by Zoran Nechev
Ursula von der Leyen switched to French when her 2026 State of the Union address reached enlargement. The passage was short and easy to miss among the heatwave plan, the Kids Act and the standing ovation for Mark Carney. It was also her clearest signal on enlargement.
Language choices in a speech like this are rarely accidental. France is where the next enlargement is most likely to be won or lost. Under Article 88-5 of the French Constitution, a new accession treaty must go to a referendum unless both chambers of Parliament choose a vote in Congress, where a three-fifths majority is needed. In France, as in the Netherlands, voters have grown warier of enlargement. By speaking French, the Commission President was addressing more than the candidate countries and the MEPs in Strasbourg. She was speaking to the people who will decide in the end whether Montenegro or Albania joins.
Read as a whole, though, the speech did more than repeat the enlargement formula she has used since 2023: their future lies in our Union, the process is merit-based, and values are non-negotiable. It sketched a Europe of concentric circles. The idea has deep French roots, from Jacques Delors and François Mitterrand’s European Confederation to Emmanuel Macron’s European Political Community.
The picture comes closest to the four tiers proposed in 2023 by the Franco-German Group of Twelve in its report Sailing on High Seas: an inner circle, the EU, associate members and the European Political Community. There, the associate tier was meant for the EEA countries, Switzerland “or even the UK,” built around the single market and kept apart from accession, “as membership in them can be permanent.” Von der Leyen has now stretched that outer circle across the Atlantic.
The “associate” idea has, however, travelled in two directions. In the concentric-circles reading, it is an alternative to membership. In another reading, it is a step towards it: only four months ago, Chancellor Friedrich Merz proposed associate membership for Ukraine as an interim stage on the road to full membership. The offer to Canada, a country that will never join, clearly belongs to the first reading. How these circles relate to one another, and whether a country can move from an outer circle towards the core, will shape enlargement over the coming years.
The enlargement passage: merit, values and roadmaps
The passage itself was brief. The EU’s neighbours, from Ukraine and Moldova to the Western Balkans, show that the European idea is alive. “Because their future lies within our Union, we will soon be proposing roadmaps for the candidates that have made most progress. This process will continue to be based on merit. And […] respect for our values is non-negotiable.”
Each element answers a worry that is strongest in Paris and The Hague. “Merit” answers the fear of geopolitical shortcuts. “Values” answers the fear of importing rule of law problems the Union still struggles with at home. She then gave the point an example: the images from Ratko Mladić’s funeral in Belgrade, which Serbian ministers attended, and her statement that glorifying war criminals “simply has no place in the European Union.” But “non-negotiable” has to mean consistent: rewarding each candidate’s real progress, sanctioning regression when it happens, and holding member states that flout the same values to account. Otherwise the Union loses the authority to demand them of others.
The only new commitment was the roadmaps, and they matter. Montenegro has opened all 33 negotiating chapters and provisionally closed 18. Albania has opened all 33 and closed three. Both are leaving the phase of opening negotiations and entering the much harder one of finishing them. After that, an accession treaty still has to pass 27 national ratifications. A roadmap that is only a checklist of closing benchmarks will add little. A useful one would set out the endgame: what the candidate still has to deliver, which safeguards apply after accession, what the EU reforms on its own side, and how ratification will be prepared. As Christophe Hillion argues, candidates should also have a seat in the new Congress of Europe that von der Leyen announced, as the candidate countries of the early 2000s had in the Convention on the Future of Europe. Those who will live under the reformed Union should help shape it.
From Kyiv to Ottawa: the life of “associate membership” so far
The Canada announcement cannot be read apart from what came before it. In May, Merz wrote to the EU leadership proposing “associate membership” for Ukraine. It would give Ukraine seats without votes in the European Council and the Council, a Commissioner without a portfolio, non-voting MEPs, an associate judge, gradual access to the EU budget and the protection of Article 42(7). A snap-back mechanism would freeze the status if Kyiv backslid on the rule of law. Merz described it as an interim step alongside accession negotiations, “not membership-lite.” For Moldova and the Western Balkans, he proposed privileged access to the single market and observer status in EU institutions. In June, France and Germany went further. Their joint non-paper, A New Momentum for Enlargement, envisages “full participation in the Single Market under an ‘EEA+ model’” as the final stage of gradual integration before full membership, for candidates that have adopted the acquis in the relevant clusters. The EEA, once the model for a permanent outer circle, has become for Paris and Berlin the anteroom to membership. Around the same time, Italy, Austria, Czechia, Slovakia and Slovenia circulated a non-paper on gradual sectoral access to the single market for candidates that meet high standards of alignment and rule of law.
