1. Security

   Despite the large number of voters of different national communities who, for the first time since war broke out in Mostar in May 1993, crossed the Neretva River into the predominantly Croat and Bosniak sides of East and West Mostar in order to vote in their former neighbourhoods, there were no significant security problems. However, a few minor incidents were reported. In one incident, a Croat attempted to drive his car with a large Croatian flag, at high speed, across the river to East Mostar, but was stopped by an IFOR patrol and released soon after. In a second incident, a fight broke out in a polling station in West Mostar between a member of the polling committee and an SDA representative observing the proceedings. In another incident, a Serb refugee arriving from Belgrade was beaten when he stepped out of the bus in East Mostar. None of these minor incidents had any impact on the outcome of the election proceedings.

   The secure environment in Mostar on June 30th may be due primarily to the vigilant and visible presence of more than 2,500 IFOR troops in armoured personnel carriers and other military hardware, deployed throughout the city. In addition to the roving patrols, IFOR was present near most of the 97 polling stations. Furthermore, IFOR checkpoints were set up on the roads leading to Mostar.

   IPTF and WEU Police combined forces to provide international police presence at every polling station. Local police were also deployed at every polling station.

   These security arrangements, combined with more than 100 international civilian monitors stationed at every polling station and patrolling the city to observe possible elections related violations, may have had a crucial role in deterring any disturbances and maintaining the high degree of calm which prevailed throughout the day. Thugs, hooligans, and other criminal elements had little if any opportunity to disturb the calm.

2. Freedom of Movement

   Some 7,000 Mostar residents crossed the river to vote in their former neighbourhood polling stations. This figure was estimated based on the distribution of national communities throughout Mostar during the 1991 census and the vote patterns on election day. In addition, according to IFOR some 18,000 displaced persons from Mostar returned to the city on the day of election or the preceding days for a brief visit in order to cast their ballots. Among those were some 200 Bosniak refugees who travelled in four bussed from Vienna and cast their ballots in predominantly Croat West Mostar without any incidents; some 150 Serb refugees travelled in buses from Belgrade; and an unspecified number of Croat refugees travelled from Croatia. While some of the vehicles, especially the buses, bringing Mostar refugees to the city were provided IFOR and IPTF escorts, others arrived unescorted, and no significant security problems were reported to the City Commission. On June 30th in Mostar, many psychological barriers were overcome and the elections served as an impetus for Mostar residents to cross the river and intermix, meet their former neighbours, and dismiss some of their abstract fears. While many who crossed the river were welcomed warmly by their former neighbours and tolerated by displaced persons who have settled in their former neighbourhoods, there were also incidents of harassment and intimidation, but none so serious as to prevent the voters from casting their ballots and returning to their side of the river in safety.

3. Voters Omitted from Electoral Lists

   According to the EUAM Decree on the conduct of elections in Mostar, Article 4(4), voters had until June 15 (15 days prior to the date of elections) to petition the CEC for corrections of the electoral lists. In Mostar, the deadline was observed. However, in the four polling stations abroad (Germany, Switzerland, Norway, and Sweden), the electoral lists were made available to voters for inspection on 25 June, 10 days after the deadline and five days before the elections day. Refugees in those countries and others where no polling stations were provided were denied an adequate opportunity to inspect the electoral list and petition for corrections.

   As a result, some citizens of Mostar were denied the right to vote. In the 20 polling stations in Mostar monitored by ICG, the percentage of registered voters who were disenfranchised averaged 2.8%. An overall percentage of voters denied the right to vote throughout the city is not available. In some cases, names appeared on the voter register, but were omitted from the polling station excerpts. In other cases, names were altogether omitted from the voter register. Polling Committees' reaction to such cases were not uniform. Citizens were directed to the CEC, MEC, and EUAM to complain, or were merely dismissed, resulting in frustration, and in some cases angry exchanges with officials.

   In the case of citizens whose names appeared on the voter register but were omitted from the polling station excerpts, the CEC issued a directive to the MECs instructing them to provide those citizens with a document certifying that their name appeared on the voter register. The document then was used by those citizens to cast their ballots in the appropriate polling stations. However, the CEC reached this decision at about 10:00 on June 30th, and some citizens who had tried to vote earlier never returned to cast their ballots.

