Middle Powers Rising: Carney, the Highest IQ (Western) Politician?

There are moments in geopolitics when an apparently eccentric proposal reveals more about the direction of the international system than a treaty or a war, and Ursula von der Leyen’s suggestion that Canada might become the European Union’s first “associate member” may prove to be one of them. It may, of course, prove to be nothing at all. There is currently no such category in the EU treaties, its rights and obligations remain undefined, and several of the governments whose agreement would ultimately be required appear to have been surprised by the announcement, which is hardly unusual in a Union that has a distinguished history of allowing grand strategic language to disappear into committees, legal objections and institutional process. Yet the significance of the proposal lies partly in the fact that it has been made before its architecture exists. Von der Leyen did not simply propose closer relations with Canada, for which there are already ample mechanisms, but described Canada as potentially the “first” associate member of the Union. What initially sounds like an unusually ambitious Canadian partnership may therefore contain the beginnings of a new category in the international system. For the moment, however, that is a hypothesis rather than an institution: the October Canada-EU summit is expected to begin giving substance to a concept whose content remains conspicuously uncertain.

Such a category would fill a curious gap. NATO provides collective defence, the G7 coordinates advanced industrial democracies, BRICS offers an increasingly heterogeneous collection of states a forum that is not centred upon Western institutions, and the Shanghai Cooperation Organisation connects much of Eurasia through a looser political and security structure. The European Union is something deeper than all of them in institutional terms, a supranational political, economic and legal system whose extraordinary degree of integration has historically made membership both valuable and difficult. What does not really exist is an institutional layer capable of combining substantial economic integration, technological cooperation, defence-industrial coordination, access to strategic supply chains and a degree of political alignment without requiring a country either to become an EU member or to rely upon a conventional military alliance for everything that now follows from the increasingly elastic word “security”. Canada may be the experiment through which Europe discovers whether such a layer can be constructed.

The immediate genealogy of the idea is slightly misleading because Friedrich Merz had already proposed “associate membership” earlier in 2026 primarily as a solution to the problem of Ukraine, envisaging an intermediate status through which Ukraine could participate in parts of European institutional life while its eventual accession proceeded. There was some ingenuity in this, but it remained essentially an antechamber to membership, which helps explain Ukrainian anxiety that the waiting room might quietly become the destination. Canada transforms the concept because Canada cannot plausibly be understood as a country waiting patiently for geography to permit it to become Belgium. If Canada becomes an associate member, association ceases to mean incomplete accession and becomes something conceptually different: participation in a European-centred strategic system by a sovereign state that has neither the possibility nor the intention of becoming an ordinary EU member.

This is where Mark Carney becomes much more interesting, although it would be premature to identify him as the sole intellectual architect of what is emerging. Ideas of middle-power cooperation are hardly new, and the European Union has been developing strategic partnerships, defence arrangements and economic-security instruments independently of Canada, but Carney’s argument provides a coherent intellectual framework through which his subsequent diplomacy can be understood. He has repeatedly argued that the international order has moved into a world in which integration itself can be weaponised and sovereignty consequently requires not withdrawal from international systems but greater control over the terms on which a country participates in them. His language at Canada’s Investment Summit this week was unusually explicit: Canada’s objective in constructing new partnerships is to increase “strategic autonomy”, precisely because integration has become capable of being used as leverage, and because sovereignty now reaches into areas such as artificial intelligence, payments, space, critical minerals and energy that would once have been regarded as primarily commercial. Canada, he noted, has concluded more than twenty trade and security agreements across five continents in the past year and more than fifty critical-minerals agreements with over fifteen countries, while seeking to deepen relationships with particularly trusted partners.

It is useful to think of this as Carney’s Middle Powers Doctrine, not because foreign policy needs another doctrine bearing a politician’s surname, but because the phrase identifies both the group of states to which the argument is addressed and the structural problem it is trying to solve. A middle power cannot individually defeat a hegemon at the hegemon’s preferred game: Canada cannot militarily balance the United States, Australia cannot economically balance China, Norway cannot construct an independent European security order, Britain cannot recreate the strategic scale of either America or the EU, and Japan and South Korea, despite their enormous industrial sophistication, inhabit a security environment shaped overwhelmingly by larger states. Their potential strength lies not in somehow becoming individually larger but in becoming collectively more difficult to coerce, which requires something more complicated than a traditional alliance. Rather than merely aggregating military capabilities against an identified adversary, middle powers can construct alternative markets, technologies, supply chains, defence relationships and sources of capital so that no larger power possesses an indispensable choke point over them.

