Part I — Situation overview

Anita Orbán, deputy prime minister and foreign minister, announced on the morning of 8 September 2026 that on her instruction the competent head of the Ministry of Foreign Affairs had informed Russia’s ambassador in Budapest: in the assessment of the Hungarian authorities, ten staff members of the Russian mission had pursued activity which, under the Vienna Convention on Diplomatic Relations, is not acceptable from a diplomat. The Hungarian side called on those concerned to leave the country. The minister emphasised that the decision does not mean breaking off diplomatic relations with Moscow. The step was taken within governmental — more precisely, ministerial — competence: it is not an act of the head of state, and it does not require a parliamentary decision. The Russian foreign ministry reacted the same day: the spokesperson Maria Zakharova promised a “harsh and painful” response, while the Russian embassy in Budapest held out the prospect of a “proportionate response”.

The weight of the announcement comes from the historical context. András Rácz, deputy director-general for research and academic affairs at the Hungarian Institute of International Affairs, recorded in his assessment given to MTI that “activity incompatible with diplomatic status” traditionally means intelligence activity in professional language, and that in the roughly four and a half years since the start of the Russian–Ukrainian war more than 600 Russian intelligence officers working under diplomatic cover have been expelled from the member states of the European Union. Hungary has so far stayed out of this wave. As part of the coordinated western action following the poisoning of the former Russian intelligence officer Sergei Skripal, a single Russian diplomat departed in 2018. And in 2021 — when the Czech Republic held the Russian secret services responsible for an ammunition depot explosion with two fatalities — Hungary was the only one of the Visegrád countries not to expel anyone. According to Rácz, 47 Russian diplomats are currently accredited in Hungary on the official list of the Ministry of Foreign Affairs, accompanied by a further 41 spouses; a Russian investigative portal was able to link 15 of them to one or another Russian secret service on the basis of public data. According to Telex’s report, after the Tisza Party’s election victory in April there was already a quiet expulsion in May, but it carried no public political message.

MIAK’s reading is as follows. The decision is defensible in substance, and from the point of view of the allied yardstick it was long overdue: an intelligence presence operating on the territory of a member state does not stop at the national border, and operations carried out in the Schengen area affect the security of the neighbours too. The character of the problem, however, is not whether the step was taken, but whether there is a rule behind it. An expulsion is in itself a single event: its predecessor could have decided with the same competence not to act, and its successor may likewise decide in either direction. A foreign policy in which a decision of this weight depends on the discretion of the minister of the day remains just as unpredictable as one that did not act for years — only the sign has changed.

Part II — Foundations in the literature

Behind the facts stand three complementary conceptual frames. Geoff Berridge, Maurice Keens-Soper and Thomas Otte, in their volume working up the classic texts of diplomatic theory (Diplomatic Theory from Machiavelli to Kissinger, 2001), quote Abraham de Wicquefort’s seventeenth-century definition, according to which the position of the ambassador is that of the “honourable spy” — obtaining information is not a by-product of diplomacy but a basic function of it; while François de Callières, the other classic author treated in the volume, also explains why states tolerate this on their own territory, and where the limit of tolerance runs. Henry Kissinger, former secretary of state of the United States and one of the defining authors of the theory of international order, writing about cyberspace in his volume World Order (2014), warns that there the threat is hard to identify, the identity of the attacker can be denied, and the proportionality of the response cannot be determined on the model of traditional deterrence either — this bears directly on the question of what remains of the intelligence risk after the physical presence has been dismantled. And the textbook Foreign Policy and Diplomacy by Valentin Naumescu and Dan Petrut lists item by item the functions of a diplomatic mission under the Vienna Convention, among them that the mission should ascertain “by all lawful means” the processes taking place in the receiving state — this phrase “lawful means” is the legal handhold along which permitted and non-permitted activity part company. The detailed treatment of the literature — author by author, with quotations — can be found in section 6.4 Literature in detail.

Part III — MIAK’s concrete proposal

MIAK proposes three measurable measures.

