Intelligence Oversight and the Michel Samaha Case in Lebanon

Secrecy is embedded in intelligence work, since an operation cannot succeed if those being investigated know exactly who is collecting information about them. However, the ability to uncover a threat does not place intelligence methods beyond scrutiny. One needs to distinguish between an operation that produces useful information and an operation whose methods can withstand independent examination. Although these objectives can support one another, they are not automatically the same. Born and Leigh (2005) argue that intelligence accountability requires responsibility to be shared among executive, parliamentary, judicial and external review institutions. The Michel Samaha case in Lebanon provides a concrete example of why this matters. A covert operation involving an informant and secret recordings led to prosecution, while the defence challenged the informant’s role in the alleged plot. The case therefore raises a central question: how can intelligence services act before an attack occurs while ensuring that their own intervention remains subject to meaningful oversight?

To understand this issue, it is essential to distinguish between collecting intelligence and producing evidence for a criminal trial. Intelligence collection seeks information that can identify threats and guide decisions, whereas criminal proceedings require the court to determine whether the accused committed the offences charged. A recording's value depends entirely on how it was obtained and the context surrounding it. An informant can provide access that investigators would otherwise lack, while also becoming an active participant in the exchanges later presented as evidence. The United Nations Office on Drugs and Crime (UNODC, 2018) identifies authorization limits and appropriate conduct as central questions in undercover operations. Thus, oversight should examine the relationship between the source and the agency directing the operation, as well as the information obtained from the target. A convincing recording cannot by itself explain every decision that preceded it.

The distinction becomes particularly important when the defence alleges entrapment. Broadly understood, this concerns whether state involvement induced an offence that the person would not otherwise have committed, rather than simply uncovering existing criminal activity. UNODC (2018) notes that jurisdictions approach undercover operations differently and commonly restrict incitement. It would therefore be inaccurate to assume that one universal legal test applies everywhere. Nevertheless, the underlying problem is relevant to intelligence oversight: a source tasked with observing a threat may influence the conduct being observed. Norris and Grol-Prokopczyk (2015), studying American terrorism prosecutions, identify indicators of potential entrapment and argue for examining the likelihood of offending without government prompting. Their findings cannot establish what happened in Lebanon, but they show why source conduct deserves systematic examination. An allegation of entrapment should not automatically discredit an operation, just as an accusation of terrorism should not make its methods immune from challenge.

The Samaha case began in 2012, when the Internal Security Forces’ Information Branch used Milad Kfoury as an informant in an investigation involving former information minister Michel Samaha. In an account reproduced by IMLebanon (2016), Ashraf Rifi stated that Kfoury approached the branch with information about proposed attacks and subsequently recorded meetings with Samaha using concealed equipment. Rifi’s account describes the source returning to the branch after the meetings so the recordings could be retrieved. Since this is a participant’s explanation of the operation, it should be treated as an attributed account rather than a complete independent audit. Its importance is that it describes organized source handling, rather than an individual merely passing on an isolated tip. The intelligence issue is therefore not only what Kfoury knew, but how the agency supervised the role he played in gathering further evidence.

Samaha was arrested on 9 August 2012 and initially sentenced on 13 May 2015 to four and a half years in prison for transporting explosives intended for attacks in Lebanon. The Military Court of Cassation accepted the prosecution’s request to overturn the first judgment in June 2015, opening the way to retrial (L’Orient-Le Jour, 2015). He was released on bail in January 2016 while proceedings continued, a decision that generated sharply opposing political reactions (Reuters, 2016). On 8 April 2016, the Military Court of Cassation imposed a sentence of thirteen years with hard labour and deprivation of civil rights (Noujeim, 2016). This sequence matters because the case was not resolved through a single unquestioned intelligence account. Evidence, intention, and the informant’s conduct became subjects of judicial dispute. Equally, disagreement over the sentences should not be treated as proof that a particular court decision was politically dictated.

The defence’s challenge went beyond denying that recorded conversations had occurred. It sought to contest the circumstances in which the exchanges developed and who initiated contact. Reporting on an October 2015 hearing describes Samaha’s request for telephone records covering six months before his arrest, rather than three. The court accepted that request, which the defence intended to use to show that Kfoury had contacted Samaha rather than the reverse (L’Orient-Le Jour, 2015). The distinction is important, but it is not decisive by itself: initiating a telephone call does not automatically establish that an informant created a criminal intention. Instead, the request illustrates how oversight can turn a general accusation of manipulation into questions that can be tested against records. For intelligence governance, this is a useful example of why preserving the chronology of source contact matters alongside preserving the most incriminating exchanges.

A second dispute concerned whether the informant could be heard in court. Hamza (2016), reporting on a March 2016 hearing for Legal Agenda, describes the court’s refusal to hear Kfoury. The report states that the court relied on his formally recognized status as a confidential informant and referred to Article 257 of the Code of Criminal Procedure. This records the court’s stated justification; it does not settle whether every possible method of testing the source’s conduct was adequate. The broader intelligence problem is the relationship between protecting a source and examining the operation in which that source participated. Confidentiality may be necessary for safety and future cooperation, but it also limits direct questioning. One therefore needs to ask what other material permits a court to assess the source’s role, including recordings, contact records and documented instructions. The argument is for meaningful scrutiny, rather than automatic public exposure of an informant.

The final outcome must also be represented accurately. According to Noujeim’s (2016) account of the judgment, the Military Court of Cassation rejected the defence’s claim that Kfoury had induced Samaha’s criminal intention. The court considered that the relevant intention already existed and found the defence’s explanation unconvincing. Consequently, the article cannot present entrapment as an established fact or describe the intelligence operation as judicially discredited. However, the rejection of that defence does not make the oversight question irrelevant. A criminal judgment determines the issues before the court, while intelligence oversight can also examine institutional arrangements such as authorization, source supervision and record keeping. The broader lesson is that the credibility of a covert operation depends partly on whether its methods can be examined empirically, rather than accepted solely because an agency invokes national security.

