Who Watches Those Who Watch Us?
When the state invokes security, we entrust it with powers we would grant to almost no one else: to gather information, monitor threats, and sometimes act without the public knowing its plans.
An interpretive illustration of who oversees institutions with secret powers. It does not depict a real institution.
When the state invokes security, we entrust it with powers we would grant to almost no one else: to gather information, monitor threats, and sometimes act without the public knowing its plans. Without such confidentiality, some legitimate duties could not be carried out. Yet that same space beyond public view can conceal abuses of authority. The serious question, then, is not whether the state may keep secrets, but who can verify that it uses them in the public interest. History offers proof that secret activities are not beyond the reach of democratic oversight—and, at the same time, a warning that establishing an oversight body does not in itself mean that power has actually been restrained.
When Secrecy Is Necessary—and When It Becomes Dangerous
A security service sometimes needs to protect a source, an investigation, the identity of a person at risk, or a method whose public disclosure would genuinely increase danger. That is not automatically a sign that abuses are being concealed. The public's right to information is not the same as a right to every operational detail at every moment. But the opposite extreme is equally dangerous: if an institution can invoke secrecy whenever it is asked to account for its actions, without anyone independently reviewing that decision, an exception becomes a way of governing rather than a limited means of protection.
The distinction between legitimate secrecy and abuse is not determined by whether we approve of an action or trust the current director. What must be examined is the legal basis for the intrusion, its purpose, necessity, proportionality, duration, the handling of information obtained, and accountability when something goes wrong. These distinctions are particularly important in intelligence activities, where the individual affected may never even learn that an intrusion occurred. If that person cannot request a review, there must nevertheless be a reliable means of independent scrutiny. [13]
Official reports, statutory powers, and statistical reviews are valuable, but they answer different questions. A law tells us what an overseer may demand; a report may tell us what the overseer examined; a documented corrective action shows whether oversight had consequences. None of these levels, by itself, establishes what happened at the other two. When an agency publishes summary statistics, for example, the public receives part of the picture, not direct insight into every secret decision. That makes it equally important to ask who verifies whether the published data are complete. [15] [16]
Power that never has to explain itself eventually risks mistaking its own needs for the public interest. But the public, too, must accept responsibility: unlawful conduct cannot be proved merely by pointing out that a piece of information remains classified. A researcher is entitled to demand records of decision-making, clearly defined authority, and meaningful oversight, but must not declare every unpublished document evidence of a conspiracy. The following chapters explore precisely this middle ground: how to preserve necessary secrecy without turning it into an unlimited license to exercise power.
The Church Committee: When the Senate Opened Closed Doors
On January 27, 1975, the U.S. Senate established a special investigative committee led by Senator Frank Church. Its task was not to investigate a single scandal but to examine the conduct of intelligence and security agencies that had operated for years with relatively little public scrutiny. The inquiry covered the CIA, FBI, NSA, and other federal institutions. The Senate's historical account records extensive hearings, the examination of large quantities of documents, and the publication of the final report on April 29, 1976. These are documented milestones, not a conclusion later inferred from theories about hidden power. According to the official Senate history, the committee held 126 full-committee meetings and 40 subcommittee hearings, interviewed approximately 800 witnesses, and examined around 110,000 documents; its report contained 96 recommendations. These figures come from the official historical synthesis, not from my own independent count of the entire archive. [1]
The committee exposed an extensive range of troubling conduct: domestic political surveillance and disruption of organizations associated with the COINTELPRO program, as well as interception of communications under the SHAMROCK and MINARET programs. Individual cases are addressed elsewhere in this series; what matters for our question is that a legislative investigation actually obtained material concerning activities the public had previously been unable to scrutinize to a comparable degree. The committee was not a court, however, and its report does not amount to a criminal conviction of everyone named in its records. [1] [3]
The way the committee worked challenges the simple notion that democratic oversight is possible only through complete public disclosure. Most hearings took place behind closed doors, while the public proceedings were especially prominent in September and October 1975. One reason was to protect sources and methods that indiscriminate disclosure could have endangered. This does not mean everything must remain secret; rather, even proceedings held behind closed doors must be subject to scrutiny: who had access, what evidence they examined, what questions they asked, and on what basis they reached their conclusions. The Senate Intelligence Committee's official catalogue now provides access to extensive historical material, but this article does not claim that every volume and appendix has been examined page by page. [1] [3]