   In the case of citizens whose names were not on the voter register, they were provided a document by the MEC, certifying that they had attempted to vote but that their names were not on the voter register. However, the decision to provide such a document was reached by the CEC at 16:15 on June 30th based on an appeal from the EUAM Ombudsman. Consequently, the number of such documents issued to citizens cannot be considered a reliable basis to determine the total number of citizens of Mostar who were denied the right to vote.

   In Germany, 4,085 refugees from Mostar cast their ballots and 421 were denied the right to vote. In Sweden, only 30 could not vote as opposed to 1,356 who cast their ballots. Data on those who were denied the right to vote in Switzerland and Norway were not available as of this writing, but 762 voted in Switzerland and 1,223 in Norway. On the eve of the elections, some 1,000 refugees who could not travel to the four polling stations to inspect the voter register had appealed to the Ombudsman to verify on their behalf the voter register in Mostar and insure that their names were properly registered. However, the request could not be accommodated in the short time remaining before election day.

4. Balloting Process
   1. Ballot boxes

      The ballot boxes for the elections in Mostar were provided by the EUAM. The boxes were made of cardboard, were assembled at the polling stations, and then secured by adhesive tape. Each polling station had three clearly marked ballot boxes in which the white, yellow and pink ballots were to be cast. However, the ballot boxes were not properly sealed with a tamper proof stamp. While the presence of international monitors and list representatives at every polling station may have prevented any tampering with the ballot boxes and the CEC received no complaints of such incidents, nonetheless the possibility existed that, during brief absences of monitors from the polling stations, the ballot boxes could have been tampered with and the adhesive tape replaced.

   2. Ballot secrecy

      Article 59 of the 1990 Law On Election of the Board Members and Delegates for Assemblies of Social-Political Communities of BiH (Law on Elections) requires that a special place be provided in each polling station for voters to complete their ballots in secrecy. Only one of the 20 polling stations monitored by the ICG team reported that this requirement of the law was violated - in polling station number 18 in West Mostar, a screened area was not provided. In another polling station where a screened area was not provided, the Polling Committee members themselves improvised and built three screened booths. The CEC did not receive any other complaints in this regard.

      The secrecy of the ballot was also violated, albeit innocently, when in some cases more than one voter, often husband and wife, were allowed to approach the screened area to mark their ballots.

   3. "Agitation" on Election Day

      Article 57 of the Law on Elections prohibits any political "agitation" on the day of elections and the 24 hours preceding the date. While candidates seemed to have respected this requirement of the law, there were many, what seemed to be spontaneous, displays of emotional symbols which could be considered "agitation," including vehicles flying large Croatian flags speeding around West Mostar, and caf�s in the immediate vicinity of polling stations in West Mostar playing belligerent marches and songs at extremely high volume. Such "agitation" may have discouraged some from crossing the river to cast their ballots.

   4. Assistance to Voters

      In general, Polling Station Committee members and their deputies were perceived to assist voters in a neutral and professional manner. However, in a small number of polling stations, party or coalition list representatives were observed providing instructions to voters and attempting to influence their choice, instead of staying within their mandated role and observing the proceedings.

   5. Other Irregularities

   In one case, an international monitor noted that an individual attempted to vote on behalf of a family member who could not be present. Despite the monitor's reminder that the law did not permit such voting, the Polling Committee Chairperson allowed the individual to proceed and vote for the family member, acknowledging at the same time that she was making an exception.

   Despite the prohibition found in Article 74, paragraph 5, of the Law on Elections that no one should enter a polling station with guns or other dangerous weapons, in many polling stations, police were observed entering the premises to cast their ballots while still wearing their side-arms. While such incidents were most likely innocent, nonetheless the law requires that when armed persons enter a polling station, the proceedings must stop until they are escorted out. This provision was not respected.

   With the large number of people in Mostar injured during the war, many physically disabled voters were discouraged from voting because of the inaccessibility of most polling stations to the physically disabled. Nonetheless, some disabled voters were assisted by friends or family members and with great difficulty did cast their ballots. The elderly voters also experienced similar difficulties.

   In some locations, more than one polling stations were placed too close to each other and in the same room, resulting in confusion and overcrowding. As a result, some voters were observed casting their ballots in the wrong ballot boxes, which could have resulted in the annulment of the particular polling station results.

