By Robbin Laird and Ken Maxwell

For the past several years, Western policymakers have comforted themselves with a story: that economic interdependence with China would eventually civilize the relationship, that trade would do the work diplomacy could not, and that rising Chinese power would express itself the way past rising powers did, through fleets, flags, and borders redrawn by force. We did not write our new book, The Australian, Brazilian, and Chinese Dynamic: An Inquiry into the Evolving Global Order, to confirm that story. We wrote it because that story is wrong, and because getting it wrong is expensive for every resource-exporting democracy, not just the two we studied most closely.

China is not building an empire the way empires have historically been built. What China has built instead is quieter and, in some ways, more durable: a structure of dependence assembled out of ore refineries, container terminals, shipping contracts, and processing monopolies, an architecture engineered so that the cost of defying Beijing rises steadily over time, while the cost of accommodation quietly disappears into the ordinary texture of doing business. We call this China’s informal empire. It does not need to conquer anyone. It only needs patience, and enough control over the chokepoints of trade that most nations find resistance an increasingly difficult case to make to their own citizens.

To understand how this actually works not as theory but as lived national experience, we built the book around a close comparison: Australia and Brazil. Two democracies. Two resource-rich, commodity-exporting economies in the Southern Hemisphere. Two nations that share a great deal structurally, and that nonetheless arrived at different points on the spectrum between resistance and accommodation. That is not a verdict on either country’s character or wisdom. It is a case study in how differently the same architecture of dependence can play out depending on the specific commodities involved, the alliances already in place, and decades of prior investment that neither country’s current leadership fully controls. Every middle power holding commodities China needs should read both halves of this story as a mirror of its own possible future.

Retiring a Useless Term

Before making that comparison honestly, we had to clear away a piece of vocabulary that has outlived its usefulness: the “Global South.” It is one of the most overworked and analytically misleading phrases in international commentary, a term that implies a coherent bloc of the historically marginalized standing in solidarity against Western dominance, when in practice it has become a rhetorical costume that great powers put on when it suits them.

Australia and Brazil are, in our framework, the genuine article: nations shaped by histories of colonial extraction, still structurally dependent on commodity exports, still exposed to the pricing power of whoever happens to be buying their ore or their soybeans this decade. China, Russia, and increasingly India are something else, established or rising great powers with imperial histories, permanent seats at the table of global institutions, and, in China’s and Russia’s cases, an active project of rewriting the borders and rules of the current order. Calling all of these actors members of the same “Global South” doesn’t clarify anything. It flattens the world into a false symmetry in which a soybean exporter and a nuclear-armed permanent member of the UN Security Council are treated as fellow travelers. They are not. And that confusion is not accidental. It is one of the softer instruments in the informal empire’s toolkit, useful precisely because it discourages countries like Australia and Brazil from recognizing how much they actually have in common as targets of the same strategy.

The Architecture of Dependence

Strip away the diplomatic language, and China’s strategy toward resource economies rests on a single insight: control the processing stage, not just the resource. Extraction is not where the leverage lives. Refining is.

Take rare earths. China accounts for roughly two-thirds of global extraction, but it controls more than four-fifths of global refining capacity and it has already shown a willingness to weaponize that gap, restricting exports of gallium and germanium, inputs critical to both advanced manufacturing and defense production. Or take lithium, where Australia produces a substantial share of the world’s raw supply and yet finds the economic value of that resource largely trapped until the ore is routed through Chinese refineries. Owning the mine turns out to matter far less than owning what the ore becomes. That is the entire logic of vertical integration, and it applies with equal force to any commodity exporter, regardless of which government happens to be in Brasília or Canberra in a given year.

Layer onto this a physical logistics grid, a Chinese-financed deep-water port in Peru designed to anchor a transcontinental corridor linking Pacific and Atlantic trade, clustered Chinese-linked operations at both entrances to the Panama Canal, and a pattern emerges: infrastructure quietly engineered so that certain trade flows become, over time, not just convenient but effectively mandatory.