Merz’s step-towards-membership logic is not new either. It is as old as the Treaty of Rome, which provided for association with third states (Article 238 EEC, today Article 217 TFEU). In 2024, Sébastien Maillard of the Jacques Delors Institute proposed reviving that logic as an “Associated State” status, a bold step towards enlargement, not an alternative to it. It was meant for candidates that had made substantial progress in negotiations, and would offer participation without voting rights in the Foreign Affairs Council, an observer delegation in the European Parliament, and access to EU funds well beyond pre-accession aid. The status would be reversible, and Maillard explicitly kept it apart from countries like Norway or Switzerland, which have no intention to join. Merz applied the same logic to the most politically sensitive candidate.
Kyiv’s answer was clear: Ukraine wants full membership, not a symbolic substitute. Candidates saw a waiting room with better furniture.
Von der Leyen’s version for Canada follows the other logic, that of the outer circle. She offered associate membership to Canada, a country that has not applied for membership and, under Article 49 TEU, cannot. Candidates will notice how quickly a new category appeared for a non-applicant while the debate on phased integration for their own countries was pushed down the road. Still, the move can also bring clarity, if the EU draws the right line. Associate membership becomes a final destination for close partners who do not want to join, not a holding pen for countries that do. Norway, Iceland or a future British government could one day fit that description. Ukraine, Moldova and the Western Balkans do not.
That line has to be stated explicitly, and the roadmaps are where to state it. Each roadmap should point clearly to full membership, and the EU must keep Merz’s interim logic separate from Canada’s permanent one.
The UK: a laboratory for sectoral integration
The most instructive case, though, may be the United Kingdom. After 2016, the EU’s position was that the single market is indivisible: no cherry-picking and no sector-by-sector deals. That stance hardened into a general reluctance to offer partial integration to anyone, candidates included.
That position is now changing. The UK–EU reset is being built sector by sector: an agri-food agreement, electricity market cooperation, linking carbon markets, youth mobility, and the 2025 Security and Defence Partnership. It has also shown where such arrangements hit their limits. Talks on UK participation in the SAFE defence instrument collapsed in November 2025 over the size of Britain’s financial contribution, while Canada negotiated its way in and the Council concluded that agreement in June. With Andy Burnham now in Downing Street and both sides aiming for another summit before the end of the year, the question is how far this sectoral model can go.
In a paper for Carnegie Europe last year, Stefan Lehne, Richard Grieveson and I argued that this shift matters directly for enlargement. As the trauma of Brexit recedes, the EU has become more flexible. The 2024 package with Switzerland, which covers much but not all of the single market and combines arbitration with the Court of Justice’s exclusive right to interpret EU law, shows that sectoral participation can be legally designed and governed. We argued that the same logic should apply to candidates. Once a country reaches a sufficient level of compliance and the Commission confirms it, it should be allowed to participate in that policy.
The UK and Canada talks are building the toolbox the candidates need: sectoral single market participation, dynamic alignment, governance and dispute settlement, financial contributions, and termination clauses. The EU is designing these instruments for a country that left and a country that will never join. It has no good reason to withhold them from countries that are negotiating to join and have reached the required level of compliance.
Still, the record so far should temper expectations. Piecemeal access, a programme here and a sector there, has brought neither visible success nor real leverage, and it has carried little political weight in the candidate countries. The gradual integration approach should therefore be far more ambitious: to bring as many candidates as possible up to EU standards across all 33 chapters, as quickly as possible, even when the formal accession process is not moving.
Three lessons for candidates, and three differences
The UK and Canada experience gives three practical lessons for gradual integration.
First, sectoral access is possible without breaking the single market, provided governance is robust. The candidates’ legal frameworks should be brought into line: the Stabilisation and Association Agreements with the Western Balkans are weaker than the Association Agreements with Ukraine and Moldova, and should be upgraded.
Second, security is now the fastest route to integration. Canada joined SAFE before being offered associate membership, and the UK’s relationship is being rebuilt largely through defence. Candidates should be brought into the European Defence Fund, the defence industrial programme and Readiness 2030, in return for full alignment with the Common Foreign and Security Policy.
Third, reversibility is what makes integration sellable. Merz’s snap-back clause and the termination clauses in agreements with third countries point the same way. The Union can reduce or suspend participation if the rule of law erodes. That is the argument most likely to reassure a French or Dutch voter.