   The Law on elections, Article 76, allows representatives of party and coalition lists to enter in the official minutes of the polling station their comments or objections regarding any incident or decision of the polling committee. In some cases, the Polling Station Committee members or the Chairperson denied this right.

   In some polling stations, voters used pencils to mark the ballots. Indelible ink pens would have been the best choice to mark the ballots in order to prevent any possibility of tampering with the marked ballots, however even simple pens would have been a better choice than pencils.

   However, none of the minor irregularities described above were significant enough to impact the outcome of the elections.

5. Results in Polling Station in Germany
   1. Discrepancy

      According to the voter register, 4,085 refugees from Mostar voted in Germany. However, when the ballots in the boxes were counted, it was discovered that the total number of ballots cast were 26 more than the number of voters according to the register.

      After the second count in Mostar confirmed that the ballots received from Germany included the additional ballots, with a four to three decision and based on Article 77 of the Law on Elections, the CEC decided on 3 July to annul the voting results from Germany. After a complaint was lodged against the CEC ruling, the issue was forwarded to the EUAM Ombudsman for a final determination.

   2. The Law

      The Law on Elections, Article 77(5), clearly states that "If it is established that the number of ballots in a ballot box is higher than the number of voters who cast their vote, the polling committee is dismissed and a new one is appointed, and the voting at that polling station shall be repeated. The result of voting at that polling station is ascertained after the repeated voting." (Emphasis added). According to Article 97 of the same law, the vote in the polling station at issue should be repeated within 30 days. In addition, Article 95 of the same law requires that, until the repeat voting takes place, the electoral results for the entire unit, in this case Mostar, cannot be finalised.

   3. Decision of the Ombudsman

      On 6 June, the Ombudsman decided to reverse the City Commission's ruling to annul the election results from Germany and reinstated the results. The ombudsman's decision included the following findings: "the irregularity should be attributed to material errors committed in the procedure by the Polling Committee rather than to fraudulent intentions"; "voters abroad did not vote in what is to be considered a polling station within the meaning of the 1990 Law" on Elections; and since "Article 77(5) makes reference to irregularities in the results of the counting for individual polling stations" (italics in the decision), therefore "It could be said that, voting abroad for their respective ward (i.e. City-Municipalities), the citizens cast their votes as they would have done in the polling station where they would have been registered, had they been in Mostar at the time. But in practice, it is obviously impossible to determine the exact influence exercised by the additional votes cast in Bonn on the results in each polling station within the Municipality concerned. Therefore, in order to obtain a correct application of Article 77(5), some of the 26 ballot-papers in excess of the number of voters should be accounted for in the returns of the voting of all the polling stations within those Municipalities. This would impair the validity of the elections in each of the three Municipal Councils concerned (North, South-West and West), as well as regarding the four municipal representatives to the City Council. Considering that one additional ballot paper for the City-wide list was also found, one could reach the absurd conclusion of having to repeat the entire electoral procedure of June 30th not only in the said Municipalities, but in the entire Mostar, and just for the sake of one single additional vote."

      The Ombudsman's decision also refers to Articles 126 and 128(2) of the Law on Elections and concludes that "the decision of the invalidation [sic] the polling should take into account if the irregularities have a significant influence on the legality of the elections and on the elections' results. This clearly did not happen in the present case...."

   4. Decision Based on Incorrect Interpretation of the Law

      As discussed above, the Ombudsman has correctly concluded that the discrepancy should be attributed to error rather than to fraudulent intention. The vote in Germany (Bonn) was supervised by a special polling committee of 30 members, 15 Bosniaks and 15 Croats. In addition, an EUAM legal advisor, party or coalition representatives, and an international monitor observed the proceedings. The latter has submitted a report to the EUAM Ombudsman. Under the circumstances, it is unlikely that fraud could have been committed. The discrepancy may be due to errors committed while processing a large number of voters in one polling station, compounded by the difficulties inherent in checking voters' names in a voluminous voter register which included all registered voters for the Mostar elections, unlike the excerpts available for each polling station situated within Mostar.