Canada’s deeper engagement with the EU matters because it begins to turn this proposition from rhetoric into infrastructure, although even here it is important to distinguish what already exists from what is merely being imagined. Canada signed its agreement to participate in the EU’s SAFE defence instrument in February, it has been provisionally applied since April and formally concluded by the Council in June, making Canada the first non-European country admitted to the programme. SAFE itself is not a military alliance but an EU financial instrument intended to expand defence-industrial production and common procurement, which makes Canadian participation conceptually interesting: Canadian companies and Canadian-origin products can participate in a European mechanism designed to increase Europe’s capacity to manufacture the material of defence.

That distinction between military protection and the industrial foundations of military power is central to the wider argument. NATO remains overwhelmingly the institution that answers the question of who assists if a member is attacked, whereas an emerging EU-centred economic-security architecture could increasingly answer the less dramatic but no less consequential questions of who manufactures the radios and drones, who refines the minerals, who builds the batteries, who finances the factories and whose industrial capacity can be mobilised when peacetime supply chains cease to behave as expected. One small example already exists. Montréal-based Marconi Technologies has been selected with a Polish defence partner to provide tactical communications equipment to the Polish military in a programme linked directly to SAFE, with deliveries beginning this year. It is not strategically transformative in itself, but it demonstrates physically what the abstraction means: Canadian industrial capability entering a European procurement system in order to strengthen a military on NATO’s eastern flank.

Von der Leyen’s proposed expansion of the relationship suggests a much broader ambition, encompassing advanced manufacturing, technology, integrated defence-industrial bases, energy, critical minerals, batteries, artificial intelligence, quantum technology, cybersecurity, economic security and the Arctic. This is not really the vocabulary of a conventional free-trade agreement. It describes strategic capacity, although whether that capacity eventually becomes institutionalised under the title “associate membership” remains precisely what has yet to be decided.

The relationship is consequently too easily described as Canada escaping excessive dependence upon the United States by moving closer to Europe, because that captures only one half of the bargain. Europe has spent the past decade discovering its own concentrations of dependence, first through Russian hydrocarbons, then through the extraordinary Chinese position in manufacturing and critical-mineral processing, and increasingly through questions surrounding American military capability, technology platforms and capital. In each case the underlying lesson has been similar: an economic relationship optimised for efficiency during benign periods can become a strategic vulnerability when geopolitical conditions change. Canada therefore offers Europe something Europe cannot manufacture through regulation: physical strategic depth. Canada possesses enormous energy resources, major uranium production, critical-mineral deposits, agricultural capacity, advanced aerospace and nuclear industries, vast territory, three-ocean access and an Arctic geography of increasing military and economic importance, while Europe brings almost the inverse portfolio of roughly 450 million consumers, sophisticated manufacturing, regulatory reach, deep capital markets, advanced research institutions and increasingly large defence demand. The complementarity is unusually strong.

Critical minerals illustrate the point particularly well. Canada possesses substantial rare-earth resources but has historically not been a major commercial producer, while China remains overwhelmingly dominant, particularly in processing. The strategic importance of Canadian development therefore lies less in fantasies of replacing Chinese volume than in constructing alternative chains at precisely the stages where concentration creates vulnerability. Emerging Canadian capacity in magnet rare earths such as neodymium-praseodymium and the heavier dysprosium and terbium matters because these materials feed permanent magnets used across electric motors, wind generation, robotics, aerospace and defence. Europe does not need Canada to replace China; it needs Canada, alongside other suppliers and processors, to help ensure that China is not the only realistic answer when particular strategic materials are required.

Australia strengthens the logic considerably because Canberra has already been deepening economic and security cooperation with Europe and has publicly indicated that it is watching the Canadian initiative closely. Australia contributes enormous mineral resources, established mining capacity and an Indo-Pacific geography that Canada cannot provide. If it eventually followed Canada into some form of association, the implications would be difficult to dismiss as an extension of European neighbourhood policy. Canada can, with sufficiently imaginative cartography, be described as belonging to an extended North Atlantic community; Australia cannot. Australian participation would establish that association was no longer fundamentally geographical but had become a mechanism through which selected non-European states could embed themselves deeply in a European-centred economic-security system without pretending to become European. Britain and Norway then become obvious members of one possible circle, albeit from very different existing relationships with the EU, while Japan, South Korea and New Zealand become imaginable in another. Ukraine and Moldova remain fundamentally different because their stated destination is full EU membership, while Georgia occupies the more awkward position of being formally a candidate and already deeply economically associated with the Union but with its accession trajectory politically stalled.