There is also a darker layer beneath the formal one. In Brazil, low container-inspection rates and the opacity of vertically integrated port operations have created conditions exploited by transnational criminal organizations like the Primeiro Comando da Capital, which move illicit goods alongside legitimate agricultural exports. This is not a Brazilian failure of governance so much as a structural byproduct of a logistics system designed elsewhere, for someone else’s benefit. It echoes an eighteenth-century Brazilian colonial economy documented in the Arte de Furtar — “The Art of Stealing” — in which condoned illegal trade was not a lapse but an embedded feature of an externally imposed system, one that kept local elites dependent on, and loyal to, an outside power. The Methuen Treaty tells a version of the same story: privileged access secured through informal channels, dependency dressed up as commerce. China did not invent this pattern. It has rebuilt it at a scale no eighteenth-century empire could have imagined, and it is a pattern that has been imposed on resource economies from the outside for centuries, Brazil’s experience of it today has deep roots that predate any recent policy choice.

Brazil: A Deep Integration, Not a Simple Choice

The scale of Brazil’s integration with China is genuinely difficult to overstate: bilateral trade exceeded $181 billion in 2023. By 2024, China was absorbing 73 percent of Brazil’s soybean exports and more than 71 percent of its iron ore. That same year, oil became Brazil’s single largest export product, with 44 percent of shipments bound for China.

It is important to be honest about how this came to be. Brazil did not simply decide, one morning, to subordinate its economy to Chinese demand. The Cerrado, a vast interior savanna once considered too acidic for serious agriculture, was transformed over decades, through enormous technological and infrastructural investment, into one of the most productive farming regions on earth, at a moment when Chinese demand for soy and protein was surging and few other buyers could absorb volumes at that scale. Brazilian companies like Vale, JBS, and CBMM, the latter controlling roughly 97 percent of the world’s niobium supply, are genuinely formidable global players in their own right. Brazil’s integration with China reflects real comparative advantages in agriculture and mining, developed and pursued in good faith as a legitimate development strategy, not merely a surrender of sovereignty.

The cost of that trajectory, though, deserves equally honest treatment. We track a documented process we call territorial reprimarization, growth concentrating in commodity-export regions while the industrial core contracts. Manufacturing’s share of Brazilian GDP fell from 25.6 percent in 2000 to 20.8 percent in 2022, and a trade surplus in manufactured goods has become a substantial deficit. Chinese investment in Brazilian EV manufacturing is often framed domestically as a sign of progress; we think it deserves a closer look, since early-mover advantages captured by Chinese firms can make it harder for Brazil to build that same industrial capacity independently down the road.

None of this makes Brazil’s diplomatic positioning —including its stance on the Ukraine peace framework, its criticism of dollar dominance, or its reversal on Huawei’s role in 5G simply a matter of principled nonalignment, as it is often described in Brasília. But nor is it simply capitulation. It reflects the genuine and difficult position of a country whose leading agricultural and mining constituencies depend heavily on Chinese demand, operating within a democratic system that has to answer to those constituencies at the ballot box.

This is the Middle Power Trap we describe in the book: not a single bad decision by any one government, but an accumulation of individually rational choices, made under real economic pressure, that collectively narrow a country’s future room to maneuver. It is a trap that could close around any commodity-dependent democracy, including Australia, in sectors where it does not currently hold the leverage it holds in iron ore.

Australia: A Different Set of Cards

Australia’s experience offers an instructive counterpoint — not because Australian policymakers were wiser or more resolute, but because Australia happened to be holding a different hand.

In 2020, Canberra called for an independent investigation into the origins of COVID-19. Beijing’s answer was swift and severe: sweeping restrictions on Australian barley, wine, coal, beef, lobster, copper, and timber, calibrated to inflict maximum economic pain and force a political retraction. Australia weathered the pressure. Exporters redirected shipments, coal cargoes changed course mid-voyage, winemakers rebuilt distribution across the UK, India, and Southeast Asia. Canberra pursued what we call managed dualism: deepening security commitments through AUKUS, the Quad, and Five Eyes, while maintaining pragmatic commercial engagement with Beijing wherever trade remained open.