      However, the remaining conclusions of the Ombudsman are based on an incorrect interpretation of the law. The Ombudsman concluded that voters abroad, including in Germany, did not vote in what can be considered a "polling station" within the meaning of the Law on Elections, but "voting locations" defined by the Agreement of May 25th, 1996 postponing the elections in Mostar, initially set for May 31st. The final version of the Decree based on the Agreement of May 25th, uses the term "voting locations abroad" once in Article 15, but reverts back to the term "polling station" for refugees abroad in the next Article (16), without altering its meaning in any way. The two terms are used interchangeably.

      Article 59 of the Law on Elections states that "Polling stations are determined regarding the number of voters, i.e. distance, and in a way that the number of voters is appropriate so the voting can be carried out with no difficulties at the time determined for voting." Article 58 of the same law further states that "Polling stations are determined by the municipal election commission.... Each polling station is marked with a number..." The polling stations abroad were set up precisely with due regard to the number of voters and so that the voting could be carried out without undue difficulty to refugee voters. Nothing in Articles 58 and 59 exclude the polling stations abroad from the definition.

      The Ombudsman further argues that those polling stations abroad were the result of an exceptional arrangement and were, according to Article 16 of the Decree, excluded "from the definition of the electoral wards." Article 16 of the Decree states, "Instituting the polling stations abroad does not affect the definition of the electoral wards ... from Article 3" (emphasis added) - Article 3 describes the manner in which Councillors are to be elected to the City and six Municipal Councils. Clearly, Article 16 refers to "polling stations" and not "voting locations." Furthermore, the Article 16 clause referenced in the Ombudsman's Decision does not change the character of the polling station, nor does it exclude the polling stations abroad from the definition of electoral wards as stated by the Ombudsman, but merely does not affect the definition of the wards so far as the manner in which Councillors are elected to the City Council and the six Municipal Councils.

      The Ombudsman states that "Article 77(5) makes reference to irregularities in the results of the counting for individual polling stations. But voters abroad did not vote in what is to be considered a polling station within the meaning of the 1990 Law." (italics in the original Decision). The Ombudsman's argument in the second sentence quoted has already been discussed above and determined that the polling stations abroad were within the meaning of the 1990 Law on Elections. Therefore, the provision of Article 77(5) applies squarely. However, the Ombudsman argues unconvincingly that, since it is impossible to determine the "exact influence" the additional ballots in Germany have on each polling station within the municipalities concerned, and since Articles 126 and 128(2) of the Law on Elections require a finding of "significant influence", therefore the discrepancy should be ignored and the results of the voting in Germany reinstated.

      Article 77, however, does not require a finding of substantial or significant discrepancy for the results of a polling station to be annulled. The provision of the law comes into effect when there are a greater number of ballots in the box than the number of voters who cast their ballots according to the register. There is no requirement for a substantially or significantly greater number of ballots, even one extra ballot will suffice. The provision of the law is not ambivalent.

      The Ombudsman's reliance on Articles 126 and 128 of the Law on Elections is not warranted. Article 126 states, "If the competent electoral commission establishes, on the basis of the electoral regulations, irregularities in the electoral process in certain polling stations or on the whole territory of a municipality substantially affect the results of the elections, voting shall be annulled in such polling stations or on the whole of the territory of the municipality, and repeated voting shall be organised." Article 128 restates the same in the case of a complaint filed.

      The Law on Elections prescribes a series of provisions for the conduct of elections, including rights of voters, rights of candidates, polling station material, secrecy of the ballot, the safe environment for voters, and more. With the exception of Article 77, none of these provisions contain prescribed remedies or language requiring that the election results shall be annulled if it is determined that the slightest infraction has occurred. Not even other provisions of the law regarding the counting of the votes include a similar remedy. Articles 126 and 128 address violations of provisions in the Law on Elections for which a remedy is not defined, and is applicable if the violations "substantially affect the results of the elections." However, the violation described in Article 77 was determined to be so egregious as to warrant the annulment of the results even if the vote count was inconsistent by a single vote.

      Since Articles 126 and 128 do not apply to the violation defined in Article 77, then the spectre raised in the Ombudsman's Decision of an "absurd conclusion of having to repeat the entire electoral procedure ... in the entire Mostar" can be dismissed, and the election should have been repeated only for the polling station in Germany. As such, the City Election Commission's decision to annul the voting results from Germany according to Article 77(5) should have been confirmed by the Ombudsman.

