V. Criminal Cases against Migrant Domestic Workers
Human Rights Watch reviewed 84 cases where the MDW was accused of a crime—61 of which dealt with accusations of theft. The analysis does not include cases where the migrant was guilty solely of violating immigration rules, and revealed numerous violations of international standards of due process and fairness.
A. Frequent Pre-trial Arrests
At least 64 of the 84 MDWs (76 percent) who were accused of a crime were detained prior to trial. Deprivation of liberty pending trial must be an exceptional measure to preserve the principle of presumption of innocence. However, a review of judicial records shows that the Public Prosecutor’s office regularly orders the detention of MDWs even when the evidence against them is weak. This practice increases the danger that MDWs are presumed guilty before a court finds them innocent, rather than the other way around.
Of the 18 MDWs in the sample who were eventually found not guilty, 15 (83 percent) were detained during trial. On average, these workers spent three months in detention before a judicial decision for their release was issued; however, in four cases, they spent at least eight months in jail before the court found them not guilty. MDWs were held pre-trial even in cases where they were accused of stealing money or goods valued at less than $1,500.
However, these MDWs are likely to have stayed even longer in jail because MDWs, like all foreigners, are referred to General Security at the end of their sentence, even if they are found not guilty and have valid residency papers. We do not know how long they would have spent at General Security, since the court documents do not reflect this. In our sample, there were only two cases where the judge explicitly asked that the jailed domestic worker be released instead of being referred to General Security.
Pre-trial arrest was particularly lengthy in felony cases before the Criminal Court (Mahkamat al-Jinayaat). Human Rights Watch reviewed 13 cases of MDWs accused of a felony and detained. On average, the worker spent 21 months in pre-trial detention.
B. Abuse during Interrogation
Research for this report focused on reviewing court documents, rather than investigating how MDWs are treated during interrogation. However, in the course of research, a number of testimonies were collected from workers who alleged they were abused during interrogation.
A Filipina woman whom police interrogated in July 2009 about a robbery at her employer’s house told Human Rights Watch:
They took me to the police station. Every question they would slap me. Madame and Mister were in the room, as well as two investigators. Then the police took me to another room and they started beating me with a plastic stick on my hand and feet. The police man would say, “I will count till five” and then slap me. At one point they threatened to beat me with an iron rod. They made me see it but did not use it. When I kept denying, they said, “If you don’t answer, I will put my cigarette in your eyes.” The interrogation lasted 30 minutes. They kept me in the police station for three days and then took me back to my employers.
The Lebanese Center for Human Rights (CLDH) conducted interviews in May 2009 with 82 migrant women in Lebanon’s Tripoli jail. Two interviewees said police had beaten them during interrogation, while a third said that a policeman had raped her.
C. Lack of Defense Lawyers
Of the 84 MDWs accused of a crime in our sample, 37 (44 percent) did not have a defense lawyer.
This percentage is even higher if we exclude cases before the Criminal Court dealing with felonies, where a defense lawyer is mandatory. Human Rights Watch reviewed 66 cases before the First Instance Criminal Court (also known as the Sole Penal Judge). MDWs did not have a lawyer in 32 of those cases, and Human Rights Watch could not verify 10 other cases—meaning MDWs lacked a lawyer in at least 48 percent of cases before the First Instance Criminal Court.
D. Access to Interpreters
The majority of MDWs arrive in Lebanon with either limited or no Arabic language skills. While many eventually learn enough Arabic to get by in their daily work, most are not fluent and are unlikely to understand the vocabulary used in a police interrogation or trial. MDWs rely on ad hoc arrangements for interpretation in police stations and court proceedings. In at least 57 of the 84 cases that Human Rights Watch reviewed, the MDW did not have access to a certified translator. Court documents and police reports clearly indicate that an interpreter was present in only 11 cases. Of those, three were official sworn translators (certified by the court); one a passerby; and one a migrant. Three were provided by the embassies of the MDW; one by Caritas, an NGO helping MDWs; and two by the employer.
Interpreters were rare even when the MDW was accused of a serious crime. Human Rights Watch reviewed 13 cases where the worker was accused of a felony: only seven had an interpreter present.
