The controversy surrounding NAZA has brought an old question in investigative journalism back to the center of the debate in recent days: How much weight can be given to the testimony of a source who is not revealed to the public?
One of the main criticisms directed at the film by Yuval Abraham and Rachel Szor is that it relies on anonymous testimonies from 24 Israeli soldiers and intelligence personnel.
The film does conceal the identities of the interviewees, and the IDF rejected its claims, arguing, among other things, that anonymous sources make it difficult to independently verify the information. This is a legitimate argument that deserves serious examination. But the mere anonymity of a source is not enough, by itself, to determine that the information they provide is unfit for publication. The more important question is what was done to examine, cross-check, and verify it.
An anonymous source to the viewer is not necessarily an anonymous source to the journalist. The journalist knows who is standing in front of them. They should know where the person served, what they did, what information they had access to, whether their account matches other facts, whether others describe similar events, and whether documents, messages, or other evidence support their testimony.
The public receives the information without the name because revealing it could cost the source their job, career, legal standing, and sometimes much more, especially with the International Criminal Court looming over the situation. People can support or oppose this approach, but investigative journalism is sometimes difficult to conduct without it.
Anonymous sources take down men accused of sexual misconduct
Take the case of Moshe Ivgy, for example. In February 2016, Walla published an investigation in which six women described alleged sexual misconduct by him. At the time of publication, none of them had yet filed a police complaint, and most of the women who spoke publicly did so anonymously. If we had applied the “no name, no story” test back then, we could have simply closed the browser and gone to lunch.
Instead, police began examining the allegations following the report, women came forward, an investigation was opened, and at the end of the process Ivgy was convicted of indecent acts and sexual harassment and sentenced to 11 months in prison. The women’s anonymity at the beginning did not make their testimonies false. It simply allowed them to begin speaking.
The case of Yehuda Meshi-Zahav may also be examined. In March 2021, Haaretz published a major investigation based on six people who described alleged sexual abuse attributed to him. All of the people interviewed appeared anonymously in the investigation. The next day, police began examining the allegations, and Lahav 433 later reopened the investigation. In this case, the allegations ultimately did not lead to a judicial ruling. But one thing is clear: The fact that the sources did not appear by name did not prevent authorities from taking the information seriously and investigating it.
Alon Kastiel provides another interesting lesson. In November 2016, Sharon Shpurer began publishing information and testimonies in The Hottest Place in Hell about his conduct. Some of the women were identified only by initials, and the original exposure was also based on people from the nightlife industry who described complaints that had accumulated against him.
One woman whose story was published decided, following the report, to go to the police. From there, an investigation developed that eventually included 12 complainants. Kastiel admitted guilt and was convicted in a plea agreement of attempted rape, forcible indecent assault, indecent assault, and sexual harassment. He was sentenced to four years and nine months in prison.
The Bar-On-Hebron affair takes us even further back. In January 1997, Ayala Hasson appeared on Israel’s Channel 1 with a dramatic report about the circumstances surrounding the appointment of Roni Bar-On as attorney general. The public was not given a detailed list of her sources. In fact, the Seventh Eye later reported that Hasson did not initially even share her sources with the editor of the Mabat program, but instead presented the findings to News Division director Rafik Halabi.
Four days after the broadcast, police transferred the matter for investigation. A State Attorney’s Office review confirmed key parts of the report and drew more complex conclusions about the criminal responsibility of those involved.
If Hasson’s sources had been required that evening to appear before Haim Yavin, provide their names, ranks, and phone numbers, the Bar-On-Hebron affair might have remained a hallway conversation.
Then came the Harpaz document, which is important here for the opposite reason. In August 2010, Amnon Abramovich and the late Roni Daniel revealed what was then called the “Galant document” on Ulpan Shishi. Abramovich defended the person who provided him with the material and described him on air as “a reliable and productive source for three decades.” The source was not revealed to viewers.
The report quickly led to a review by the attorney general and a police investigation. But investigators later discovered the document itself was forged. Boaz Harpaz was later convicted of forging and using it.
And that does not undermine the argument. It reinforces it. An anonymous source is not a certificate of authenticity, but it is also not a certificate of disqualification. A journalist who receives information from a confidential source cannot simply say, “They told me, therefore it is true.” They have to work harder precisely because the public cannot independently examine who the person is.
NAZA directors must be asked about verification of their sources
The Harpaz document shows what can happen when journalists give too much weight to material from a source considered reliable before establishing its authenticity. Ivgy and Kastiel show the other side: sometimes anonymity is exactly the condition that allows truthful information to emerge.
Therefore, the question that should be asked of Abraham and Szor regarding NAZA is not “Why don’t the soldiers show their faces?” That is the easy question.
The much harder questions are: How did you verify that they actually served in the roles they describe? Did you cross-check each claim with more than one source? What documents or data did you see? Were there testimonies that contradicted the film’s thesis and were not included? What did you present to the IDF before publication, and how did it respond? Did the 24 people come from different systems and units, or are some of them describing the same incidents? When two sources provide different versions of events, which version made it into the film, and why?
These are journalistic questions. “Show me their faces” is a far less interesting demand.
The other side must also acknowledge something that may be less comfortable for those defending the film: The fact that there are 24 anonymous sources does not automatically make NAZA true. Twenty-four testimonies can still be biased, incomplete, or taken out of context. A documentary film is an edited work. Someone decides what to include, what to leave out, where to cut, and which sentence will follow another.
That is precisely why it should be watched and examined. But rejecting a film simply because its interviewees are anonymous would mean rejecting, in hindsight, a significant part of investigative journalism.
Moreover, we tend to accept anonymous sources when they expose people we dislike, but suddenly view them as an unacceptable ethical problem when they reveal something we do not want to hear. That is less a journalistic principle and more a matter of convenience.
After watching NAZA, we may find that the filmmakers carried out every possible verification process. We may also watch it and find major holes. At this stage, both possibilities remain open. The fact that the interviewees are not identified does not decide between them. It simply raises the standard placed on the filmmakers.
That is what should be remembered amid all the noise surrounding the film: The problem was never that a source is anonymous. The problem begins when the verification process is also anonymous.