SPECIAL · FAMILY

Encryption, Turkish case law and game and document chats put much of a teen’s messaging beyond a parent’s reach, so trust must cover the rest.

ABOUT 27 MINUTES

Can a parent or guardian read a teenager’s messages? Mostly no, and rarely in the way the question imagines. The apps teens use most keep message content out of reach of the phone’s own parental controls. Turkish case law suggests that reading the messages of a child over 15 can carry criminal risk. In the U.S. the answer changes at the state line. That is not a reason to look away. A parent or guardian can watch openly and closely: see whom your teen talks to, block the worst images before they land, set the rules in writing and know exactly whom to call. But no control reaches every channel a teenager can open. In this editor’s judgment, the one protection that works on all of them is the teenager’s trust, and the aim of everything else is that your teen comes to you before a problem becomes irreversible.

The impulse is easy to understand. In 2025 the CyberTipline run by the U.S. National Center for Missing & Exploited Children (NCMEC) received 1.4 million reports of online enticement, including more than 800 in which an offender traveled to meet a child and more than 80,000 about sextortion ^[1]. Enticement reports rose 158% from 2024, a rise NCMEC says was “impacted by the 2024 enactment of the REPORT Act,” which requires platforms to report enticement ^[1].

Financial sextortion, in which an offender obtains an intimate image and demands money, is the fastest-moving part: more than 50,000 NCMEC reports in 2025, about 137 a day, up from more than 36,000 in 2024 ^[2]. The FBI says victims are “typically males between the ages of 14 to 17, but any child can become a victim,” and that offenders are usually outside the U.S. ^[3].

Turkey counts differently. According to the statistics agency TÜİK, security units recorded 23,123 children as victims of sexual crimes in 2025, 19,689 of them girls ^[4]; Anadolu Agency put that at 10.3% of all child crime victims recorded ^[5]. These are children who came to police and gendarmerie attention, online and offline, not a measure of how common abuse is.

One more number matters most. When the relationship to the victim was known, 94% of the sextortion reports NCMEC received from the public came from the child or a parent, most from children themselves ^[2]. Children report. Whether they report to their parents depends on what happens at home when something goes wrong.

The stranger overseas is real. NCMEC notes that offenders increasingly move conversations “more quickly to private messaging and encrypted apps,” where detection is harder ^[2].

The other half of the risk is someone the family already trusts: a coach, a tutor, a teacher or a family friend. The most cited U.S. estimate is old. A 2004 review for the U.S. Department of Education by Charol Shakeshaft, drawing on survey data collected in 2000, estimated that 9.6% of students in grades 8 to 11 had experienced some form of educator sexual misconduct, broadly defined to include non-contact behavior ^[6]. Read it as an order of magnitude, not a current rate.

The reporting machinery around schools is firmer. The National Education Association says “all 50 states have enacted laws that require educators to report suspected child abuse and neglect” ^[7]. In Turkey, the Child Protection Law (Law 5395) requires judicial and administrative authorities, police, health and education institutions and civil society groups to notify the social services authority of a child in need of protection ^[8], and Ministry of National Education (MEB) guidance sets out how schools report violence against children ^[9].

WARNING SIGNS OF GROOMING

NCMEC’s NetSmartz program lists red flags it teaches children to watch for ^[10]. A parent can watch for the same pattern:

None of these proves abuse. Together, and especially with secrecy, they are a reason to talk and, if needed, to report.

U.S. federal law starts from the Wiretap Act, which bars intercepting a communication unless a party to it consents ^[11], and the Stored Communications Act, which covers unauthorized access to stored messages ^[12]. Neither mentions parents. Some courts accept “vicarious consent,” a parent consenting for a child: a federal court in Utah in 1993 ^[13] and the Sixth Circuit in Pollock v. Pollock (Sept. 1, 1998), which required “a good faith, objectively reasonable basis for believing that it is necessary and in the best interest of the child” ^[14]. In 2023 the Ninth Circuit said “the vicarious-consent doctrine is not the law of Nevada or this Circuit” ^[15].

State law adds a second layer. Some U.S. states, including California, Florida, Illinois, Pennsylvania and Washington, require everyone’s consent to record a conversation, according to a compilation by RecordingLaw.com ^[16]. Unlawful recordings can be barred as evidence ^[17], and the people recorded can sue ^[18].

