to such persons is included in Annex 13.
4.15 Of the people included in the ‘List of Eligible Voters in the 2013 Presidential Election’ 4032 were found to have permanent addresses different from their addresses listed in the DNR Database. 2830 of these people voted. People whose permanent addresses listed in the ‘List of Eligible Voters in the 2013 Presidential Election’ were different from that listed in the DNR database are included in Annex 14.
4.16 Of the people included in the ‘List of Eligible Voters in the 2013 Presidential Election’, 319 people’s gender was listed as different from that listed in the DNR Database. Information relating to these people is in Annex 15.
4.17 Of the people included in the ‘List of Eligible Voters in the 2013 Presidential Election’, 1164 individuals were found to have names different from the DNR Database. Registry of Voters shows that 952 such people voted on 7 September 2013. Information relating to people whose names were different is in Annex 16.
4.18 It was noted that carelessness when writing in identity card numbers of those who voted from ‘List of Eligible Voters in the 2013 Presidential Election’, meant some digits were missing or illegible. The information of 162 people were noted this way. Moreover there were 815 people in the registry of voters whose ID card numbers written in pen by the polling stations did not match the ID card numbers printed in the ‘List of Eligible Voters in the 2013 Presidential Election’. This figure includes the 162 Identity Card numbers previously mentioned. This information can be seen in Annex 07 of this report.
4.19 Of people in the ‘List of Eligible Voters in the 2013 Presidential Election’ 07 people were completely missing from the DNR Database. Of those 07 people, one person voted in the 07 September 2013 presidential election. As this person has a passport, it is believed that the person voted using the passport. Nevertheless, no information relating to this person is in the DNR database. Information relating to the 7 people is included in Annex 17.
4.20 It was noted that there were differences between the ‘List of Eligible Voters in the 2013 Presidential Election’ published in the government gazette and list given to Jumhooree Party candidate by the Elections Commission. As such, while the total number in the gazetted amended Voters registry (on the Elections Commission website) was 239956, the total number of eligible voters as per B.P.R.S was 239593. The list Elections Commission gave the Jumhooree Party contained the number 239593. The voters registry sent to the Supreme Court by the Elections Commission (on a CD) contained information relating to 239971 individuals. However, the booklets (used in the polling stations, printed on paper) containing ‘List of Eligible Voters in the 2013 Presidential Election’ that the Elections Commission submitted to the Supreme Court to show the registry of people who voted contained a total of 239603.
4.21 Discrepancies were noted in results from ballot boxes. As such, there was a difference of 251 votes between the total number given in the ‘List of People Who Voted in the 2013 Presidential Election’ and results announced by the Elections Commission on its website and other forums. Total number of votes from 66 boxes was less than the number published by the Elections Commission. While 284 boxes had no problems, 120 boxes had more voters than announced by the Elections Commission. While the number which voted less is 123, the number of votes cast more than the total at the boxes was 254. Annex 18 contains a list which shows the discrepancies in the vote boxes.
5.1 Of the issues noted in No.4 of this report, records show 773 people were allowed to vote despite discrepancies in identity card numbers, 7 people whose names were not on the list were added by pen, 18 people voted who were listed as dead in the DNR registry, according to the Registry 07 children voted, 3 people voted repeatedly, 225 people voted who were marked as ‘repeat’ in the DNR and were not issued with identity cards, 2830 people were given the opportunity to vote even though their permanent addresses did not match, 952 people voted despite discrepancies in their names, 7 people voted whose records were non-existent in DNR, and 819 people were included whose identity card numbers in the printed Voter Registry did not match the identity card numbers entered with insufficient care by Elections Commission officials. This is a total of 5623 votes. In relation to the announced results of this election, we see this as a massive figure.
5.2 While officials in the presidential election 2013 were given training, a particular procedure was set for them to follow, and an ‘Officials’ Handbook’ was printed and shared with them, it was noted that there was no consistency in how voters’ identity card numbers were entered and in drawing the required line with highlighter markers to denote people who voted. This created additional difficulties in entering this information into the computer.
5.3 We did not receive enough technical information necessary for analysing the Ballot Progress Reporting System. Therefore, we cannot explain the system’s process and data flow mechanisms and did not have the information to assess the importance of using such a system or to identify the vulnerabilities of the system if any. When we asked for the system’s server access log and audit trail, we only got the access log. Adequate information in this kind of work cannot be gathered this way. However, as access log shows that the software has been accessed by several IPs, it is our view that these access points be identified and their legitimacy established. From analysing the access logs, it can be seen that the Ballot Progress Reporting System is a server hosted in a way that allows access to it by the general public. That is, the server had been accessed by a large number of domestic and foreign IP addresses. As such, records show that it had also been accessed several times by IPs in countries that did not host ballot boxes on that day. The IP counts of the access log also shows up information that makes us believe that some Nets used at polling stations were also connecting to the server. It is also our opinion that additional people (third-party), too, can enter the server. As the Elections Commission did not provide the audit trail of the Ballot Progress Reporting System, we could not do the work to answer complaints raised by the public about its stoppage at various times on polling day.
5.4 As the database of the Ballot Progress Reporting System could not be obtained, we could not further verify the ‘List of Eligible Voters in the 2013 Presidential Election’ submitted to the Supreme Court by the Elections Commission.
5.5 As the clerical evidence obtained cannot get to the reality of the claims that votes had already been cast in their names when some people arrived at polling stations, and because there is no time to carry out such work, and as we do not have information on complaints filed at complaints bureaus, it was not possible for this report to get to the origins of those suspicions. For the same reasons the validity of the ‘List of 1187 people included in the Eligible Voters List who were registered at addresses without knowledge of homeowners’ could not be adequately verified.
5.6 Although the figure shown in point (5.1) is not a large percentage of the eligible voters, there is sufficient room to believe that it can affect the second round of the election. But, it must also be said that in relation to the error-margin this is not that big a number. When considering what has happened, it has to be said that there is vast opportunity to say that these things were done intentionally rather than unknowingly. That is, things like the inclusion of 1804 people in the Voter Registry to whom the Department of National Registration had not issued identity cards, a large number of dead people (690) being included in the Eligible Voters Registry and some of them casting votes, 39 children who were not 18 years of age according to the Roman calendar and 07 of them casting votes, 2830 people whose permanent addresses did not match being included in the Voters Registry, including in the Eligible Voters Registry 1164 people whose names contained huge discrepancies are issues which could have been, but were not, easily clarified from the Department of National Registration. For these reasons, it cannot be believed that this was an election in which work of the Elections Commission staff was sufficiently and cleverly supervised.
5.7 Since the issues listed above creates the room in which some people can see them as actions carried out for the benefit of a particular candidate, it is our view that ensuring that such matters are not repeated in future elections is vital for maintaining people’s trust in the young Maldivian democracy.
5.8 There is a difference of 251 people between the election results announced by the Elections Commission and the ‘List of Eligible Voters in the 2013 Presidential Election’. Although we cannot say directly how this difference has emerged, this will also be added to the Registry of people who voted. But, as all officials did not mark the voters registry as outlined in the Officials’ Handbook, and did not act with consistency or use the same method in noting names of people on the list who turned up to vote [sic]. And, it could also have happened because people who were not on the registry were also given the opportunity to vote this time, differences existed in the various copies used at even the one ballot box, or some other reason that has not been noted in this work. Moreover, judging from other problems encountered in the Registry, it is our view that it is also possible that people could have cast extra votes.
This proposal was researched by a team of people with knowledge in various fields. Thus, document examiners, computer forensic analysts and technical staff participated.
03 October 2013
Assistant Commissioner of Police Hussein Adam