3.1 A public foreign policy doctrine document (by the submission of the 2027 budget)

Programme point KP12 — the public documentation of foreign policy philosophy — is precisely the missing element that would turn the present decision from an event into a policy. MIAK proposes that by the submission of the 2027 budget the Ministry of Foreign Affairs publish a concise doctrine document of at most thirty pages answering three questions. First: on the basis of what set of criteria does a foreign diplomatic presence qualify as an intelligence risk, and by what procedural route does this assessment reach the decision-maker. Second: which commitments does Hungary regard as irrevocable — the obligations arising from EU and NATO membership, allied information sharing — and on which questions does it maintain independent room for manoeuvre. Third: at what intervals does it review these principles, and who is entitled to initiate the review. The value of the document lies not in the specific positions it contains but in the fact that it makes them answerable for afterwards: the KP4 principled pragmatism works if the principles are written down. The phrase “by all lawful means” quoted by Naumescu and Petrut (see 6.4.3) shows that the dividing line is given in law — the task of the doctrine is to fix the Hungarian interpretation of it as well.

3.2 A public summary of the retaliation scenario (within 30 days)

According to the crisis management protocol of programme point KP7, foreign policy decisions must be preceded by scenario analysis. MIAK proposes that within thirty days of the announcement the government publish a public summary of the preparation made for the expected Russian countermeasures. It has to cover three channels: the diplomatic mirror step — how many departing Hungarian diplomats the government reckons with, and which consular services this affects in terms of availability for Hungarian citizens living in Moscow; energy-side exposure — the current share of natural gas and crude oil imports from Russian sources, together with the price shock scenario for which the government has prepared; and contractual economic exposure, including ongoing investments with Russian participation. A public summary does not require the release of classified details: naming the risk categories and indicating the state of preparedness is sufficient. Withholding sensitive details is legitimate; concealing whether any analysis was made at all is not.

3.3 A cyber-defence surface survey after the reduction of the physical presence (within 90 days)

The departure of the ten diplomats reduces the physical intelligence presence, but not the digital attack surface. MIAK therefore proposes that within ninety days, in the framework of the D5 cybersecurity strategy, a targeted survey be prepared on which state systems have been the targets of confirmed or probable foreign intrusion in the past five years, and in which of these the remedying of the vulnerabilities has been completed. Let the aggregated result of the survey — without institution names and technical details — be published, because without it the citizen cannot judge whether the diplomatic step has brought a real improvement in security, or has merely displaced the risk. Related to this, the HV11 strategic communication programme point fixes the task that the public explanation of the state’s response to hybrid threats is itself part of defence: the population’s being informed is itself a defence capability. Kissinger’s warning about the unidentifiability of the source of the threat (see 6.4.2) means precisely that in cyberspace preparedness cannot be demonstrated by a single spectacular step — only by continuous, documented work.

The three proposals are bound together by the same principle: the credibility of a foreign policy decision comes not from the firmness of the announcement but from whether there is a written rule, a risk analysis and a subsequently verifiable preparation behind it. The doctrine says on what criterion we decide. The scenario says what we have prepared for. The cyber-defence survey says how much safer we have become. None of them restricts the government in what foreign policy it pursues — only in doing so without explanation.

Part IV — Expected effects and risks

Dimension Expected effect Risk
Foreign policy Hungary aligns with allied practice, which may bring a measurable improvement even in the short run in the confidence feedback of NATO and EU partners; one node of operations carried out in the Schengen area narrows If no doctrine follows, the step will qualify as a government-dependent gesture, and partners will attribute to it a lifespan extending to the next change of government; the narrowing of diplomatic channels makes crisis communication harder
Energy security A public inventory of exposure is in itself disciplining: diversification decisions gain a measurable basis The Russian side may respond with retaliation along the energy supply and investment threads; the effect is most sensitive at the start of the heating season, and may appear in household prices too
National security and cyberspace Intelligence capacity operating under diplomatic cover is reduced, and physical operational opportunities narrow The capability may partly relocate into cyberspace and to actors in third countries, where identification is harder; the spectacular step may create a false sense of security if it is not accompanied by a survey of the digital surface

The main tension of the package runs between firmness and predictability. In handling an intelligence risk, speed is a real value: long deliberation is itself a risk. At the same time every firm step to which no criterion fixed in advance is attached later turns into political room for manoeuvre — and the next government may use the same freedom in the opposite direction. The proposal tips towards the risk side if the doctrine document is not produced, and the expulsion remains a stand-alone, unrepeatable episode. The second question for consideration is the extent of publicity: national security sensitivity is a real limit, and MIAK is not asking for the release of details. Naming the risk categories and indicating the state of preparedness, however, does not injure any national security interest, and at the same time makes it possible to settle the debate about whether any analysis was made at all.