According to Lebanese criminal procedure, anyone is entitled to a translator during interrogation (article 81 of Code of Criminal Procedure). This requirement is even printed on the forms used during interrogation. But in practice, many MDWs do not have access to interpreters and, in many cases, investigators simply record that the worker understood enough Arabic for the interrogation to proceed. In reality, few migrant workers master the language to the point where they feel comfortable responding to an investigator’s questions.
Human Rights Watch reviewed a number of police reports where the MDW indicated that she only speaks “a little Arabic,” yet police proceeded with the interrogation without an interpreter. For example, P.V. (not her real initials), a Sri Lankan domestic worker was accused by her employers of theft. The police did not request a translator, even after P.V. stated, “I understand Arabic a little and I can only answer a few things in Arabic.” P.V. had only been in Lebanon for eight months when she was interrogated, making it highly unlikely that she understood enough Arabic to comprehend the details of the case against her. Her confessions in the initial deposition were eventually used to convict her of theft. The court sentenced her to three months in jail and ordered her to pay back her employer LBP1,500,000 ($1000) in restitution for the jewelry taken. In another case involving a Sri Lankan worker accused of theft, the police contacted the Sri Lankan embassy to ask for a translator only at the end of the interrogation, for reasons that were unclear in the police report.
When translators are provided, the arrangement is often ad hoc and problematic. In one case, the police relied on a Sudanese national to translate the interrogation of an Ethiopian migrant worker, M.W. (not her real initials), accused of theft by her employer. Since M.W. spoke no Arabic, the Sudanese national translated the questions into English, which she spoke only a little. In at least two other cases, the police relied on a translator provided by the employer in cases where the employer was accusing the worker of committing a crime. In both cases, the courts eventually recognized the conflict of interest inherent in relying on such an interpreter, but not before the worker had spent time in jail based on the initial translation.
The employer of D.A. (not her real initials), 26, a Filipina migrant worker, accused her of stealing jewelry from his house. During D.A.’s interrogation, the police relied on the lawyer of the employer, as well as a friend of the employer, to translate D.A.’s deposition. In her preliminary investigation, D.A. confessed to the theft. However, in subsequent interrogations, D.A. was able to obtain her own translator, and called for her first confession to be put aside because she claimed after it had been made, the police beat her. In its decision, the court stated that “the translation of the defendant’s testimony by the accuser’s lawyer and personal friend, both of whom are not neutral or objective, leads the court to put aside this confession.” In the absence of any other material evidence, the judge found D.A. innocent. Despite a positive final outcome, the worker spent nine months in pre-trial detention.
In another case, with more serious charges, the police relied on a worker employed in the household of the employer’s friend to translate the deposition of a domestic worker accused of attempting to murder the employers’ children. The first instance Criminal Court found the domestic worker guilty of attempted first degree murder and sentenced her to four years in jail. The lawyer at Caritas Migrant successfully brought an appeal to have a retrial, during which the accused worker had the services of a sworn translator. She was acquitted.
E. Evidentiary Burden
Human Rights Watch interviewed many MDWs, as well as officials at the embassies of MDW-sending countries, who complained that courts favor the employer, and convict workers on the basis of nothing more than the employer’s testimony.
In fact, Lebanese courts have a mixed record. While the courts convicted the MDW based solely on the employer’s accusation in a number of cases, in others the court required further evidence. Out of the 61 cases reviewed where the employer accused an MDW of theft, the court found the MDW innocent in 18 of the cases because of lack of evidence.
A positive precedent was set in 2002 when the Court of Appeal of the Metn district found Jenet Teklo, the Ethiopian domestic worker accused by her employer of theft, not guilty because the charge of theft “has to be supported with external evidence like fingerprints, witnesses, or finding the stolen goods, which are not present in this case.” However, the standard outlined in the Teklo case has not been followed widely. In a number of cases reviewed, the courts convicted the worker solely based on the employer’s accusation, even in certain cases where it seemed weak or vague. For example, on December 29, 2006, a penal judge in Beirut sentenced a Sri Lankan worker to six months in jail for stealing $5,000 worth of cash and jewelry from the employer. The worker, who had escaped from her employer, consistently denied stealing anything. The employer did not give any details about the stolen jewelry or money in his initial complaint, did not bother to attend any of the trial sessions, and did not request any civil compensation.