But intellectual honesty requires naming the structural advantage underneath that success. Iron ore, the one commodity conspicuously absent from Beijing’s sanctions list, supplies a significant share of the raw material behind China’s own construction and military-industrial base. That is not a market Beijing can walk away from without inflicting serious damage on itself. Australia’s leverage came, in large part, from geology it did not choose and decades of prior alliance investment made by governments long since out of office. Brazil’s principal exports, soy, iron ore sold into a more diversified but still China-dependent chain, and now oil, have not, so far, handed it a comparably irreducible chokehold on any single Chinese sector. That is a difference in the cards each country was dealt, not a difference in resolve.

Why This Is Everyone’s Test, Not Just Theirs

It would be a mistake to read this as a story about which of two countries got it right. It is a story about the terms on which every resource-exporting democracy will eventually be tested, and those terms are set almost entirely by factors, geology, existing alliances, the specific commodity mix a country happens to export, that no single election or administration can quickly change. The global transition to green energy is about to sharply increase demand for cobalt, lithium, and rare earths, precisely the minerals whose processing capacity sits concentrated inside China’s informal empire. Every middle power holding those resources is walking toward the same set of hard choices that Australia and Brazil have already begun navigating, and most will not have Australia’s particular structural luck.

There is also a handicap specific to democracies that deserves more attention than it gets, and it applies equally in Brasília and Canberra. Elected governments operate on three-to-four-year cycles and face constant pressure to deliver visible economic results now. The infrastructure of the informal empire, a deep-water port, a dominant refining position, takes fifteen years or more to build. By the time any elected government fully recognizes how far the walls have closed in, the concrete, quite literally, is already dry. That is not a verdict on any particular leader’s judgment. It is a mismatch between the electoral clock and the imperial one, and every democracy exporting raw materials shares it.

The Verdict

Australia and Brazil are not opposite case studies in virtue and vice. They are two data points on the same curve, shaped by the same architecture of dependence and separated mainly by which commodities each happened to hold, and what alliance and industrial investments prior generations had already made by the time the test arrived. Australia’s advantage rests on a geological accident and an alliance structure that must be continuously maintained not a permanent or portable guarantee against more calibrated future pressure aimed at sectors where it holds no such leverage. Brazil’s position is not fixed either. Rebuilding manufacturing capacity and diversifying its trade relationships would require the kind of sustained, multi-decade investment that democratic electoral cycles make genuinely difficult to sustain, in Brazil or anywhere else.

What both cases make unmistakably clear is that the informal empire does not operate through conquest. It operates through the patient accumulation of structural leverage, in ports, in refineries, in trade contracts signed years before anyone notices the pattern, that can erode the sovereignty of any nation not paying close enough attention, regardless of that nation’s intentions or resolve. The room for strategic ambiguity that middle powers once enjoyed is closing for all of them. The question every resource-exporting democracy now faces, Brazil and Australia very much included, but far from alone, is not whether it will eventually be tested. It is whether it will have built, well before that test arrives, the structural leverage and the political room to choose its own answer.

The Australian, Brazilian, and Chinese Dynamic: An Inquiry into the Evolving Global Order, co-authored with Kenneth Maxwell, is out this week.

A Tale of Two Middle Powers

By Robbin Laird

A recent CSIS conversation between the Honorable Michael Obadal, Under Secretary of the Army, and the Honorable Ryan McCarthy, former Secretary of the Army moderated by Jerry McGinn of CSIS’s Center for the Study of the Industrial Base offered a candid look at how senior Army leadership is thinking about the service’s future. I listened to it shortly after finishing my own book, Building the Impact Force: Marine Corps Transformation in an Age of Chaos, which tracks how the Marine Corps has spent the past year reworking its own concept of what it is for.

Reading the two side by side turns out to be a genuinely useful exercise. The Army and the Marine Corps are answering the same set of pressures, Chinese and Russian modernization, the cost-curve problem in drone and missile warfare, the compression of decision timelines, but they are doing so from very different starting points, with very different institutional histories bearing down on the answers they can give.

What follows is that comparison, and then a closer look at the one piece of the Marine Corps story that does not yet have a real Army analog: what Marine aviation now delivers to make the “impact force” concept work in practice rather than on a briefing slide.