The investigation did not unfold under ideal conditions of impartial cooperation. An archival review published by the National Security Archive describes disputes between the White House, the CIA, and investigators over document requests and the limits of disclosure. This is demonstrable historical friction between branches of government, not proof that every restriction was unlawful or that the committee was denied all significant records. That distinction is itself instructive: oversight that an institution can exercise only with the permission of the very body it is investigating is particularly vulnerable; at the same time, any request for access also raises a legitimate question about protecting information. [4]
The history of the Church Committee is therefore not a fairy tale in which good investigators defeated the intelligence services once and for all. It is a documented example of how investigators can scrutinize a closed system, uncover genuine misconduct, and open a debate about lasting safeguards. Its greatest lesson is not the number of programs exposed, but the change in the public question. Instead of asking, “Should we trust the agencies?”, it became possible to ask, “Who can independently verify that they have complied with the law?” This shift does not eliminate the need for trust, but it requires trust to rest on verifiable procedures. [1] [4]
From an Extraordinary Investigation to Permanent Oversight
On May 19, 1976, the U.S. Senate established a permanent intelligence oversight committee, and the House of Representatives created a corresponding committee in 1977. The creation of these institutions was politically contested. Supporters argued that without continuous access to information, the executive branch would enjoy an excessive advantage; opponents feared the disclosure of sensitive information and reduced effectiveness of intelligence work. Both concerns are understandable, but neither answers the decisive question on its own: does oversight have sufficient powers, expertise, and independence? Establishing a committee changes the legal framework; it does not measure all the committee's subsequent achievements. [2]
U.S. law distinguishes several forms of reporting obligation. Under 50 U.S.C. § 3091, the President must ensure that the congressional intelligence committees are promptly informed of illegal intelligence activities and of corrective action taken or planned; this provision does not itself require the committees to approve every significant operation in advance. Section 3092 governs reporting on other significant intelligence activities and failures, while § 3093 specifically addresses presidential approval and reporting concerning covert actions. The legal conditions and exceptions differ; notifying Congress is not the same as making information public, and sending a notification does not itself prove that legislators received complete information on time. [5] [6] [7]
There is also an important difference between being informed and granting approval. An overseer may be entitled to information without deciding on every future operation; an intelligence activity is not necessarily unlawful simply because citizens were not informed about it. For covert actions, U.S. law provides a specific procedure for presidential authorization and prescribed notification; in exceptional circumstances it also provides for limiting the circle of people who receive the information. Such an exception must be read within the law, not as evidence of an unlimited right to conceal activities. Restricted notification in covert-action matters does not extinguish the obligations the law imposes on the executive branch and oversight bodies. If oversight is reduced to automatically approving the notifications it receives, its democratic purpose is lost, even if the formal steps have been followed. [5] [7]
The legislature is not the only overseer. Under 50 U.S.C. § 3033, the Inspector General of the Intelligence Community has responsibilities that include reviews, audits, investigations of allegations, and reporting. Such an oversight channel, situated within the institutional system but governed by law, may acquire knowledge that an external committee does not always have readily available. For that very reason, a new question arises: does the investigator have access, autonomy over staffing, and the ability to report unwelcome findings? Authority granted by statute is not the same as unrestricted independence. To demonstrate oversight in practice, we need specific reports, identified violations, recommendations issued, and evidence of what was actually corrected afterward. [8] [16]
People who notice irregularities from within an institution have a special role. The U.S. provisions in 5 U.S.C. § 416 and 50 U.S.C. § 3234 establish distinct procedures for certain protected disclosures and safeguards against retaliation, subject to specific conditions concerning eligible persons, reporting channels, and classified information; a separate set of protections applicable to FBI employees, referenced in § 3234, must also be taken into account. Access to classified material does not automatically give an employee the right to publish it wherever they choose. But a state that calls for loyalty must also provide a safe way for someone to report unlawful conduct without being left exposed to retaliation. Without such a channel, the rules of oversight can collapse at the first serious warning from within. [9] [10]
Who Oversees the Overseers?