There are also three differences, and the roadmaps must reflect them.
The direction. For the UK (at least for now, until a new referendum is called) and for Canada, sectoral integration is the end point. For candidates, each step must lead towards full membership.
The money. The SAFE episode showed that participation for third countries is pay-to-play. Candidates cannot pay their way in, and they already receive only a fraction of what new members get. Narrowing that funding gap has to be part of the package.
The seat at the table. Candidates need structured participation in EU institutions that grows as they progress. Once the Commission confirms a sufficient level of compliance in a given area, they should take part as observers in the corresponding Council working groups, Commission expert groups and European Parliament committees, well before the accession treaty is signed rather than only afterwards. This is not to give them a say before membership, but to prepare their administrations for it.
Where enlargement will be decided
The State of the Union showed a Commission willing to rethink the EU’s partnerships. That openness is welcome, but it cuts both ways. If the EU designs its most creative arrangements for Canada and the UK, and offers candidates only the classic formula, it will confirm the suspicion that enlargement has become the least flexible part of European policy.
By speaking French, von der Leyen acknowledged that the next enlargement will be decided in national parliaments, such as the German Bundestag and the Dutch House of Representatives (Tweede Kamer), and possibly in a French referendum, rather than in Brussels, Podgorica or Tirana. The roadmaps are the chance to win that audience. They should draw on the toolbox now being built for London and Ottawa, give candidates tangible benefits before accession, keep the Union able to reverse them, and point clearly to full membership.
They should also settle the endgame. By the time the accession treaty is signed, a candidate should already be inside much of the single market, on the EEA+ logic that Paris and Berlin now accept, and sitting as an observer in the rooms where it has proved compliance. The treaty would be signed in the classic way, and from signature the country would become an active observer: the status acceding states have held in past enlargements, attending the Council at every level, and the Parliament, with a voice but no vote. Its own ratification would then bring full application of the single market and the budgetary arrangements, the last step rather than the first. Member states could ratify at their own pace, without the pressure of an in-or-out vote and, the political decision to enlarge having been taken, without national objections that are not properly justified. If the roadmaps are written that way, the French passage may be remembered as the moment the Commission began preparing ratification, not just negotiations.
Spoken in French, written for Paris: the State of the Union and Europe’s new circles
by Zoran Nechev
Ursula von der Leyen switched to French when her 2026 State of the Union address reached enlargement. The passage was short and easy to miss among the heatwave plan, the Kids Act and the standing ovation for Mark Carney. It was also her clearest signal on enlargement.
Language choices in a speech like this are rarely accidental. France is where the next enlargement is most likely to be won or lost. Under Article 88-5 of the French Constitution, a new accession treaty must go to a referendum unless both chambers of Parliament choose a vote in Congress, where a three-fifths majority is needed. In France, as in the Netherlands, voters have grown warier of enlargement. By speaking French, the Commission President was addressing more than the candidate countries and the MEPs in Strasbourg. She was speaking to the people who will decide in the end whether Montenegro or Albania joins.
Read as a whole, though, the speech did more than repeat the enlargement formula she has used since 2023: their future lies in our Union, the process is merit-based, and values are non-negotiable. It sketched a Europe of concentric circles. The idea has deep French roots, from Jacques Delors and François Mitterrand’s European Confederation to Emmanuel Macron’s European Political Community.
The picture comes closest to the four tiers proposed in 2023 by the Franco-German Group of Twelve in its report Sailing on High Seas: an inner circle, the EU, associate members and the European Political Community. There, the associate tier was meant for the EEA countries, Switzerland “or even the UK,” built around the single market and kept apart from accession, “as membership in them can be permanent.” Von der Leyen has now stretched that outer circle across the Atlantic.
The “associate” idea has, however, travelled in two directions. In the concentric-circles reading, it is an alternative to membership. In another reading, it is a step towards it: only four months ago, Chancellor Friedrich Merz proposed associate membership for Ukraine as an interim stage on the road to full membership. The offer to Canada, a country that will never join, clearly belongs to the first reading. How these circles relate to one another, and whether a country can move from an outer circle towards the core, will shape enlargement over the coming years.
The enlargement passage: merit, values and roadmaps
The passage itself was brief. The EU’s neighbours, from Ukraine and Moldova to the Western Balkans, show that the European idea is alive. “Because their future lies within our Union, we will soon be proposing roadmaps for the candidates that have made most progress. This process will continue to be based on merit. And […] respect for our values is non-negotiable.”