The newer U.S. laws regulate platforms, not parents. COPPA covers children under 13 ^[19]. Florida may now enforce HB 3, which bars accounts for under-14s on covered platforms and requires parental consent at 14 and 15, after the 11th Circuit lifted an injunction on Oct. 14, 2026; the case continues ^[20]. Texas’s SCOPE Act duty on platforms to monitor and filter content for minors remains blocked by a district court injunction; on Oct. 28, 2026, the full Fifth Circuit agreed to rehear the case and vacated a July panel opinion, with argument not yet scheduled ^[21][22][23]. The Senate’s Kids Online Safety Act (S.1748) stalled after Sen. Rand Paul blocked a fast-track vote on Sept. 30; the Senate returns for legislative business on Nov. 9, and any bill not enacted by Jan. 3, 2027 dies with this Congress ^[24][25].

Turkey starts from the Constitution, whose Article 22 says “Haberleşmenin gizliliği esastır,” the privacy of communication is fundamental ^[26]. The Penal Code (TCK) makes violating it (Article 132), recording private conversations without consent (Article 133) and violating private life (Article 134) crimes ^[27]. The Civil Code (TMK) gives parents custody but lets minors with the capacity to judge exercise strictly personal rights on their own (Article 16) ^[28].

Turkish case law suggests that reading the messages of a child over 15 can carry criminal risk. The reasoning appears in a Nov. 3, 2020 majority decision of the Yargıtay Ceza Genel Kurulu in a case between spouses, not a parent and child, which also said a parent acting to protect a child from danger may rely on the “mistake about wrongfulness” provision of TCK Article 30/4 ^[29]. This desk has seen only a law firm’s reproduction, not the official text. Separately, Law No. 7578 bars social networks from serving children under 15, and its implementing regulation took effect today, Nov. 1, 2026 ^[30][31][32]. New accounts require an age check through the e-Devlet government portal, existing accounts must be verified by March 31, 2027, and accounts found to belong to under-15s are to be closed. Accounts of 15- to 17-year-olds are private by default, can receive messages only from accounts they follow and cannot livestream, and platforms must offer parental control tools ^[33].

The tools built into phones shape what reaches a child; they do not show parents what the child wrote.

On Apple devices, Communication Safety is on by default for children under 18 in a Family Sharing group ^[34]. It detects nudity in photos and videos in Messages, AirDrop, FaceTime video messages and some third-party apps, blurs the image and suggests messaging a trusted adult. With a Screen Time passcode set, children under 13 cannot view flagged content without a parent or guardian entering it. The analysis runs on the device; gore detection and reporting to Apple require version 27 of Apple’s systems ^[34]. None of it lets a parent read the conversation.

On Android, Google’s Family Link manages apps, screen time, filters and device location, but Google says parents can’t remotely listen to calls or see the screen. A parent can change the child’s password, but Google then notifies the child, so it cannot be done quietly ^[35]. Contact limits cover only Google’s and Samsung’s phone and messaging apps; with apps like WhatsApp, Google warns, children “can still call and text numbers that aren’t in their parent-managed contacts” ^[36]. A child under 18 needs a parent’s approval to end supervision ^[37], but on iPhones, web browsers or other unsupervised devices the child is only partly supervised ^[38].

Instagram’s supervision shows whom the teen messaged in the last seven days, but parents “will not be able to see the content of the actual messages,” and message requests and deleted chats do not appear ^[39]. Instagram dropped end-to-end encryption for direct messages from May 8, 2026, so Meta can now see that content, and Meta pointed users who want encryption to WhatsApp ^[40]. Neither gives parents access.

Third-party monitoring apps fill some of the gap, with conditions. Google Play allows them only if they are built and marketed exclusively for parental or enterprise use and declare themselves, and it shows users a warning ^[41]. Bark, a U.S. service, says on its support page that it scans iPhones through a sync with a home computer or its own device and that alerts are not instantaneous ^[42].

Covert spyware is a different and riskier category. TechCrunch reported in July 2024 that mSpy suffered its third known data breach, with about 2.4 million customer email addresses exposed, and noted that users can face prosecution under wiretapping laws ^[43]; it has since reported exposures at Spyzie and Catwatchful ^[44][45]. A parent who installs such an app risks handing a child’s private life to strangers.