When we establish a commission or inspectorate, the question of oversight does not end; it moves to another level. Who appoints the overseers? Who can restrict their budget, access to evidence, or time for an investigation? Are they allowed to publish an unwelcome finding, and can anyone verify what happened after their recommendation? An oversight body is not necessarily independent simply because its name includes the word “independent”; nor is it necessarily biased merely because it was created through a political process. Independence must be assessed through specific powers and evidence of how they have been used. [8] [16]
A contemporary example is the U.S. Privacy and Civil Liberties Oversight Board (PCLOB). In a review published on April 2, 2026, its staff examined changes to the surveillance program under Section 702 of FISA, particularly following the 2024 RISAA reforms. The official summary and the introductory sections of the full report examined directly describe new restrictions, requirements for prior approval of certain queries, and expansions of certain aspects of the framework. A procedural detail is important: this is a PCLOB staff report, which was not adopted by a vote of a quorum of the Board. In the report authors' assessment, the reforms have so far improved compliance with certain procedures and protections for privacy, but there has not yet been enough time or data to assess all their effects conclusively. It would therefore be unfair to claim either that the risks have disappeared or that the report proves oversight is generally ineffective. The same legislative change can produce different kinds of consequences. [14] [18]
Also in April 2026, the Office of the Director of National Intelligence (ODNI) published a public statistical report on the use of national-security surveillance authorities during calendar year 2025. ODNI states that it has published reports of this kind annually since 2014; the existence of regular public reporting is itself important for traceability. Statistics, however, show only what their methodology captures; they do not directly measure the total number of unjustified intrusions. To reach that conclusion, we would also need definitions, independent verification, and information about unreported cases. Different forms of transparency must therefore reinforce one another: statistical publication, oversight audits, the right to lodge complaints, and public accountability for identified errors. [15] [16]
An additional problem arises when information or operations cross borders. In a research and advocacy report, Privacy International highlighted gaps in oversight of intelligence-sharing between countries. Its position is not a judicial finding that all international exchanges are unlawful, but it raises a legitimate investigative question: can the overseer in one country discover how information was used once another country received it? If accountability disappears at the boundary between jurisdictions, each individual agency may claim to have acted within its own rules while the person affected is denied a complete account. [17]
The Slovenian Dimension and the Right to Privacy
Slovenia is not outside this question. The Parliamentary Oversight of Intelligence and Security Services Act (ZPNOVS) governs the powers of the parliamentary commission KNOVS and the obligations of the agencies it oversees. The Rules of Procedure of the National Assembly supplement the commission's position within parliamentary work. These provide a basis for institutional oversight, not proof that the commission obtained all documents in every particular case or remedied every irregularity it identified. ZPNOVS also provides for legally defined restrictions and the possibility of a temporary deferral of oversight of particular measures under specified conditions. Such an exception is not in itself evidence of abuse, but it is essential to understanding the reach of oversight. The official consolidated text from 2007 and the parent-act record in PISRS were reviewed; this was not an exhaustive assessment of every possible subsequent change to the legal framework or of KNOVS's actual reports through November 2026. [11] [12] [19]
International comparisons are useful only when we respect each country's legal framework. Reporting to the U.S. Congress, oversight by a Slovenian parliamentary commission, and review by European courts are not three names for the same power. Parliament examines political and legal accountability within a particular framework; judicial review addresses legality and rights within the court's jurisdiction; and an inspectorate or other oversight body may examine internal procedures. If we expect one of them to do everything, we will overlook the gaps between them. Of particular importance is access to a remedy for someone who learns of an intrusion upon their rights only after a delay—or never learns of it at all.