At this point, however, a sceptic should reasonably interrupt the argument, because there is a danger of mistaking a new label for a new geopolitical system. Canada already has CETA. It already has an EU Security and Defence Partnership. SAFE already provides a route into European defence procurement. Horizon, Erasmus, regulatory dialogues, critical-minerals agreements and bilateral military cooperation can all be expanded without inventing “associate membership”. European officials themselves have acknowledged that no associate category currently exists, while no comprehensive institutional blueprint has yet been agreed.

The strongest sceptical interpretation is therefore that nothing structurally new is happening at all. Canada and Europe share interests, American policy has accelerated their cooperation, Brussels likes institutional language and von der Leyen has attached an arresting title to relationships that could perfectly well continue without it. There is already criticism that the proposal may be largely symbolic and that it does little by itself to solve Europe’s more fundamental security problems.

That objection is important because it gives us a test. The significance of associate membership cannot be judged by the phrase itself but by whether it eventually produces things that the existing Canada-EU relationship could not easily produce. If the October summit merely generates another declaration of strategic partnership, additional working groups and an enlarged list of cooperative sectors, then “associate member” will have been branding. If, however, Canada acquires standing mechanisms for regulatory consultation before European rules are made, privileged and durable access to EU procurement and industrial programmes, meaningful labour or professional mobility, participation in selected European institutions, integrated strategic investment mechanisms or other rights and obligations sufficiently deep to require a genuinely new legal category, then something different will have occurred. The distinction between branding and architecture is observable rather than semantic.

This matters because a genuinely new category would create consequences far beyond Canada. Brussels has spent years insisting that the benefits of deep European economic integration cannot simply be cherry-picked by outsiders. Brexit made the principle almost existential: Britain could leave the political institutions but could not expect identical Single Market privileges without accepting corresponding obligations, while Norway receives extraordinary access in exchange for accepting substantial European regulation without equivalent voting power and Switzerland has spent decades negotiating its own complicated network of bilateral arrangements. If Canada is now offered privileged access to European programmes, procurement, research, defence markets, professional mobility or elements of the Single Market without obligations comparable to those expected of European outsiders, London will immediately ask why, Oslo will ask what additional benefit it receives for its much deeper regulatory alignment, Switzerland will examine the precedent, and Ukraine may reasonably wonder why a wealthy transatlantic democracy is being offered an attractive new status while Ukrainians fight for eventual membership.

This could become the greatest institutional danger in the project, but it may simultaneously constitute its greatest opportunity because the European Union has spent decades trapped by a surprisingly binary constitutional imagination in which a country is either inside an extraordinarily deep system of integration or remains a third country negotiating bespoke arrangements from outside it. The EEA, Switzerland, candidate status, neighbourhood policy and post-Brexit Britain have produced numerous exceptions without producing a coherent outer architecture. Canada may force Europe finally to invent one, creating concentric layers in which full members retain voting rights, budgetary obligations, free movement and supranational law while associates participate selectively but deeply in defence industry, research, energy, critical minerals, technology, procurement and perhaps particular economic sectors without pretending to possess identical political rights. If that happens, associate membership will represent constitutional evolution disguised as Canadian foreign policy and potentially provide Europe with an answer to one of its central geopolitical problems: how to increase European strategic scale without endlessly enlarging the European Union itself.

It is only once that institutional possibility is established that the idea of network power becomes more than a metaphor. Such a structure would not resemble BRICS, whose extraordinary breadth is purchased partly through institutional shallowness, nor NATO, whose essential proposition remains collective military defence, nor a federalising European Union, because its participants would retain separate sovereignties, currencies, foreign policies and military commitments. Instead, strategic capabilities could remain distributed among sovereign states while being connected in ways that make them progressively more valuable to one another: a Canadian mine feeding a European processor, an Australian lithium project supplying a European battery manufacturer, Canadian communications equipment entering Polish military procurement, European defence financing expanding Canadian industrial capacity, Norwegian energy supporting European manufacturing and British aerospace technology entering multinational weapons programmes. No participant would control the whole system, but each would acquire a growing interest in preserving it, with the political consequence that coercing one node would increasingly impose costs upon others.