Kid-safe phones such as Gabb, Troomi and Pinwheel limit the device rather than watch it. Bark, which sells its own phone and so is a competitor, says in a comparison that Pinwheel “gives parents full access to kid’s texts” ^[46]. Confirm any such claim, and any price, on the maker’s own site; this desk has not tested these devices.

The Other Post has no affiliation with any company, firm or product named in this piece, including the security firms and services in “If Budget Is No Object,” and receives nothing for mentioning them. Nothing here is a ranking or endorsement.

Every tool above works only on the devices and accounts enrolled in it. Even a teenager who sincerely agrees to supervision can end up in a channel none of them covers, and moving a child to such a channel is a common step in grooming. What follows draws on news reports, not instructions.

Second phones and accounts. Controls follow the device and the account: Family Link only partly supervises other devices ^[38], and Instagram’s supervision covers only the linked account ^[39]. Reporting on New York City’s 2025 school phone ban described students carrying burner or second phones ^[47]. Police in Fairfax County, Virginia, and Kenner, Louisiana, have warned parents about apps disguised as calculators that can hide photos and chats, according to local news reports ^[48][49].

Documents as chat. Fox News and the New York Post have reported students turning shared Google Docs into chatrooms during school phone bans ^[50][51]. In 2012 the BBC, citing AP, reported that former CIA Director David Petraeus left messages in the drafts folder of a shared email account ^[52].

Games with chat. NBC News found in 2023 that adults had been prosecuted in 35 cases over six years for kidnapping, grooming or sexual assault allegedly involving communications on Discord, a chat service popular with gamers. The Canadian Centre for Child Protection told NBC that offenders meet children on Roblox or Minecraft and move them to Discord ^[53]. Bloomberg Businessweek reported in July 2024 that U.S. police had arrested at least two dozen people since 2018 accused of abducting or abusing victims met or groomed through Roblox; Roblox disputed parts of that reporting ^[54]. A federal judicial panel in December 2025 centralized private lawsuits against Roblox in one proceeding, MDL No. 3166; its order summarizes allegations that offenders made contact on Roblox and then moved children “to a second platform such as Discord, Snapchat, or Instagram, or via texting” ^[55]. These are allegations, not findings. Roblox says it has required age checks to chat since January 2026 ^[56].

The move itself. In a 2022 survey by Thorn, a U.S. nonprofit that builds tools against child sexual abuse, 65% of minors said someone they knew only online had invited them “to move from a public chat into a private conversation on a different platform” ^[57]. The new channel is, by design, one the parent’s tool does not watch.

Turkey. Turkish courts have blocked both platforms. On Aug. 7, 2024, a criminal judgeship of peace in Adana blocked Roblox at the request of the Adana prosecutor’s office, citing content that could lead to the abuse of children, according to Anadolu Agency ^[58]. On Oct. 9, 2024, a court in Ankara blocked Discord; Justice Minister Yılmaz Tunç cited “sufficient suspicion” of child sexual abuse and obscenity offenses, and the transport minister said the company had not shared information with Turkish security units ^[59]. Both blocks were still in place as of Oct. 21, 2026, according to a Turkish news report. The transport minister said on June 5, 2026 that Discord had met Turkey’s content-removal criteria, but no lifting has been announced, and in June the telecoms regulator’s registry still listed the Adana order against Roblox ^[60][61].

In this editor’s judgment, open monitoring still matters. But no parent can block every channel, and the warning sign that matters most, a request to move somewhere private, reaches a parent only if the teen decides to say so.

The research on monitoring is thinner than the marketing. A 2018 University of Central Florida study of 215 U.S. parent-and-teen pairs found that parental-control app use was associated with more, not fewer, online risks for teens, including unwanted explicit content, harassment and sexual solicitations ^[62]. That is an association, not cause: parents may install apps because trouble has already begun. In 736 children’s reviews of such apps on Google Play, about 79% gave two stars or fewer ^[62].

The American Academy of Pediatrics (AAP) favors family input in setting rules and warns that “relying on restrictive rules or monitoring technologies makes it harder for them to problem-solve and develop the autonomy they need” ^[63].

The trade-off is practical as well as emotional. The invitation Thorn found 65% of minors had received arrives where no parental tool is watching ^[57]; a parent learns of it only if the teen says so. Children are often the first to report sextortion ^[2], and a teen who has found a hidden app is less likely to come to the parent who installed it. NCMEC’s message to children is that “the blackmailer is to blame, not the child” ^[2]; it lands only in a home where it is believed.