Each element answers a worry that is strongest in Paris and The Hague. “Merit” answers the fear of geopolitical shortcuts. “Values” answers the fear of importing rule of law problems the Union still struggles with at home. She then gave the point an example: the images from Ratko Mladić’s funeral in Belgrade, which Serbian ministers attended, and her statement that glorifying war criminals “simply has no place in the European Union.” But “non-negotiable” has to mean consistent: rewarding each candidate’s real progress, sanctioning regression when it happens, and holding member states that flout the same values to account. Otherwise the Union loses the authority to demand them of others.
The only new commitment was the roadmaps, and they matter. Montenegro has opened all 33 negotiating chapters and provisionally closed 18. Albania has opened all 33 and closed three. Both are leaving the phase of opening negotiations and entering the much harder one of finishing them. After that, an accession treaty still has to pass 27 national ratifications. A roadmap that is only a checklist of closing benchmarks will add little. A useful one would set out the endgame: what the candidate still has to deliver, which safeguards apply after accession, what the EU reforms on its own side, and how ratification will be prepared. As Christophe Hillion argues, candidates should also have a seat in the new Congress of Europe that von der Leyen announced, as the candidate countries of the early 2000s had in the Convention on the Future of Europe. Those who will live under the reformed Union should help shape it.
From Kyiv to Ottawa: the life of “associate membership” so far
The Canada announcement cannot be read apart from what came before it. In May, Merz wrote to the EU leadership proposing “associate membership” for Ukraine. It would give Ukraine seats without votes in the European Council and the Council, a Commissioner without a portfolio, non-voting MEPs, an associate judge, gradual access to the EU budget and the protection of Article 42(7). A snap-back mechanism would freeze the status if Kyiv backslid on the rule of law. Merz described it as an interim step alongside accession negotiations, “not membership-lite.” For Moldova and the Western Balkans, he proposed privileged access to the single market and observer status in EU institutions. In June, France and Germany went further. Their joint non-paper, A New Momentum for Enlargement, envisages “full participation in the Single Market under an ‘EEA+ model’” as the final stage of gradual integration before full membership, for candidates that have adopted the acquis in the relevant clusters. The EEA, once the model for a permanent outer circle, has become for Paris and Berlin the anteroom to membership. Around the same time, Italy, Austria, Czechia, Slovakia and Slovenia circulated a non-paper on gradual sectoral access to the single market for candidates that meet high standards of alignment and rule of law.
Merz’s step-towards-membership logic is not new either. It is as old as the Treaty of Rome, which provided for association with third states (Article 238 EEC, today Article 217 TFEU). In 2024, Sébastien Maillard of the Jacques Delors Institute proposed reviving that logic as an “Associated State” status, a bold step towards enlargement, not an alternative to it. It was meant for candidates that had made substantial progress in negotiations, and would offer participation without voting rights in the Foreign Affairs Council, an observer delegation in the European Parliament, and access to EU funds well beyond pre-accession aid. The status would be reversible, and Maillard explicitly kept it apart from countries like Norway or Switzerland, which have no intention to join. Merz applied the same logic to the most politically sensitive candidate.
Kyiv’s answer was clear: Ukraine wants full membership, not a symbolic substitute. Candidates saw a waiting room with better furniture.
Von der Leyen’s version for Canada follows the other logic, that of the outer circle. She offered associate membership to Canada, a country that has not applied for membership and, under Article 49 TEU, cannot. Candidates will notice how quickly a new category appeared for a non-applicant while the debate on phased integration for their own countries was pushed down the road. Still, the move can also bring clarity, if the EU draws the right line. Associate membership becomes a final destination for close partners who do not want to join, not a holding pen for countries that do. Norway, Iceland or a future British government could one day fit that description. Ukraine, Moldova and the Western Balkans do not.
That line has to be stated explicitly, and the roadmaps are where to state it. Each roadmap should point clearly to full membership, and the EU must keep Merz’s interim logic separate from Canada’s permanent one.
The UK: a laboratory for sectoral integration
The most instructive case, though, may be the United Kingdom. After 2016, the EU’s position was that the single market is indivisible: no cherry-picking and no sector-by-sector deals. That stance hardened into a general reluctance to offer partial integration to anyone, candidates included.