Part V — Measurability and summary

5.1 What is worth following? (suggested KPIs)

MIAK proposes four performance indicators (KPIs, in full: Key Performance Indicators) from which it will be visible in 6, 12 and 24 months whether the decision has become a policy:

  • The appearance of the public foreign policy doctrine document: the suggested deadline is the submission of the 2027 budget. The existence or absence of the document is in itself an easily followed indicator.
  • The number of Russian diplomats accredited in Hungary: against the present starting value of 47, it is worth following where the headcount settles after the expulsion, and whether there is replenishment. The figure is public from the official list of the Ministry of Foreign Affairs.
  • The share of natural gas and crude oil imports from Russian sources: the suggested target is quarterly public publication of the diversification path, so that the extent of the retaliation risk can be followed in figures.
  • The first edition of the KP23 alliance credibility audit: it is worth following whether the annual review is produced that sets out item by item the divergences between Hungarian positions and allied commitments.

5.2 Summary

MIAK’s request in a single sentence: let the government publish the rule on the basis of which it took this decision. Concretely: a doctrine document by the submission of the 2027 budget, a public summary of the retaliation scenario within thirty days, a cyber-defence surface survey within ninety days. None of these requests is about the government deciding differently — they are about the decision being judgeable afterwards.

Two MIAK foundational values are in play here. One is being ideology-free: MIAK applies the same yardstick to the present decision that it applied to the previous one. The fact that in recent years Hungarian foreign policy failed to act stemmed from the same absence as the firm step does now — there was no written criterion, only discretion. The right direction does not make the rule superfluous. The other is accountability: the national security justification of an expulsion legitimately remains classified, but the question whether the government has prepared for the consequences is not a national security matter but one of government performance — and to that the citizen has a right to an answer.

Part VI — Reasoning and further sources

6.1 The framing of the press by spectrum

The liberal-left band framed the news in terms of historical omission. 444.hu tied the announcement directly to the practice of earlier years: it recalled that Russian intelligence had also worked through the IT system of the Hungarian foreign ministry, and that even three days after the election a Russian diplomat linked to the secret services had visited the ministry. Telex, by contrast, made international comparison the backbone of its article: it arranged the Hungarian response to the 2018 Skripal affair, the 2021 Czech case and the post-2022 European wave of expulsions into a single time series, and indicated that the Hungarian side stayed out of the wave throughout. Both framings derive the significance of the present step from the missing antecedents — the emphasis is on the question of “why only now”.

The economic and public affairs band, by contrast, concentrated on expert interpretation and on consequences. Portfolio published the facts of the announcement without commentary, highlighting that the step does not mean breaking off relations. 24.hu brought two expert statements as well: András Rácz’s interpretation of the rules of diplomatic language and of the accredited headcount of 47, and the assessment of the former foreign minister Péter Balázs, according to which a strong signal had been given. This framing is the most informative, because instead of the political content of the announcement it makes its operational consequences — headcount, deadline, expected mirror step — examinable.

The conservative band presented the consequences from the side of risk. Magyar Nemzet highlighted the statement of the secret service expert László Földi, according to whom the Russian response is “practically certain”, may mean the sending home of at least ten Hungarian diplomats, and may be followed by economic countermeasures alongside the diplomatic response — with a separate emphasis on Russian energy imports and on investments implemented with Russian participation. By MIAK’s yardstick of being ideology-free this consideration is substantive and must be answered: energy-side exposure is indeed asymmetric, and the risk does not disappear because the step was justified. MIAK’s answer to this is that it is precisely for this reason that the retaliation scenario has to be made public: presenting the exposure in figures does not weaken but strengthens the position, because it shows that the decision-maker was aware of the price. MIAK does not regard the description of a “traitorous government” formulated in the article as a policy statement, and does not respond to it — the energy security concern behind it, however, is well founded in itself.

6.2 Facts and data

Data Value Source
The day of the announcement 8 September 2026 24.hu, 8 September 2026
The number of Russian diplomats concerned 10 persons Portfolio, 8 September 2026
The legal frame invoked the Vienna Convention on Diplomatic Relations (1961) communication of the Ministry of Foreign Affairs, 8 September 2026
Number of Russian diplomats accredited in Hungary 47 persons, with a further 41 accompanying spouses András Rácz, MTI / 24.hu, 8 September 2026
Diplomats identified as linkable to a secret service 15 of the 47 (from public sources, analysis by a Russian investigative portal) András Rácz, MTI / 24.hu, 8 September 2026
Russian intelligence officers expelled from the European Union since the start of the war more than 600 persons, in about 4.5 years András Rácz, MTI / 24.hu, 8 September 2026
Expulsions from Europe and North America in the first half-year of the war nearly 600 persons Telex, 8 September 2026
Earlier Hungarian expulsion after the Skripal affair 1 person, 2018 Telex / 24.hu, 8 September 2026
Hungarian participation in the action following the 2021 Czech case no expulsion; the only one among the Visegrád countries 24.hu, 8 September 2026
Expulsion in May 2026 1 person (third secretary) and spouse 24.hu, 8 September 2026
Russian reaction “harsh and painful response” (Maria Zakharova, foreign ministry spokesperson); the embassy promised a “proportionate response” 444.hu / Telex, 8 September 2026