Some families can pay for far more than an app: a professional network at home, phones managed like company devices, a security firm on call. Money does buy more visibility. But it buys visibility into the devices and networks a family controls, not into the channels described above, and the law sets the same limits for a wealthy parent as for anyone else.

In this editor’s judgment, the best use of a large budget is a disclosed engagement with a firm that can test the home and the family’s devices, test the school with the school’s written permission, and investigate if something goes wrong, with the teenager told first and the scope set in writing. Of the firms this desk reviewed, Kroll fits that brief best ^[64][65][66]. A home-network filter, set up with your teen’s knowledge, is the sensible base layer ^[67][68]. The five tiers below set out what each option sees, what it cannot and how to deploy it lawfully in the U.S. and Turkey.

TIER 1: THE HOME NETWORK

NextDNS is a filtering service that blocks categories and named apps and games, schedules “Recreation Time,” enforces SafeSearch and YouTube’s Restricted Mode, and keeps logs a parent can retain for anywhere from one hour to two years, or switch off, stored in the U.S., the EU, the UK or Switzerland ^[67]. Because it runs over encrypted DNS, it can work “on all devices and on all networks” when it is set up on the device itself, not only on home Wi-Fi ^[67]. Its Pro plan costs $1.99 a month or $19.90 a year “for personal and close family use,” and a free tier covers 300,000 queries a month ^[81].

Firewalla, a hardware firewall, sets rules by device, group, network segment and schedule, with categories such as gaming, social, porn and VPN; its help page gives the example of blocking YouTube on a child’s iPhone from 7 to 9 p.m. Rules for separate network segments need its Gold or Purple models ^[68]. This desk did not check hardware prices.

The limits are built into the technology. Firewalla notes that URL rules are “not possible without unwrapping https sessions,” and that doing so would “break end to end trust,” and that rules may not work for traffic that does not pass through the box ^[68]. A phone on cellular data or a friend’s Wi-Fi is outside a home device unless the filter also runs on the phone. Apple warns that network-based filtering and parental controls may conflict with iCloud Private Relay ^[69]. A network layer shows which services your teen reaches and when, never what they write.

Deploying it lawfully. Because it does not capture message content, a filter that logs only domains stays furthest from the interception rules in both countries ^[11][27]. Turkey’s data-protection law, KVKK, does not apply to processing by people within their own family living in the same dwelling, provided the data is not disclosed to third parties and kept secure; the EU’s GDPR likewise excludes “purely personal or household activity” ^[70][82]. Keep logs short, share them with no one and tell everyone who uses the network.

TIER 2: PHONES MANAGED LIKE COMPANY DEVICES

Mobile device management (MDM) is how companies run staff phones, and a family can use the same tools. Apple says “supervision generally denotes that the device is owned by the organization.” Supervising an iPhone manually with Apple Configurator requires the phone in hand, and “the device is erased and all data is lost.” A supervised iPhone then tells its user it is supervised and that the organization “can monitor your internet traffic and locate this device” ^[71]. Jamf Now, marketed as “MDM for Small Business” with web protection, content filtering and remote lock, warns that devices “will be wiped when bringing them under supervision” ^[73][72]. This desk did not confirm Jamf Now’s price on Jamf’s own pages. Jamf Parent is a different product: it manages only school-issued devices, and only during the time periods the school configures ^[83].

Deploying it lawfully. Supervision is disclosed by design: the phone shows the banner ^[71]. The parent who bought the phone owns it, but the photos, chats and accounts on it are your teen’s, so wipe it only after your teen has backed it up and agreed. In Turkey, a minor with the capacity to judge exercises strictly personal rights on their own (TMK 16) ^[28], and case law suggests that reading the messages of a child over 15 can carry criminal risk, with TCK 30/4 a possible excuse for a protective parent ^[29]. Supervision does not open your teen’s encrypted chats in any case. A school-issued device belongs to the school; manage it only through the school’s own system ^[83].