Two Services, One Inflection Point

Obadal frames the Army’s task around two durable outputs that only a dominant land force can provide, persistent presence and decisive control, and around a picture of the Army as the foundation the rest of the joint force stands on. It seizes and holds the terrain, the ports, and the airfields that let the other services exercise dominance in their own domains. Read against that framing, my book describes the Marine Corps arriving from a very different direction at a strikingly similar self-conception.

Brigadier General Christopher Haar’s account of the Force Design Update captures the shift in a single line: “The original document was all about killing ships for the stand-in force. And then we realized, you know what, we’re better as the JTAC for the joint force, using our sensors, using our eyes, killing occasionally, but using our time and our capabilities to enable the rest of the joint force.” Both services, in other words, have concluded from opposite ends of the force-design spectrum that their highest-value contribution in a peer fight is not what they can shoot themselves, but how much more lethal, aware, and coordinated they can make everyone else.

That shared conclusion runs through several parallel threads in both conversations.

Mass over exquisite performance. Obadal describes a deliberate 60/40 shift toward low-cost, producible-at-scale munitions over the smaller, exquisite systems the Army has spent decades perfecting, an explicit trade of unit performance for magazine depth, made because the cost curve in drone and missile warfare now favors the attacker. The Marine Corps reached for the same logic by a different route. The book’s account of the October 2025 Force Design Update describes a Corps that chose to preserve breadth rather than build an ever more specialized “ship killer” stand-in force, retaining the 4th Marine Regiment as conventional infantry rather than converting it into a third Marine Littoral Regiment, precisely so the service is not over-optimized for a single theater and a single threat. Retired Commandant Gen. Charles Krulak pressed that argument hard in the book’s account of the decision, and it is worth noting how directly it echoes Obadal’s own logic: both services independently concluded that over-specializing against today’s most visible threat is itself a form of risk.

Drones and AI as enablers, not replacements. Obadal and McCarthy insist that Ukraine has not made the tank obsolete; what it has shown is what happens when neither side has an actual maneuver combined-arms force to bring to bear. Their lesson is that drones unburden soldiers rather than replace them, and that the Army’s new data-centric C2 effort is compressing fires workflows while deliberately preserving human decision points rather than automating them away. My book draws an almost identical lesson from Steel Knight 25, the Marine Corps’ December 2025 exercise: Ukraine’s real lesson for Marines was adaptation speed, not a tactical template to be copied wholesale. Decision-Centric Aviation Operations, or DCAO, is built explicitly around that same instinct, treating data as the primary weapon while keeping decision authority delegated to the edge rather than centralized in a rear headquarters.

Command and control as the actual bottleneck. Both conversations locate the hardest problem not in hardware but in culture. Obadal’s Army C2 effort aims to replace some fifteen stovepiped warfighting-function systems with a single data layer accessible through one common set of applications, a project that is as much about breaking organizational habits as it is about software. Steel Knight 25 exposed the Marine Corps’ version of the same friction: modern C5ISR gives a captain commanding a distributed node access to F-35, MQ-9, and allied sensor data, and the ability to coordinate naval, Air Force, and Army fires but, as the book puts it, “capability without clear tasking and appropriate authority becomes wasted potential.” Both services have arrived at the same uncomfortable conclusion: the limiting factor on the new way of war is less technology than institutional willingness to delegate.

Where the Paths Diverge

The similarities are real, but so are the differences, and they follow directly from scale and mission. McCarthy’s description of the Army as “the show that never ends”, 60 percent of combatant commanders’ worldwide requirements on a good day, plus hurricane response, border missions, and disaster relief at home, describes an institution whose baseline identity is standing global demand. It cannot choose when to show up; it is already everywhere, all the time, by design. The Marine Corps’ impact-force argument, by contrast, is built around a much narrower proposition: that a comparatively small force can generate effects disproportionate to its size by persisting forward as sensing and decision-enabling connective tissue inside contested space, not by being everywhere at once. Scale, in other words, is not a detail here. It is the reason the two services can share a philosophy and still end up with different force designs.