   5. Functioning of the Commissions

      In general, the Polling Station Committees, the MECs and the CEC performed their duties professionally and with diligence. However, in some polling stations the Polling Station Committee members could have organised the proceedings more efficiently in order to avoid unnecessary crowding and long delays. There were also misunderstandings on the part of some Polling Station Committee members regarding the role of international monitors, resulting in less than cordial interaction.

      Considerable problems were experienced when a confusion developed regarding the procedure to be followed after the ballots were counted at the polling stations. The correct procedure was as follows: once the ballots were counted and tabulated at the polling stations and the minutes prepared, the results were to be transported to the respective MEC by the Polling Station Committee members and escorted by the international monitors present, party and coalition list representatives, and police; the MEC then was to tabulate the results pertaining to the particular municipality and forward the two results to the CEC for the final tabulation of city-wide election results. However, the EUAM Basic Information for Observers contained an error and instructed international monitors that the results were to be transported from the polling stations directly to the CEC. This error resulted in polling station committees and international monitors receiving contradictory instructions throughout the day. The error was corrected at around 17:30, just before the polling stations closed.

      Some of these problems could have been addressed more effectively had EUAM provided for a central focal point for the organisation of the monitoring effort. The focal point could have included representatives from EUAM and each of the organisations providing monitors, and could have responded to developing problems in a concerted, co-ordinated, and timely manner.

The June 30th elections in Mostar marked a turning point for freedom of movement, albeit under the watchful eye of some 2,500 IFOR troops and an additional number of international police. A large number of Mostarians were able, for the first time since war broke out in the city in 1993, to cross the Neretva River and visit the segregated neighbourhoods where they once lived. The voting proceeded without any serious security incidents. This secure environment for voters could not have been achieved without the visible presence of IFOR and international police. The presence of international monitors in all polling stations and throughout the city to observe the freedom of movement was a significant contributing factor for the elections to proceed without major violations or irregularities being reported to the City Election Commission.

The most significant problem encountered was the difficulty with the missing names from the voter register. As a result, a number of Mostarians were denied the right to vote. In the 20 polling stations monitored by ICG out of a total of 97 stations, 2.8% of registered voters were disenfranchised as a result of such problems.

There were also some problems with the facilities provided for the balloting - ballot boxes were not sealed properly to prevent possible tampering, the secrecy of the ballot was not always respected, in some polling stations ballots were marked with a pencil instead of a pen, there was some political "agitation" outside polling stations, assistance provided to voters by the polling station commissions was not always perceived as neutral, party or coalition list representatives were observed providing instructions to voters instead of monitoring the proceedings, and other minor violations which were not significant enough to substantially impact the outcome.

The major problem occurred in Germany, one of the four polling stations provided for refugees, where more ballots were found in the ballot boxes than voters registered to have cast their ballots. The Bosnian law in such a case is clear - the results of voting in the polling station at issue would have to be annulled and the vote repeated within 30 days. In accordance with this provision, the City Election Commission annulled the results of the vote from Germany. However, the EUAM Ombudsman reversed the CEC decision and reinstated the results of the voting in Germany. The Ombudsman's decision cannot be justified in law. It was based on a faulty interpretation of the provisions of the Law on Election and the Decree on the Conduct of Elections in Mostar, and an even more faulty and convoluted argument.

With the exception of the voting results discrepancy from Germany and the Ombudsman's decision to reverse the CEC's decision in this regard, and despite the minor problems and irregularities discussed above, in general the election day proceedings in Mostar on June 30th can be considered successful - voting took place in a safe environment, significant irregularities were not reported and the proceedings were administered equitably. However, this conclusion does not take into consideration events preceding the elections day - the campaign process, the extent to which opposition groups were able to compete, the role the media played, and human rights conditions in general. Some of those issues are examined in another report prepared by ICG.

Unfortunately, the Ombudsman's decision on the voting in Germany appears to have been based on political expediency rather than the letter of the law and may provide a precedent for other such decisions in the forthcoming elections. As the EUAM Ombudsman's decision is final according to the Decree on the conduct of elections (Article 11), there is no recourse to re-open the matter. However, the Ombudsman's decision is likely to remain as a source of future complaints and difficulties whenever the political parties are in dispute with each other, and may jeopardise the constitution of effective City and Municipal Councils.