TIER 3: DISCLOSED MONITORING SERVICES

Bark charges $14 a month or $99 a year for its app, and its own phone plans start at $29 a month plus device payments ^[74]. It states that “Apple’s privacy protections mean third-party apps can’t read what’s on your child’s device,” and that “for some social media platforms, monitoring requires entering your child’s login credentials.” Its app is offered only in the U.S., South Africa and Australia, so it is not an option in Turkey ^[74]. Qustodio, based in Barcelona, lists its Complete plan at $104.95 a year on its English-language page; calls and messages monitoring on iOS needs a Windows or Mac computer, and message alerts for Instagram, TikTok, Snapchat and X are Android-only ^[75].

Deploying it lawfully. U.S. federal law and the Turkish Penal Code both make it a crime to access a computer or information system without authorization (18 U.S.C. § 1030; TCK 243) ^[76][27], and the Stored Communications Act covers unauthorized access to stored messages ^[12]. A parent who signs in to a teenager’s account behind the teen’s back is on the wrong side of the consent question. The defensible form is the teen entering their own password, knowing what is scanned and who sees the alerts.

TIER 4: WORLD-CLASS SECURITY FIRMS

A family that wants professionals to test its home and its children’s school, with everyone’s agreement, needs a firm that does both and can investigate what it finds. In this editor’s judgment, two fit, and Kroll fits best.

1. Kroll. It runs a family office and private client service, offers security consulting for “specialty schools, college preparatory schools, colleges and universities,” and says its threat and vulnerability assessments are “routinely conducted at corporate headquarters, executive residences” and schools ^[64][65]. CREST accredits it for penetration testing and incident response, and lists “over 700 cyber experts across 19 countries” ^[84]; its staff include former FBI agents and forensic computer scientists ^[65]. Its incident-response unit works 24x7 and follows a “strict chain-of-custody protocol” ^[66]. Why it wins: one firm covers house, school and any investigation that follows. Limitation: it has no page for home-network testing, so write those tests into the contract.

2. Stroz Friedberg, a LevelBlue company. It offers to “test home networks, monitor credentials, and maintain protection over time” ^[77], and its testers write that an executive’s “initial foothold” for attackers “is often obtained by targeting devices used by their family members” ^[85]. It works with “K-12 systems safeguarding minors’ data” ^[86], is CREST-accredited in the UK ^[87] and offers 24/7 incident response with “no retainer required” ^[79]. Why it wins: the most detailed published home-network testing. Limitation: it lists no office in Turkey ^[88].

Also consider: Control Risks, for a family visibility profile built from open sources “without any form of hacking or intrusion” ^[89]; S-RM, London, for private client work that includes digital forensics and 24/7 kidnap, ransom and extortion response ^[90]; BlackCloak, U.S. home and device protection for whole families, which says “we never monitor their internet activity, accounts, or the content of their messages” ^[78].

None of these firms publishes its prices on the pages this desk opened.

None of the five publishes prices or a child-safeguarding protocol on the pages this desk opened, so ask for a written quote and a written protocol before any work begins.

Deploying it lawfully. A firm tests systems only with the authorization of whoever controls them: the parents for the home, school leadership for the school. Testing without it is the conduct TCK 243 and § 1030 address ^[27][76]. Once findings leave the household, the family exemptions in KVKK and GDPR no longer cover them, and the firm handles the family’s data under KVKK’s consent and security duties or the GDPR ^[70][82]. Children’s school records fall under FERPA in the U.S., where a parent’s rights pass to the student at 18 or on entering college ^[91]. England’s guidance requires schools to set safeguarding requirements in contracts and to check contractors who work there ^[92].

TIER 5: WHEN SOMETHING HAS HAPPENED

If abuse or a crime has already happened, the order matters. In the U.S., report to NCMEC and the FBI; in Turkey, to the police or the prosecutor ^{[93][3][94][95]}. A forensic firm supports that work and does not replace it. Kroll offers digital forensics, onsite response “within hours” and expert testimony ^[66]; LevelBlue’s Stroz Friedberg unit says it has investigated more than 9,000 incidents and will “maintain chain of custody of evidence” ^[79]; S-RM says it helps families “preserve critical digital evidence” ^[90]. Tell any firm in writing, before it starts, that if it finds sexual images of a minor it stops, does not copy them and goes to the police; U.S. law treats possession seriously, with only a narrow defense for prompt reporting or destruction ^[96][80]. In Turkey, findings obtained unlawfully are barred as evidence, so how the device was examined matters as much as what was found ^[26].