This is where some rather dry academic work on interdependence becomes unexpectedly useful. Albert Hirschman understood that two countries can benefit enormously from trade while remaining unequally dependent upon it because political power resides partly in the relative cost of exit, while Keohane and Nye subsequently distinguished between sensitivity to disruption and vulnerability to it. The distinction provides a useful theoretical foundation for what Carney appears to be attempting because dependence is not simply the existence of a relationship; it becomes strategically dangerous when there is no adequate alternative to that relationship. For middle powers, strategic autonomy therefore cannot sensibly mean autarky, which would be economically ruinous and in most cases physically impossible. It can instead mean constructing enough alternative relationships that the loss of any single one, although painful, does not become existential.

Canada is an almost perfect laboratory for the proposition because few advanced Western countries are more dramatically constrained by geography. Its border with the United States stretches for almost 9,000 kilometres, its population is a fraction of America’s, its industrial systems are deeply intertwined and its defence relationship is not merely close but structurally integrated through NORAD and a defence-industrial relationship built over generations. Energy provides the clearest numerical measure of the resulting interdependence. In 2025 Canada supplied 63.4% of all crude oil imported by the United States, close to 100% of imported natural gas, 97.9% of imported natural-gas liquids and 81.3% of imported electricity. Yet the same flows of crude, refined products, natural gas and NGLs sold south were worth C$157.5 billion to Canada, equivalent to 20.2% of all Canadian goods exports worldwide. Canada therefore possesses something more complicated than an energy weapon: an extraordinary relationship of asymmetric mutual dependence in which American vulnerability to disruption is substantial but Canadian exposure to the relationship is itself enormous.

At first sight Ottawa could therefore threaten to turn off the taps and impose significant costs on American refineries, particularly those configured around Canadian heavy crude, but historically the threat injured its owner almost as seriously as its target because Canada had nowhere comparably efficient to send the displaced volume. The resource existed without producing equivalent leverage because the exit route did not. This is why pipelines are not simply infrastructure but foreign policy constructed from steel. The Trans Mountain expansion altered Canadian geopolitics because Alberta oil acquired substantially greater access to the Pacific, and the first full year after expansion already shows the scale of the change: Canadian crude exports to destinations other than the United States rose 132.6% in 2025, reaching 10.9% of total crude exports, more than three times the average share recorded over 2016–24. The United States nevertheless still received 222.7 million cubic metres of Canadian crude, which illustrates both the speed of diversification and how far Canada remains from anything resembling energy independence from its neighbour.

Nor should complete energy independence be the objective. Selling Alberta crude into the enormous adjacent American market will often make more economic sense than shipping the same barrel across the Pacific, and a Canada capable of redirecting every barrel to China would merely have exchanged one concentration for another. What matters is the ability to redirect a sufficiently large marginal volume that Washington can no longer assume Canadian energy is physically captive to the North American system. The political value of an exit route begins before anybody actually uses it.

That same logic makes Canada’s relationship with China considerably more complicated than a simple democratic-versus-authoritarian division of the world would suggest. China is simultaneously one of the principal reasons Canada and Europe want more resilient critical-mineral and manufacturing supply chains and one of the markets capable of helping Canada reduce its excessive dependence upon the United States. Canada may therefore need deeper economic engagement with China in some sectors in order to acquire greater independence from America while cooperating with Europe to acquire greater independence from China in others. This is not necessarily incoherent; it is what diversification looks like when a state is attempting to preserve room for manoeuvre rather than merely exchange one dominant relationship for another.

That also exposes the weakness in imagining Carney’s Middle Powers Doctrine as simply the creation of a new Western bloc. If it is coherent, the doctrine should not require an anti-Chinese Canada any more than European association should require an anti-American one. Canadian oil sold into Asian markets can increase Canada’s bargaining power within North America, European processing of Canadian minerals can reduce Chinese leverage, and American defence integration can continue protecting Canada while European defence-industrial integration reduces the exclusivity of that American relationship. The resulting arrangement will often look inconsistent if judged by the standards of a rigid alliance, but the inconsistency is partly the point.