That difference in scale produces a difference in how each service is managing its hardest modernization trade. The Army’s aviation dilemma centers on a single, enormous fleet decision: sunsetting roughly 2,000 Black Hawk helicopters in favor of the MV-75 Cheyenne tiltrotor, whose roughly two-and-a-half-times range advantage is meant to solve what McCarthy frames starkly as the Scarborough Shoal-to-Hainan problem, 550 nautical miles that the Black Hawk simply cannot cover without extensive aerial refueling. Obadal and McCarthy are candid that this will be one of the harder political sells the Army faces, precisely because the congressional committees with jurisdiction sit atop the jobs and factories tied to the existing fleet. Retiring a platform is never just an engineering decision; it is also a fight over districts and payrolls, and both men know it.

The Marine Corps, by contrast, largely already made its version of that bet when it fielded the MV-22 Osprey a generation ago. Its current aviation transformation is not about replacing a legacy platform with a new one; it is about re-purposing an already-diverse inventory — F-35, MQ-9, MV-22, CH-53K, and H-1 — as nodes in a kill web rather than as independent shooters. That is a fundamentally different kind of modernization problem. The Army is trying to decide whether to make a single, politically fraught platform swap. The Marine Corps has already made its platform bets and is now working the harder, less visible problem of getting five very different aircraft to behave as one integrated sensing-and-decision network.

The two services also differ in how they talk about their own force design logic, and the difference in vocabulary is itself revealing. The Army’s framework is explicitly categorical: every piece of equipment is sorted into legacy (divest), enduring (sustain and modernize), or modern (invest aggressively), Humvee, Black Hawk, and Group 4/5 attack-reconnaissance UAS, respectively. It is a taxonomy built around ownership: what do we keep, what do we shed, what do we buy next. The Marine Corps’ framework, as the book documents, is less about platform categories and more about a change in the metric of success itself. An F-35 sortie that discovers ten targets and cues joint strikes without firing its own weapons is now judged to generate more impact than the same sortie engaging two targets independently. The Army is still fundamentally organizing around what it owns; the Marine Corps has shifted toward organizing around what it enables. That is not a small semantic distinction.It changes how each service will eventually measure whether its own modernization program is actually working.

What Marine Aviation Delivers to the Impact Force

This is the piece of the Marine Corps story that has no close Army analog yet, and it is worth spelling out in some detail, because it is the mechanism that actually makes “impact force” more than a slogan on a strategy document.

The 2026 Marine Aviation Plan elevates Aviation Ground Support to a formal seventh function of Marine aviation, on par with assault support and offensive air support, a recognition that the ability to generate fuel, power, maintenance, and airfield services from austere, shifting locations is what actually determines whether distributed operations work at all. It is easy to overlook this kind of decision because it is unglamorous; nobody writes headlines about fuel bladders and expedient runway repair. But an impact force that cannot sustain itself forward is not an impact force, it is a stranded one, and the plan’s authors clearly understood that.

That plan organizes around Project Eagle’s three horizons — Fight Tonight (2026-2030), Bridge the Gap (2031-2035), and Future Fight (2036-2040) — and around two operational concepts that Steel Knight 25 tested under real conditions rather than on a whiteboard. Distributed Aviation Operations, or DAO, sustains combat power from a hub-spoke-node network of expeditionary sites rather than fixed bases. Decision-Centric Aviation Operations, or DCAO, treats data flow not organic weapons as the thing that actually generates impact. Put the two together and you get the operating logic behind everything that follows: disperse the aircraft, concentrate the data.

Against that framework, each platform’s role changes in ways that would have been unrecognizable a decade ago.

F-35 functions less as a strike aircraft and more as a sensor-shooter node, an airborne “brain,” in one Marine officer’s phrase, that can discover targets inside threat rings other platforms cannot penetrate, manage networks as an airborne C2 node, and cue joint fires from Navy ships and Air Force bombers, employing its own weapons only when sensor-to-shooter timelines demand immediate action. The jet’s stealth and sensor suite were built for a different era’s assumptions about air combat; what the impact-force logic does is repurpose those same attributes as a discovery-and-cueing capability that happens to also be able to strike when nothing else can.

MQ-9 delivers persistence rather than firepower, 20-plus-hour orbits that feed targeting cycles and sustain pattern-of-life analysis, turning episodic strikes into sustained campaigns. An aircraft that simply stays overhead longer than anything else in the inventory changes the nature of what can be tracked and understood over time, which is a different kind of contribution than lethality, but not a lesser one.