On May 25, 2021, the Grand Chamber of the European Court of Human Rights considered the United Kingdom's bulk interception regime in Big Brother Watch and Others v. the United Kingdom. In its examination of that regime, the Court emphasized the need for independent authorization at the outset, supervision throughout the process, and subsequent independent review. At the same time, it assessed different legal regimes separately; with regard to receiving material from foreign intelligence services, it did not find the same violation as in certain domestic interception procedures. The shortcomings identified do not therefore mean that every interception is impermissible or that the case can be equated with Slovenian law without further analysis. They do mean that extensive secret powers require concrete, reviewable legal safeguards. [13]
In a democracy, the right to privacy cannot be measured solely by whether a person has something to hide. Even a perfectly lawful life contains conversations, relationships, and thoughts that are not meant for a government database. If the authorities can construct a detailed picture of someone's contacts and habits without adequate safeguards, the consequences go beyond whether proceedings will be brought against that person. Fear alone may cause an individual to change their behavior. The question of independent authorization and subsequent review is therefore also a question of freedom of conscience, expression, and association—not merely a technical issue of data processing. [13]
| Form of oversight | What it can provide | What it does not establish on its own |
|---|---|---|
| Congressional committees (US) | Statutory notification and oversight of intelligence activities. [5] [6] [7] | That committees approve every operation in advance or receive complete information in every case. |
| Intelligence Community Inspector General (US) | Audits, investigations, reporting and recommendations. [8] [16] | That every recommendation has been implemented or every violation remedied. |
| PCLOB (US) | Specialized scrutiny of privacy and civil liberties. [14] [18] | That the April 2026 staff report was adopted by a quorum of the Board. |
| KNOVS (Slovenia) | Parliamentary oversight under ZPNOVS, subject to statutory conditions and exceptions. [11] [12] [19] | That formal powers guarantee full access in every case or effective correction of all deficiencies. |
| Judicial review of rights (ECtHR) | Review of whether the regimes examined comply with the ECHR and safeguards against abuse. [13] | That a judgment concerning the UK regime alone establishes the unlawfulness of a particular Slovenian measure. |
Different institutions scrutinize different aspects of secret powers. Legal authority, a review actually performed and an effective remedy are not the same thing. This is an editorial synthesis of the cited sources, not an evaluation of each institution's overall performance.
Accountability Without Naivety or Paranoia
What, then, would it mean to take oversight of the overseers seriously? First, there must be lawful and sufficiently broad access to evidence; the body being scrutinized must not be the only one setting the questions; findings must rest on a clearly identifiable methodology; corrective measures must exist, along with a way to verify their implementation; and the person affected must, under appropriate conditions, be able to seek independent review. These are not measured results of every existing oversight system, but editorially proposed criteria by which the public can judge its quality. No commission, court, or individual can replace all the others without limits. [1] [8] [13]
Protecting legitimate secrets is equally important. Complete publication of all personal data, the identities of sources, or information from ongoing investigations could endanger people and undermine future lawful work. Secrecy must, however, have a comprehensible purpose and be subject to independent review. Democratic accountability does not begin only when every document is made public. It begins when rules, authorized overseers, and real consequences for abuses of power exist even behind closed doors. This asks more of us than a call simply to trust the state, but also more than the conviction that every state secret is evidence of a crime. [4] [5]
Perhaps the hardest thing for a reader to accept is that open questions sometimes remain open. A document that has not been recovered is not proof of intent; equally, an official denial is not the final word if there is verifiable evidence to the contrary. What we need, therefore, is investigative patience: we must distinguish facts from suspicions, equip overseers with appropriate authority, examine their work, and refuse to let necessary criticism of power turn into automatic condemnation of everyone who performs public duties. The human right to truth is not a right to a predetermined conclusion.