Nowhere is that overlapping geography more significant than in the Arctic, which for decades was too easily treated as an environmental frontier sitting above conventional geopolitics but is becoming one of the places where resources, climate, technology and military competition intersect most sharply. Russia possesses enormous Arctic territory and extensive military infrastructure, China has sought a larger economic and strategic role despite not being an Arctic state, melting sea ice alters potential shipping routes and access to resources, and satellites, undersea cables, early-warning systems, submarine operations, critical minerals, icebreakers and northern ports increasingly belong to the same strategic conversation. Put Canada together with Denmark through Greenland, Finland, Sweden, Norway and the northern European states and a geographical structure appears that is easily missed on conventional maps: a transatlantic northern arc in which North American and European security are physically adjacent rather than separated by an ocean.

The significance of that arc lies precisely in the fact that NATO and the EU contribute different things to it. NATO can organise deterrence, surveillance, military exercises and collective defence, but it is not principally an institution for financing mineral processing, ports, telecommunications, energy networks, civilian satellite capacity or industrial investment. An EU-Canada economic-security relationship potentially fills some of that space, which explains why the Arctic has appeared explicitly in the proposed partnership rather than as an ornamental reference to shared geography. In the north, the distinction between economic infrastructure and security infrastructure becomes particularly difficult to sustain: a port can service commercial and naval vessels, a satellite network can monitor environmental conditions and military activity, and a mineral project can be simultaneously an economic investment and an attempt to remove a strategic supply-chain vulnerability.

The logic becomes much less tidy when it reaches Canada’s own federation, because most Canadian oil is not produced in Ottawa and the province containing much of it has interests that do not always coincide with those of the federal government. Alberta possesses enormous hydrocarbon resources, substantial constitutional authority over natural resources and a political culture in which federal energy policy has repeatedly become entangled with questions of provincial autonomy. Alberta separatism remains a minority position and should not be inflated into an imminent break-up of Canada, nor is there evidence that Washington is orchestrating it, but geopolitically Alberta exposes the limit of Ottawa’s strategy because Canada’s strongest potential instrument of leverage over the United States is energy while Alberta has powerful economic reasons to preserve access to the very American market Ottawa wishes to make less indispensable.

Ottawa’s sovereignty project and a hypothetical Alberta sovereignty project could consequently point in opposite directions. Ottawa can seek to use Alberta’s resources to make Canada less dependent upon America, while an Alberta separatist movement could seek to use American demand to make Alberta less dependent upon Ottawa. An independent Alberta would be landlocked, its mature pipeline network would point predominantly toward the United States and American capital, markets, security relationships and monetary gravity would be immensely powerful. None of this means that an independent Alberta would automatically seek American statehood, still less that Albertans presently favour such an outcome; it means that the economic geography of independence would create powerful incentives for close alignment with the United States.

Seen in that context, westward pipeline capacity becomes simultaneously international strategy and Canadian nation-building. The implicit federal bargain is that Alberta does not need independence to obtain world-market access because Canada can provide a Pacific outlet, while Ottawa acquires greater national strategic autonomy by giving Alberta greater provincial economic autonomy. Alberta can continue selling to America when that is most profitable while possessing Asian alternatives when circumstances make them useful, so the same infrastructure that reduces American leverage over Canada potentially weakens one of the economic arguments available to Alberta separatists. The pipeline alters Canada’s bargaining relationship with Washington, Alberta’s bargaining relationship with Ottawa and the internal political economy of the Canadian federation at the same time.

This is also why a hypothetical American attempt to annex Canada by force, although an interesting stress test against recent rhetoric concerning Canadian sovereignty, is ultimately less analytically useful than the possibility of strategic absorption through existing dependencies. There is no public evidence of American preparations to conquer Canada, and military speculation quickly becomes detached from anything observable. The United States does not require tanks in Toronto to possess enormous leverage over Canada because trade, pipelines, defence, capital, technology, aerospace, integrated supply chains and sheer proximity already provide it. The deeper sovereignty question is whether Canada can make coercion sufficiently costly without attempting the impossible task of making itself militarily equal to the United States.

European integration potentially contributes to that objective in an unusual way because Canada does not require an EU equivalent of NATO Article 5 while it remains a NATO member anyway. Instead it can make Canadian industrial capacity, resources, technologies and infrastructure progressively more important to European states, so that severe coercion against Canada would cease to be a purely bilateral Canadian-American event and begin disrupting European defence programmes, mineral supplies, investments, Arctic security and technology projects. Canada becomes more resilient not because Europe promises to reconquer Ontario but because the consequences of damaging Canada spread more widely through the systems in which it participates. What initially looks like sovereignty surrendered through deeper entanglement can therefore become sovereignty strengthened by reducing the dominance of any one entanglement.