Furthermore, it is necessary to distinguish between supervision during an operation and review after it has ended. A court may examine recordings years later, but it cannot recreate instructions or conversations that were never documented. Scheinin’s (2010) compilation of intelligence oversight good practices emphasizes independent scrutiny, access to relevant information and sufficient expertise. Applied to source handling, these principles support recording the source’s task, the limits imposed and significant changes in the operation. These are recommendations drawn from an oversight framework, not findings that particular documents were absent in the Samaha investigation. The distinction matters because a source’s actions should be traceable to decisions for which someone is responsible. Without such records, later scrutiny risks becoming a contest between competing accounts of what the source was supposed to do. Good supervision should make retrospective review possible before a controversy arises.

The case also demonstrates why political controversy and institutional accountability should not be confused. Reuters (2016) documented both condemnation and defence of Samaha’s release on bail by opposing political figures. Such reactions show how strongly an intelligence-related prosecution can become embedded in political conflict. However, anger at a sentence does not establish how the informant was supervised, just as support for the accused does not establish that the operation was fabricated. An assessment based only on political affiliation would replace scrutiny with competing loyalties. Instead, the relevant questions concern what the agency authorized, what the source did and how the resulting evidence was examined. This keeps the focus on intelligence practice while acknowledging the environment in which the operation and trial were interpreted. Accountability should be judged by the quality of the examination, rather than by whether its conclusion satisfies one political camp.

Consequently, the strongest response is to connect operational supervision with independent review. Intelligence services should maintain clear authorizations for sensitive source activity, establish limits on what an informant may do and preserve records that permit those limits to be checked. A specialist reviewer should be able to examine sensitive files confidentially, while courts should have appropriate means to evaluate evidence produced through covert activity. Parliamentary oversight can address the broader rules and institutional performance without attempting to direct individual operations. This follows the complementary approach discussed by Born and Leigh (2005). Public reporting could then explain recurring oversight problems and corrective action without revealing operational identities. These proposals do not require the public to know every intelligence secret. They require the responsible institutions to know enough to challenge an operation’s conduct and demand a response when established rules have been breached.

Ultimately, the Samaha case shows that preventing threats and scrutinizing intelligence methods are connected responsibilities. The documented disputes over contact records, the informant’s appearance and criminal intention provide a concrete basis for examining how covert intelligence becomes courtroom evidence. The analysis relies on published accounts of the proceedings and the operation, rather than direct access to the complete operational file, and therefore cannot establish every internal authorization decision. Nevertheless, the central argument remains clear: an intelligence success should be capable of surviving independent examination. Neither the seriousness of a threat nor the conviction of a target should remove the need to scrutinize source handling. Protecting a confidential informant and ensuring accountability can coexist, provided that supervision and review are designed to make the operation answerable to law.

References:

Born, H., & Leigh, I. (2005). Making intelligence accountable: Legal standards and best practice for oversight of intelligence agencies. Publishing House of the Parliament of Norway. https://www.dcaf.ch/sites/default/files/publications/documents/making-intelligence.pdf

Hamza, R. (2016, March 11). one man show محامي سماحة يتهم رئاسة المحكمة بال .Legal Agenda. https://legal-agenda.com

IMLebanon. (2016, March 4). إخبار ريفي الى الجنائية الدولية! https://www.imlebanon.org/2016/03/04/ashraf-rifi-international-criminal-issue-samaha/

L’Orient-Le Jour. (2015, October 23). La défense s’efforce d’enfoncer la « taupe » Milad Kfoury [The defence seeks to undermine informant Milad Kfoury]. L’Orient-Le Jour. https://www.lorientlejour.com/article/950948/la-defense-sefforce-denfoncer-la-taupe-milad-kfoury.html

Norris, J. J., & Grol-Prokopczyk, H. (2015). Estimating the prevalence of entrapment in post-9/11 terrorism cases. Journal of Criminal Law and Criminology, 105(3), 609–677. https://scholarlycommons.law.northwestern.edu/jclc/vol105/iss3/2/

Noujeim, S. (2016, April 9). La Cour de cassation militaire prouve l’intention criminelle de Samaha et aggrave sa peine. L’Orient-Le Jour. https://www.lorientlejour.com/article/980017/la-cour-de-cassation-militaire-prouve-lintention-criminelle-de-samaha-et-aggrave-sa-peine.html

Reuters. (2016, January 14). Lebanese ex-minister, jailed over Syria plot, is released on bail. Investing.com. https://www.investing.com/news/world-news/lebanese-ex-minister,-jailed-over-syria-plot,-is-released-on-bail-380143

Scheinin, M. (2010). Report of the Special Rapporteur on the promotion and protection of human rights and fundamental freedoms while countering terrorism: Compilation of good practices on legal and institutional frameworks and measures that ensure respect for human rights by intelligence agencies while countering terrorism, including on their oversight (A/HRC/14/46). United Nations Human Rights Council. https://www.refworld.org/sites/default/files/legacy-pdf/en/2010-5/4c0cb0ed2.pdf

United Nations Office on Drugs and Crime. (2018, May). Undercover operations. Education for Justice University Module Series: Organized Crime, Module 8. https://www.unodc.org/e4j/en/organized-crime/module-8/key-issues/special-investigative-techniques/undercover-operations.html

Follow our socials

Community

© 2026. All rights reserved.