With this, Part VI returns to its starting point. In earlier articles, we explored parallel structures, covert operations, secret resistance networks, infiltration, and blackmail; here, the question turns toward those who claim to oversee all of this. Who watches those who watch us? No single person or body can do so, and none should be exempt from further scrutiny. The answer can only be a system of distinct forms of accountability that restrain one another, permit lawful criticism, and place human dignity above the convenience of unchecked power. This is an editorial and philosophical standard, not a claim that such a system has already been fully achieved anywhere.
Sources and further reading
- U.S. Senate Historical Office (updated 2026). Senate Select Committee to Study Governmental Operations with Respect to Intelligence Activities (The Church Committee). Official historical overview of the investigation. Source
- U.S. Senate Historical Office (May 19, 2026). The Senate Creates the Select Committee on Intelligence. Official historical account of the establishment of the permanent Senate committee. Source
- U.S. Senate Select Committee on Intelligence. Intelligence Related Commissions — Church Committee reports and hearings index. Archival catalogue; not a substitute for direct examination of every document. Source
- National Security Archive, George Washington University (July 20, 2015). Church Committee, White House and CIA — declassified briefing book. Curated archival presentation with documents concerning the investigation and disputes over access. Source
- U.S. House of Representatives, United States Code (as of 2026). 50 U.S.C. § 3091 — General congressional oversight provisions. Source
- U.S. House of Representatives, United States Code (as of 2026). 50 U.S.C. § 3092 — Reporting of intelligence activities other than covert actions. Source
- U.S. House of Representatives, United States Code (as of 2026). 50 U.S.C. § 3093 — Presidential approval and reporting of covert actions. Source
- U.S. House of Representatives, United States Code (as of 2026). 50 U.S.C. § 3033 — Inspector General of the Intelligence Community. Source
- U.S. House of Representatives, United States Code (as of 2026). 5 U.S.C. § 416 — Additional provisions regarding intelligence community inspectors general. Source
- U.S. House of Representatives, United States Code (as of 2026). 50 U.S.C. § 3234 — Prohibited personnel practices in the intelligence community. Source
- Republic of Slovenia, PISRS (UPB1 consolidated text, 2007). Zakon o parlamentarnem nadzoru obveščevalnih in varnostnih služb (ZPNOVS) [Parliamentary Oversight of Intelligence and Security Services Act]. Subsequent amendments still require further direct verification. Source
- Republic of Slovenia, PISRS (2002; as amended). Poslovnik Državnega zbora [Rules of Procedure of the National Assembly]. Powers of parliamentary working bodies. Source
- European Court of Human Rights, Grand Chamber (May 25, 2021). Big Brother Watch and Others v. the United Kingdom, applications nos. 58170/13, 62322/14, and 24960/15. Judgment and legal summary concerning bulk interception and safeguards. Source
- Privacy and Civil Liberties Oversight Board (April 2, 2026). FISA Section 702 — 2026 staff report overview. Official online presentation of the staff report on the changes following RISAA 2024; see also [18]. Source
- Office of the Director of National Intelligence (April 1, 2026). Annual Statistical Transparency Report Regarding National Security Authorities, Calendar Year 2025. Announcement of the annual report, without adopting its independently unverified figures. Source
- Office of the Director of National Intelligence. IC OIG Semiannual Reports. Official index of semiannual reports, not an analysis of the entire 2026 edition. Source
- Privacy International (April 24, 2018). New Privacy International report reveals dangerous lack of oversight of secret global surveillance networks. Research and advocacy analysis requiring further documentary verification of individual cases. Source
- Privacy and Civil Liberties Oversight Board, staff (April 2, 2026). Report on the Surveillance Program Operated Pursuant to Section 702 of FISA, 61 pages, public release. The cover, introduction, and selected relevant passages were examined directly; the report was not adopted by a vote of a quorum of the Board (footnote 1 on p. 1). Source
- Republic of Slovenia, PISRS (record accessed November 18, 2026). ZPNOVS — parent act, ID ZAKO3455, and links to consolidated texts. Description of powers and exceptions verified; no claim that every KNOVS proceeding was reviewed. Source