The political sensitivity of this strategy is already visible in Washington’s hostile reaction to the proposed EU association, even though its actual institutional content remains undefined. That reaction should not be exaggerated into proof that a coherent new bloc already exists, but it demonstrates that diversification ceases to look like an abstract exercise when procurement, technology, trade and strategic relationships begin moving away from established patterns.

For middle powers the apparent paradox disappears once sovereignty is separated from autarky. Canada will always depend upon others, as will Australia, Britain, Norway, Japan, South Korea and individual European states; the strategically meaningful questions concern whether those dependencies are concentrated or distributed, reciprocal or asymmetric, and replaceable or indispensable. Carney’s Middle Powers Doctrine can therefore be understood most coherently as an attempt to convert asymmetric dependence into a broader pattern of interdependence, a problem shared by countries whose geographical and economic circumstances otherwise look remarkably different. Australia relies heavily upon Asian trade while depending substantially upon the United States for security; Japan and South Korea combine immense industrial sophistication with deep American security dependence and major Chinese economic relationships; European states discovered that cheap Russian energy could become political vulnerability; and Britain discovered after Brexit that legal sovereignty does not abolish economic geography.

The deeper problem lies in the assumptions of the post-Cold War order, which rewarded efficiency by concentrating production where it was cheapest, sourcing energy from convenient suppliers, outsourcing much European security to Washington, stretching manufacturing supply chains through China and integrating financial systems on the assumption that economic interdependence made serious conflict increasingly irrational. The flaw was not interdependence itself but asymmetric interdependence without adequate exit routes. When integration can be weaponised, efficiency and vulnerability become two descriptions of the same network, yet middle powers cannot solve that problem by recreating every supply chain domestically because Canada cannot manufacture everything Canada consumes, Europe cannot mine every mineral European industry requires and Australia cannot independently guarantee the sea lanes through which its trade travels. Autarky would make them poorer without necessarily making them safer.

Their alternative is redundancy, which is less glamorous than strategic autonomy but probably closer to what the policy requires. Several suppliers may cost more than one optimised supplier, redundant pipelines are expensive, domestic or allied processing may cost more than Chinese processing, overlapping defence relationships complicate procurement and simultaneously maintaining useful relationships with Washington, Brussels and Beijing creates diplomatic contradictions. Yet those apparent inefficiencies create choices when circumstances change. Strategic resilience is partly the deliberate purchase of capacity one hopes never to need.

This is also where Carney’s Middle Powers Doctrine must confront its own conceptual limitation. If “middle powers” simply means Canada, Europe, Australia, Japan and South Korea, the supposedly new architecture risks becoming little more than a reorganised collection of familiar advanced democracies. India is a middle power by almost any meaningful geopolitical definition but will not surrender its Russian relationship or tradition of strategic autonomy to a European project; Brazil possesses its own conception of multipolarity; Indonesia, Saudi Arabia, the UAE and South Africa navigate among larger powers rather than fitting neatly into a democratic strategic bloc; and several belong to BRICS precisely because they value institutional environments in which Western political alignment is not the price of participation.

The doctrine therefore faces a choice between becoming a relatively cohesive network of advanced democracies seeking resilience against pressures generated by America, China and Russia, which could be useful but less revolutionary than its rhetoric suggests, or developing into a broader theory of middle-power agency in which coalitions vary according to the issue. Europe and Canada might integrate deeply in defence while Canada simultaneously sells energy to China; Australia and Japan might cooperate over critical minerals; India and Europe over technology; Gulf states and Europe over capital and energy; ASEAN countries where particular interests coincide. Such an architecture would be untidy, but a network does not require every participant to maintain the same relationships, and the international order may be moving away from neat blocs at precisely the moment political rhetoric increasingly describes the world as divided into them.

If the EU associate experiment develops beyond branding, it is consequently possible to imagine by the mid-2030s a system in which Canada, Britain, Norway and perhaps Australia participate at different depths in European economic-security institutions; Ukraine and perhaps Moldova have moved further into the Union itself; Japan and South Korea are integrated into selected industrial and technological initiatives; European defence procurement draws upon factories across several continents; Canadian and Australian minerals enter increasingly diversified processing chains; Arctic infrastructure develops across a Canada-Greenland-Nordic continuum; and energy, artificial intelligence, quantum, space, payments and cyber systems are connected through agreements that stop well short of political federation. Such a system might possess no common flag, president or single founding treaty, yet still exercise considerable power because its strength would derive not from central command but from the density of its connections and the costs created by breaking them.