In many cases, the court treated the fact that a worker “ran away” from the employer as evidence supporting the accusation of theft, even though the MDWs often advanced other legitimate reasons for leaving their employers. For example, a penal court judge sentenced S.U. (not her real initials), a Sri Lankan worker, to 45 days in jail for stealing from her employer. In his decision, the judge relied heavily on the fact that the migrant worker had “escaped” from her employer, an act he saw as evidence that she had something to hide— even though the worker told the court the employer had owed her four months of unpaid wages and had refused to let her travel to Sri Lanka when her husband died.
Some cases show how easily certain employers accuse workers of theft. In one case, the employer accused the worker of running away and stealing a gold ring worth $300. The court initially tried the MDW in absentia and sentenced her to three months in jail. It then emerged that the ring had been missing three months before the worker left her employer. After she was detained, the court re-tried the MDW and found her not guilty for lack of evidence. In another case, employers accused their MDW of theft only after learning she intended to file a complaint before the Labor Court for unpaid wages. The employers accused the MDW of stealing $2,650 and jewelry worth $1,500. The court found the MDW not guilty for lack of evidence.
 Article 9(3) of the International Covenant on Civil and Political Rights, which Lebanon ratified on November 3, 1972, notes that “It shall not be the general rule that persons awaiting trial shall be detained in custody.” The Human Rights Committee has consistently held that “pre-trial detention should be the exception and that bail should be granted, except in situations where the likelihood exists that the accused would abscond or destroy evidence, influence witnesses or flee from the jurisdiction of the State party.” Communication No. 526/1993, M. and B. Hill v. Spain (Views adopted on 2 April 1997), UN doc. GAOR, A/52/40 (vol. II),p. 17, para. 12.3.
 This percentage is based on the fact that of the 18 cases where the court found the MDW innocent of any accusation of theft, 15 of them had been detained during trial.
 Case no. 1398/2006, Sole Penal Judge in Beirut (8.21 months), Case no. 322/2007, Sole Penal Judge in Keserwan (9.03 months), Case no. 482/2005, Investigative Judge in Mount Lebanon (9 months), Case no. 323/2002, Felony Court of Beirut (10 months).
 For example, in a number of cases, there was a specific request from General Security to the court to transfer the custody of a worker to General Security even after the court found the person not guilty. See for example, decision of Judge Nazem al-Khoury, March 20, 2006, agreeing to transfer the passport, work permit to General Security. In other cases, there is a standing order from general security asking that all cases of foreigners be referred to them.
 Case no. 2047/2000, Sole Penal Judge in Beirut, Case no. 387/1998, Investigative Judge in Mount Lebanon. In the case of Arlinda Sakbibit, from the Philippines, accused by employer of theft of antique piece, the investigative judge decided to drop the charges due to lack of proof and ordered her release without being referred to General Security.
 Human Rights Watch interview with Filipina worker E.V. (not her real initials), February 11, 2010.
 E-mail from CLDH to Human Rights Watch, April 23, 2010.
 In the other 16 cases, it was unclear from the court documents if there was an interpreter.
 Article 14(3) of the International Covenant on Civil and Political Rights, which Lebanon ratified on November 3, 1972, recognizes the right of everyone to “have the free assistance of an interpreter if he cannot understand or speak the language used in court.”
 Police Report, August 21, 2005 (on record with Human Rights Watch). The irony is that the paper on which her deposition was noted, has a standard language that the person being questioned has the right to a certified translator.
 Case no. 2212/2005, Sole Penal Judge in Beirut.
 Police Report no. 152/302 (Mafrazat Istiksa’ Jabal Lubnan), December 21, 2006 (on record with Human Rights Watch).
 Case no. 2374/2007, Sole Penal Judge in Beirut.
 Case no. 1398/2006, Sole Penal Judge in Beirut.
 Email from Caritas Migrant detailing case.
 Case no. 2831/2006, Sole Penal Judge in Baabda. Deposition of employer at General Security June 10, 2006 (on file with Human Rights Watch). The MDW did not have a lawyer.
 Case no. 2427/2007, Sole Penal Judge in Beirut. The employer told the court that the agency had informed them that the worker was lying in stating that her husband had died.
 Case no. 3265/2003, Sole Penal Judge in Beirut.
 Case no. 2117/2007, Sole Penal Judge in Baabda.