MV-22 and CH-53K enable impact through movement rather than fires: inserting forces, sustaining dispersed nodes, and extracting personnel under contested conditions. The CH-53K’s lift advantage, once fully fielded, changes the DAO logistics math directly by letting one sortie do what once took several MV-22 sorties which matters enormously in a hub-spoke-node concept where every sortie flown is also a sortie that a contested adversary might detect and target.

H-1, far from being a legacy platform on its way out, has gained new relevance through digital interoperability rather than through any new hardware. HMLA-267, the first Marine H-1 squadron to achieve full digital interoperability, demonstrated H-1s operating as airborne gateways that translate between legacy and modern communications systems and cue joint fires. That is proof, in a very concrete form, that the impact-force logic applies to older platforms as much as to new ones, the transformation here is not about buying something new but about rewiring what already exists.

The throughline across all five platforms is the same one the book keeps returning to: aviation’s value is no longer measured primarily by what any single aircraft can do on its own, but by how much faster, more lethal, and more survivable it makes the rest of the joint force. That is a genuinely different design philosophy than the Army’s current aviation debate, which for entirely legitimate reasons tied to the size and age of the Black Hawk fleet is still substantially a platform-replacement conversation.

The Marine Corps got to skip that conversation, in effect, because it already had it a generation ago with the Osprey. What it is doing now is the second-order work of turning a diverse inventory into a coherent network, and that work is arguably harder than any single procurement decision, because there is no single program office or committee hearing that can be pointed at as the finish line.

The Common Unfinished Business

For all the differences, both services land on the same honest caveat, and it is to their credit that neither leadership team tried to paper over it. Steel Knight 25 revealed, in the book’s words, that “capability gaps remain severe,” that “logistics is the binding constraint,” and that “culture lags technology” or the Marine Corps has platforms and networks that enable impact-force operations but not yet the command cultures and training methodologies to fully exploit them.

Obadal and McCarthy describe an almost identical unfinished agenda for the Army: contested logistics as an unsolved problem, a fires portfolio still being rebalanced between exquisite and producible-at-scale, and a political fight still ahead over trading a large, understood fleet for a smaller number of more capable aircraft. Neither service has finished the transformation it is describing; both are candid that the exercises and studies driving these changes have exposed as many gaps as they have closed. That candor, frankly, is one of the more reassuring things about both conversations. Services that claim to have already solved their hardest problems are the ones worth worrying about; services that can name their own gaps in public are the ones doing the actual work.

Looking Ahead

The most interesting open question raised by putting these two conversations side by side is whether the Army’s coming tiltrotor decision will end up rhyming with the Marine Corps’ aviation experience rather than with the Army’s own legacy framework. The MV-75 Cheyenne is, functionally, the Army’s belated version of the range and speed advantage the Marine Corps has flown with since the MV-22 entered service. If the Army follows the Marine Corps’ arc, the more consequential change will not be the airframe itself. It will be whether the Army treats the Cheyenne as a networked sensor-shooter node in a joint kill web, the way the Corps has come to treat the Osprey, rather than simply as a faster Black Hawk that happens to fly further and carry the same mission set it always has.

That is the thread I plan to pick up next: how the tiltrotor factor could reshape Army aviation in the same way digital interoperability has already reshaped what Marine aviation delivers. The airframe is the easy part of the story to tell, because it is the part with a ribbon-cutting and a production line attached to it. The harder and more consequential part, whether the institution around the airframe changes how it thinks about what aviation is for and is the part the Marine Corps’ experience suggests is worth watching most closely.

The Way Ahead for the US Army: A CSIS Baseline Discussion

By Robbin Laird

I have never had a career. I want to say that plainly, at the start, because it is the truest thing I know about my own life, and because it took me the better part of four decades to be able to say it without apology.

A career is a path. It has rungs. Someone above you has already climbed it, and if you are diligent and reasonably talented, you climb it too, a little differently, a little later, but recognizably the same ascent. A career can be described in a résumé, because a résumé is really just a map of a road that already existed before you set foot on it.