That position is now changing. The UK–EU reset is being built sector by sector: an agri-food agreement, electricity market cooperation, linking carbon markets, youth mobility, and the 2025 Security and Defence Partnership. It has also shown where such arrangements hit their limits. Talks on UK participation in the SAFE defence instrument collapsed in November 2025 over the size of Britain’s financial contribution, while Canada negotiated its way in and the Council concluded that agreement in June. With Andy Burnham now in Downing Street and both sides aiming for another summit before the end of the year, the question is how far this sectoral model can go.
In a paper for Carnegie Europe last year, Stefan Lehne, Richard Grieveson and I argued that this shift matters directly for enlargement. As the trauma of Brexit recedes, the EU has become more flexible. The 2024 package with Switzerland, which covers much but not all of the single market and combines arbitration with the Court of Justice’s exclusive right to interpret EU law, shows that sectoral participation can be legally designed and governed. We argued that the same logic should apply to candidates. Once a country reaches a sufficient level of compliance and the Commission confirms it, it should be allowed to participate in that policy.
The UK and Canada talks are building the toolbox the candidates need: sectoral single market participation, dynamic alignment, governance and dispute settlement, financial contributions, and termination clauses. The EU is designing these instruments for a country that left and a country that will never join. It has no good reason to withhold them from countries that are negotiating to join and have reached the required level of compliance.
Still, the record so far should temper expectations. Piecemeal access, a programme here and a sector there, has brought neither visible success nor real leverage, and it has carried little political weight in the candidate countries. The gradual integration approach should therefore be far more ambitious: to bring as many candidates as possible up to EU standards across all 33 chapters, as quickly as possible, even when the formal accession process is not moving.
Three lessons for candidates, and three differences
The UK and Canada experience gives three practical lessons for gradual integration.
First, sectoral access is possible without breaking the single market, provided governance is robust. The candidates’ legal frameworks should be brought into line: the Stabilisation and Association Agreements with the Western Balkans are weaker than the Association Agreements with Ukraine and Moldova, and should be upgraded.
Second, security is now the fastest route to integration. Canada joined SAFE before being offered associate membership, and the UK’s relationship is being rebuilt largely through defence. Candidates should be brought into the European Defence Fund, the defence industrial programme and Readiness 2030, in return for full alignment with the Common Foreign and Security Policy.
Third, reversibility is what makes integration sellable. Merz’s snap-back clause and the termination clauses in agreements with third countries point the same way. The Union can reduce or suspend participation if the rule of law erodes. That is the argument most likely to reassure a French or Dutch voter.
There are also three differences, and the roadmaps must reflect them.
The direction. For the UK (at least for now, until a new referendum is called) and for Canada, sectoral integration is the end point. For candidates, each step must lead towards full membership.
The money. The SAFE episode showed that participation for third countries is pay-to-play. Candidates cannot pay their way in, and they already receive only a fraction of what new members get. Narrowing that funding gap has to be part of the package.
The seat at the table. Candidates need structured participation in EU institutions that grows as they progress. Once the Commission confirms a sufficient level of compliance in a given area, they should take part as observers in the corresponding Council working groups, Commission expert groups and European Parliament committees, well before the accession treaty is signed rather than only afterwards. This is not to give them a say before membership, but to prepare their administrations for it.
Where enlargement will be decided
The State of the Union showed a Commission willing to rethink the EU’s partnerships. That openness is welcome, but it cuts both ways. If the EU designs its most creative arrangements for Canada and the UK, and offers candidates only the classic formula, it will confirm the suspicion that enlargement has become the least flexible part of European policy.
By speaking French, von der Leyen acknowledged that the next enlargement will be decided in national parliaments, such as the German Bundestag and the Dutch House of Representatives (Tweede Kamer), and possibly in a French referendum, rather than in Brussels, Podgorica or Tirana. The roadmaps are the chance to win that audience. They should draw on the toolbox now being built for London and Ottawa, give candidates tangible benefits before accession, keep the Union able to reverse them, and point clearly to full membership.
They should also settle the endgame. By the time the accession treaty is signed, a candidate should already be inside much of the single market, on the EEA+ logic that Paris and Berlin now accept, and sitting as an observer in the rooms where it has proved compliance. The treaty would be signed in the classic way, and from signature the country would become an active observer: the status acceding states have held in past enlargements, attending the Council at every level, and the Parliament, with a voice but no vote. Its own ratification would then bring full application of the single market and the budgetary arrangements, the last step rather than the first. Member states could ratify at their own pace, without the pressure of an in-or-out vote and, the political decision to enlarge having been taken, without national objections that are not properly justified. If the roadmaps are written that way, the French passage may be remembered as the moment the Commission began preparing ratification, not just negotiations.