DEPLOYING ANY TIER LAWFULLY: THE CHECKLIST

The Other Post has no affiliation with any company, firm, product or service named in this section, including Kroll, Stroz Friedberg and LevelBlue, Control Risks, S-RM, BlackCloak, NextDNS, Firewalla, Apple, Jamf, Bark and Qustodio, and receives nothing for mentioning them. Nothing here is a ranking or endorsement beyond the editor’s stated judgment.

WHAT MONEY CANNOT BUY

Every tier above stops at the same border. A firewall sees the home network, MDM sees the managed phone, a monitoring service sees what it can scan and a world-class firm sees what its written scope allows. None sees the second phone, the friend’s laptop, the shared document or the game chat your teen opens somewhere else, and the law in both countries keeps the most intrusive options off the table for good reason. In this editor’s judgment, money buys visibility into devices and networks, not into the channels outside them. Only your teen’s trust covers every channel, and the best professional engagement is the one your teen knows about, agreed to and would tell you about if something went wrong.

Do not let it go. Keep a close eye, openly and by agreement. But total visibility is not on offer: the technology does not provide it, the law in both countries limits it and the channels above sit outside it. In this editor’s judgment, the monitoring that works best is the kind your teen knows about and agreed to, because it builds the habit of telling a parent early, the only safeguard that follows a teenager into a game chat, a shared document or a friend’s phone.

BEST MOVES

WHAT TO WATCH

IF IT HAPPENS: THE FIRST HOUR

Drawn from NCMEC, the U.S. Administration for Children and Families and Turkish official channels:

Watch openly, watch closely and keep the rules in writing. Then accept what the tools cannot do. No setting covers a second phone, a shared document or a game’s chat, and no law gives a parent every message. In this editor’s judgment, what covers every channel is a teenager who knows they can tell you anything without being punished for it. The goal is that they come to you before a problem becomes irreversible.

This is general information, not legal advice. U.S. law differs by state, especially on recording and on parental consent, and Turkish law is applied case by case. Consult a lawyer where you live before acting on any of it.

The Turkish Parliament’s 1995 Aegean text never uses the word “war,” yet abroad it goes by a Latin label meaning “occasion of war.”

ABOUT 6 MINUTES

On Tuesday, Oct. 20, after a speech at the European Policy Center in Brussels, Foreign Minister Hakan Fidan took three questions from the Greek newspaper TA NEA. One was about the casus belli. His answer, as To Vima reported it in English the next day, started with the label itself: “Regarding the casus belli, many consider it a threat of war. That is the Greek interpretation. We, however, have not characterized it as a casus belli.” Then: “It is a declaration by the Turkish Parliament from 1995, not a law.” And later: “This preventive declaration has helped preserve peace to this day.”^[99]

Athens has its own sentence for the same text. Greece’s Foreign Ministry says the 1995 resolution grants the Turkish government “full and perpetual competence to declare war (casus belli)” should Greece extend its territorial waters beyond six nautical miles in the Aegean^[103]. In 2011 a ministry spokesman called it “the familiar casus belli” and said “retraction of the threat of war” was “an obligation on the part of Turkey.”^[104]

Note the chain before we start. The English quoted here is To Vima’s, reporting what TA NEA was told; we have not seen a transcript in the language Fidan spoke. And the phrase at the center of it all is neither Greek nor Turkish nor English. It is Latin.

Same text, two readings. Both capitals describe one paragraph adopted on June 8, 1995. Neither description quotes it. So, before any label: what does the paragraph say, in its own Turkish?

The Turkish Grand National Assembly adopted it unanimously. It calls itself an açıklama: a statement, or more literally an explanation. The TDK dictionary defines açıklama as “the act of explaining,” with izah and tavzih as synonyms^[100][107]. Its operative sentence hopes that Greece will not take its waters beyond six miles “Lozanla kurulmuş dengeyi bozacak biçimde,” in a way that would upset the balance set at Lausanne. Should it do so, the Assembly gives the government “tüm yetkiler,” all powers, “askerî bakımdan gerekli görülecek olanlar da dahil olmak üzere,” including those deemed necessary militarily. It ends by resolving to make this known to the Greek and world public “dostane duygularla,” with friendly feelings^[100].

The word savaş, war, does not appear. Nor does harp, its older synonym, and nor does “casus belli.” I checked the published minutes of the sitting for all three^[100]. The label is not in the text. It is what the text is called.