Two lines of the table require a separate note. The European Union aggregate of more than 600 and the figure of nearly 600 for Europe and North America in the first half-year of the war relate to different time spans and different geographical scopes, so they cannot be added together and are not directly comparable either; the table gives both with their own source attribution. And the figure of 15 relating to secret service ties is the result of external analysis carried out from public sources, not an official finding — the ten persons expelled and this circle are not necessarily identical.

6.3 Policy dimensions

  • Foreign policy (programme points) — the doctrinal foundation of the decision and crisis management: KP4 (principled pragmatism), KP7 (foreign policy crisis management protocol), KP11 (strategic balance policy), KP12 (documentation of foreign policy philosophy) and KP23 (alliance credibility audit) together provide the skeleton of the proposal;
  • Defence (programme points) — the quality of the public explanation given to hybrid threats: HV11 (strategic communication and information defence) and HV7 (societal defence resilience) provide the frame in which informing the population is itself a defence capability;
  • Digitalisation and AI regulation (programme points) — the attack surface remaining after the physical presence ends: D5 (cybersecurity strategy) is the frame of the proposed survey, while D2 (open data programme) is the form of publication of the aggregated result.

6.4 Literature in detail

6.4.1 Berridge and his co-authors: Diplomatic Theory from Machiavelli to Kissinger

The volume works up the classics of diplomatic theory, and from the point of view of the topic it dwells at greater length on two authors. Abraham de Wicquefort, the seventeenth-century Dutch envoy and political news provider — who wrote his main work in prison — characterised the position of the ambassador as that of an “honourable spy”: obtaining information is not a by-product of the diplomatic role but an essential part of it. François de Callières supplements this observation by explaining why states nevertheless tolerate this activity on their own territory. According to the volume’s summary, in Callières’s reasoning

“the true basis of the immunity of envoys is reason of state. As soon as a sovereign takes it upon himself to punish the excesses of the agents of another sovereign, his acts threaten an institution in which he himself — together with the other sovereigns — has a vital interest.”

The core of the thought is therefore not that the diplomat is inviolable, but that immunity is an institution existing out of mutual interest and capable of bearing a load — which, however, is not unlimited. In this frame the Hungarian decision is not a violation of the institution of immunity but the enforcement of the limit that the institution itself presupposes: the receiving state maintains the immunity but removes the particular person. That is precisely why it matters that the place of the limit should be fixed in advance — because the institution works predictably only if it is not day-to-day discretion that decides where it runs.

📖 Source: Geoff Berridge – Maurice Keens-Soper – Thomas Otte: Diplomatic Theory from Machiavelli to Kissinger

6.4.2 Henry Kissinger: World Order

The ninth chapter of Kissinger’s volume treats the relationship between technology and the international balance, and in its part on cyberspace it warns that this domain falls outside the frames of traditional calculations of power. Before the cyber age a country’s capabilities could be assessed from headcount, equipment, geography and economic strength; the role of intelligence services was confined mainly to assessing the opponent’s capabilities, and their activity was framed by tacit norms of conduct. In cyberspace this structure breaks down:

“Threats emanating from cyberspace are nebulous and undefined, and it may be difficult to attribute them to anyone.”

Kissinger adds that attack is easier than defence, which from the outset pushes capability development in an offensive direction, and that the symmetrical logic of nuclear deterrence cannot be applied here, because “responding in kind” may have entirely different consequences in the case of two countries.

From the point of view of the Hungarian decision this passage is the most important limiting consideration. The departure of ten diplomats reduces the physical intelligence presence, and this is a real security gain. It does not, however, touch the digital surface — there the source of the threat continues to be hard to identify, and presence does not depend on how many people work in an embassy building. This argument justifies a cyber-defence survey going with the diplomatic step: without it the step displaces part of the risk rather than eliminating it.