This is where comparison with BRICS and the SCO becomes useful without imagining that the emerging structures perform identical functions. Their expansion demonstrates demand for international arrangements that give states additional options in a world no longer comfortably organised around a single American-led system, and their appeal does not necessarily arise from ideological identification with Beijing or Moscow so much as from the desire not to possess only one institutional route through which international influence and economic cooperation can be pursued. Europe would misunderstand this if it responded to multipolarity by attempting to construct a smaller unipolar system around itself. A more consequential role would be to provide institutions through which states can acquire additional strategic capacity without having to surrender every alternative relationship.

The EU possesses something few international organisations can reproduce: institutions capable of converting political agreement into standards, procurement rules, research programmes, industrial policy, financing and enforceable economic relationships. What it increasingly lacks is sufficient strategic resources and confidence that its traditional external dependencies will remain benign. Canada possesses many of those resources, Australia others, Britain military, financial and intelligence capabilities, Norway energy and Arctic geography, and Japan and South Korea industrial and technological capacity. None of them needs to become European if deeper connection to Europe increases rather than diminishes their freedom of action.

That is ultimately the test of associate membership. If it becomes merely an arresting description for CETA, SAFE and an expanding collection of bilateral programmes, it will be politically interesting but institutionally modest. If it becomes a waiting room for countries unwilling or unable to join the EU, it will solve a different and narrower problem. If, however, it becomes a mechanism through which selected capabilities of sovereign states can be integrated into a common economic-security architecture while those states retain meaningful relationships outside it, something genuinely new may be emerging: not another superpower, empire or conventional military alliance, but a system intended to make superpowers less capable of dictating the choices available to everyone beneath them.

There is a final irony in all this because for most of the post-war period Canada benefited enormously from precisely the geography it is now attempting to make less determinative. It shared a continent with the world’s richest and most powerful country, enjoyed privileged access to its market and sheltered beneath a continental security system in which American power overwhelmingly worked in Canada’s interests. Pipelines ran south because south was profitable, factories integrated because integration was efficient, defence structures merged because the threat came from elsewhere and trade concentrated because the American market was enormous and immediately adjacent. Nothing irrational occurred. The vulnerability emerged from decades of individually rational decisions made under assumptions that subsequently became less certain, just as Europe’s dependence upon Russian gas and Chinese industrial supply chains did not begin as strategic stupidity but as economic optimisation under political conditions expected to persist.

Carney’s Middle Powers Doctrine is therefore best understood not as a doctrine of disengagement but as an attempt to insure against the possibility that today’s indispensable partner can become tomorrow’s coercive one. Insurance has a cost: redundant pipelines cost money, alternative suppliers may be more expensive, domestic or allied processing may be less efficient, multiple defence relationships complicate procurement and simultaneously maintaining useful relationships with Washington, Brussels and Beijing will produce contradictions that cannot always be reconciled elegantly. That inefficiency is not necessarily evidence of failure. It may simply be the premium paid for possessing the option.

The question confronting Canada, Europe and other middle powers is consequently no longer simply how to maximise prosperity within a stable international system, but how much efficiency they are prepared to sacrifice in order to retain meaningful choices when the system becomes unstable. Seen from that perspective, the apparently strange spectacle of a Canadian prime minister deepening his relationship with institutions created to integrate Europe begins to make considerably more sense. Canada cannot move away from the United States, Australia cannot move away from China, Europe cannot manufacture mineral deposits, Alberta cannot move its oilfields to Vancouver, Britain cannot undo the economic geography of the Channel and Japan cannot choose different neighbours. Geography remains stubbornly resistant to political aspiration.

What governments can alter is the infrastructure connecting geography to power. Pipelines can run west as well as south, minerals can be processed in Saskatchewan rather than exclusively in China, Canadian defence companies can become embedded in European procurement, Australian resources can enter European strategic supply chains, Britain can participate in European security without rejoining the Union, Canada can remain inside NORAD while becoming more deeply integrated into European defence, and countries can acquire additional strategic relationships without destroying their existing ones. The purpose is not to escape geography, which is impossible, but to prevent geography from dictating every available choice.

Geography still writes the first draft of foreign policy; Carney’s Middle Powers Doctrine is an attempt to give middle powers the right to edit it.