Ankara’s own words for it have moved over the years. In October 2018, the Turkish Foreign Ministry spokesman said in Turkish: “Yüce Meclisimizin 8 Haziran 1995 tarihli bildirisi bu bağlamda gerekli siyasi ikazı havi olup, geçerliliğini bugün de korumaktadır.”^[101] The ministry’s English version reads: the declaration “contains a necessary political warning and is still in force today.”^[102]

That one Turkish sentence runs through three registers. Bildiri is the plain modern word: TDK defines it as a text written by an official body to make a situation known to those concerned, and gives it no foreign origin^[107]. İkaz is Arabic, from īḳāẓ, and TDK glosses it two ways: “warning” and “awakening”^[107]. Havi, also Arabic, from ḥāvī, means “containing”^[107]. The English is accurate and loses two things: the stiff, old-Istanbul-flavored legal register, and the idea that a warning is also a wake-up.

So on the Turkish side the text has been an açıklama (1995), a bildiri that carries an ikaz (2018) and, in To Vima’s English, a “preventive declaration” (2026)^{[99][100][101]}. On the Greek side, in its ministry’s English, it has been “the casus belli” throughout^[103][104]. The two vocabularies do not meet.

Merriam-Webster defines a casus belli as “an event or action that justifies or allegedly justifies a war or conflict.” It marks the phrase as New Latin, “occasion of war,” and dates its first known use in English to about 1841^[105]. Etymonline gives 1840 and splits it: casus, “case,” from a root meaning “a falling,” the notion being “that which falls” as “that which happens”; and belli, the genitive of bellum, war^[106].

In the dictionary sense, then, a casus belli is an occurrence: the thing that happens. In the 1995 case, that occurrence would be an extension of Greek waters beyond six miles. Over time the label has also come to name the text that points at the occurrence. English holds both uses, and the article gives them away. “A casus belli” is an event. “The casus belli” has become a name. A careful listener can hear which one a speaker means, and neither use is wrong as English.

The dictionary also carries its own warning in two words: “allegedly justifies.” The phrase has always been a claim about a cause, not the cause itself.

Then there is my own ear. When I meet the phrase in a Turkish headline, my first reading is not Latin at all.

Casus, with the Turkish c sounded like the English j, is a spy. TDK defines it as a person who works for the secret aims of a state or organization, gives its origin as Arabic cāsūs and lists casus kamera and casus yazılım, spy camera and spyware^[107]. Belli means known, evident, out in the open. It also means “waisted,” as in ince belli, slender-waisted^[107].

So “casus belli,” met cold by a Turkish speaker, reads as “the spy is obvious.” On a lighter day, “the waisted spy.” It is a false friend: a phrase about the occasion for war that half-rings, in Turkish, as a spy who has been found out.

Put the two languages side by side and the phrase says something else again. Latin casus gave English “case”^[106]. Turkish belli means “obvious”^[107]. Read across the border, casus belli is an obvious case.

That is the trap. Each side treats its reading as the obvious one.

None of this settles what the 1995 text means in law. That is for the two governments, and both have said plainly where they stand^[101][103]. What it offers a diplomat is a map of the words.

In Ankara’s Turkish-language record, the text is a statement, a bildiri and a political ikaz. In Athens’s English-language record, it is the casus belli. The Latin works as a term of art in a Brussels corridor, but it is not the word the Turkish side uses for its own text, and Fidan said so directly^[99]. An interpreter who renders ikaz as “warning” is right, and the room still loses the wake-up in it. An aide who briefs “the casus belli” has already picked a reading, whether or not they meant to.

The useful habit is small. Ask which document is meant. Ask which word is on the page, and in which language. Then ask what it was meant to do: announce, warn, deter or threaten. The answers can differ by capital, and both sides will be sincere.

Everyone hears the same, but gets it differently. Even for the most presumedly obvious cases, it may be a matter of war and peace to ask the obvious question: what does it really mean, and why?

Sürç-i lisan eylediysek…

All sources accessed Nov. 1, 2026. Translations from Turkish are The Other Post’s own.

SOURCES

The Other Post: The Daily Briefing for the Conscientious Leader, through an Istanbul lens is independent analysis for decision-makers. Nothing here is investment, legal or medical advice. Special, Sunday, Nov. 1, 2026 · © 2026 The Other Post