📖 Source: Henry Kissinger: World Order

6.4.3 Naumescu and Petrut: Foreign Policy and Diplomacy

The textbook lists the functions of a diplomatic mission on the basis of the Vienna Convention: representing the interests of the sending state and its citizens within the framework of international law, negotiating with the government of the receiving state, ascertaining and reporting on the processes taking place in the receiving state, and developing bilateral relations. The formulation of the third function is decisive:

“ascertaining by all lawful means conditions and developments in the receiving State, and reporting thereon to the Government of the sending State”

The stipulation of “lawful means” is not an ornamental element but the dividing line itself. Obtaining information is a recognised task of the mission — which is why it is no argument to say that “diplomats gather information”. The difference between activity under the convention and an operation infringing the law of the receiving state lies in the means, not in the aim.

This distinction explains why expulsions are not usually justified in detail: the justification would bring with it the disclosure of detection methods. At the same time this is precisely why a set of criteria fixed in advance is needed. If in the particular case there can be no public argument, then the credibility of the decision comes from the decision rule itself being known and constant — this is what MIAK’s doctrine proposal would serve.

📖 Source: Valentin Naumescu – Dan Petrut: Foreign Policy and Diplomacy

6.5 International comparison

The post-2022 European wave of expulsions shows two different patterns. One is the crisis-reaction model: the state responds to a particular incident that has come to light — a poisoning, an explosion, a network intrusion — and the step is legitimated by the event itself. The Czech case of 2021 falls into this category. The other is the review model: the state bases itself not on a single incident but on a regular, periodic assessment of the foreign intelligence capacity operating on its territory, and the decision follows from the result of the assessment. The latter is more predictable, because it does not depend on whether a suitable incident comes to light — and the announced justification of the Hungarian decision also stands closer to this model, since it refers to the official assessment as the conclusion of a longer process.

The particularity of the Hungarian situation is late entry. According to András Rácz’s assessment, Hungary was the only EU country in which an expulsion on this scale had not taken place until now, and the Budapest presence may also have served as a base for operations carried out in other countries of the Schengen area — as a similar pattern was documented earlier in the case of Prague. This delay is a burden on the side of allied trust, which a single step does not discharge. The KP23 alliance credibility audit proposal is therefore not symbolic: the regular, itemised presentation of divergences is the only way for partners to perceive the change not as a one-off gesture but as a lasting direction.

In handling the risk of energy-side retaliation, the practice of the Baltic states and Poland offers a basis for comparison: there the reduction of import dependence and diplomatic action proceeded in parallel, and the governments published exposure data regularly. The lesson of the sequence is that the size of diplomatic room for manoeuvre is not independent of energy market exposure: a public exposure time series is not an admission of weakness but an honest reckoning of the price of the decision.

Foreign policy

  • KP4 — Principled pragmatism doctrine
  • KP7 — Foreign policy crisis management protocol
  • KP11 — Strategic balance policy
  • KP12 — Documentation of foreign policy philosophy
  • KP23 — Alliance credibility audit

Defence

  • HV7 — Societal defence resilience
  • HV11 — Strategic communication and information defence

Digitalisation and AI regulation

  • D2 — Open data programme
  • D5 — Cybersecurity strategy

Suggested new programme point: Retaliation preparedness summary — for the Foreign policy area: every foreign policy decision that provokes a foreseeable countermeasure should be accompanied by a public preparedness summary broken down by risk category, without the release of classified details.

6.7 List of sources

Press sources (MIAK press monitor, 9 September 2026 — topic 1):

Knowledge base references (specialist books):

  • 📖 Geoff Berridge – Maurice Keens-Soper – Thomas Otte: Diplomatic Theory from Machiavelli to Kissinger
  • 📖 Henry Kissinger: World Order
  • 📖 Valentin Naumescu – Dan Petrut: Foreign Policy and Diplomacy

MIAK internal materials:

  • MIAK policy area: Foreign policy (programme points; programme point ID: KP4, KP7, KP11, KP12, KP23)
  • MIAK policy area: Defence (programme points; programme point ID: HV7, HV11)
  • MIAK policy area: Digitalisation and AI regulation (programme points; programme point ID: D2, D5)
  • MIAK press monitor, 9 September 2026 — topic 1, score: 95/100

Supplementary public data sources:

  • The Vienna Convention on Diplomatic Relations (1961) — the functions of a diplomatic mission and the regime of declaring a person persona non grata
  • The list of the accredited diplomatic corps of the Ministry of Foreign Affairs — headcount data of the Russian mission

Generation metadata