What I have had instead is something closer to what the Greeks meant by an odyssey: not a road but a voyage, undertaken because the winds and the gods and one’s own stubbornness kept redirecting the journey, with no guarantee of the shape it would take until long after it had already taken it. You only see the shore you were making for once you have, half by accident, arrived.

I did not plan to spend my life moving between Soviet nuclear doctrine and French defense politics and NATO’s institutional grammar and now, in recent years, the architecture of a kill web that did not exist as a concept when I started. Nobody plans that. There was no professional ladder marked “Soviet affairs, then alliance transformation, then maritime autonomous systems, then Indo-Pacific chokepoints.” I ended up walking that route because each destination turned out, on arrival, to be adjacent to some other unanswered question, and I followed the question rather than the career.

This is where the phrase “thinking outside the box” has always struck me as strange, almost as if it were describing a foreign country whose customs I’d heard about but never visited. The phrase presumes a box. It presumes that most people, most of the time, are working inside some agreed structure, a discipline, a bureaucracy, a set of categories that everyone tacitly accepts and that every so often, as a special act of creativity or courage, someone steps outside it.

But I have never experienced the box as a default condition to be escaped. I have never quite known how to think inside one. When I was studying under Brzezinski at Columbia, the box would have been to treat Soviet nuclear policy as a self-contained technical subject, insulated from French politics, from the sociology of European elites, from the psychology of alliance management.

Nobody who actually wanted to understand the Cold War could afford that box. The subject itself refused to sit still inside a discipline. If you wanted to understand why French nuclear policy looked the way it did, you had to understand de Gaulle’s wounded pride about 1940, the peculiar independence of the French officer corps, the way Soviet planners thought about escalation, and the raw fact that nuclear weapons had made certain older categories of strategic thought simply obsolete. There was no box that contained all of that. There still isn’t.

So when people later called this kind of work “interdisciplinary” or praised it for its “outside the box thinking,” I always felt a small, private confusion, because from the inside it never felt like an achievement. It felt like the only way to actually see the thing I was looking at. Calling it outside-the-box thinking is a bit like calling a fish “outside the desert.” The fish isn’t achieving anything remarkable by living in water. It would be remarkable, and probably fatal, for the fish to try living in the desert because someone insisted that’s where the career path was.

I think this is worth saying honestly because so much of professional life, especially in policy and defense circles, is organized around box-shaped incentives even when the problems themselves are not box-shaped. Institutions like boxes. They fund boxes, they promote people who stay inside their box, they build career ladders that only make sense if the discipline holds still. A defense analyst who works on air power is expected to stay in air power. A person trained in Soviet studies is expected to remain a Soviet-studies person even after the Soviet Union no longer exists, or to gracefully retire once the box itself is dissolved.

I watched a great many capable people get trapped this way, not because they lacked the wit to see connections across categories, but because the categories were what their institutions paid them to defend. My own path, not having gone through those institutions in the expected way, never gave me a box worth defending.

What it gave me instead was a network — collaborators like Ed Timperlake, with whom the “kill web” idea took shape not as a single insight but as years of conversation that kept refusing to stay inside the older categories of platform-centric warfare; like Kenneth Maxwell, with whom the connections between eighteenth-century Atlantic history and twenty-first-century Indo-Pacific strategy turned out, disconcertingly, not to be a stretch at all, but a continuation of the same old questions about chokepoints, empire, and the fragility of maritime order. like my wife Murielle, whose French vantage point on European defense has spent thirty years reminding me that there is no “American” box either, only a set of American assumptions that look, from Paris, exactly as parochial as anyone else’s.

An odyssey, unlike a career, has no natural retirement age, because it was never organized around institutional time in the first place. Careers end when the institution stops needing you, or when your rung is filled, or when the pension math works out. Odysseys end when the voyage is over, and you often don’t know that in advance either.

I am, by the calendar, well past the age at which a career would have concluded. And yet the work has not concluded, because it was never a career to begin with, it is still the same voyage, now touching Guam and the Luzon Corridor and the Netherlands and Canada as connective nodes in a trans-Atlantic tiltrotor architecture, now returning to Australia for the Williams Foundation seminars, now circling back, as odysseys do, to older questions about allied cohesion that I first encountered as a young man trying to understand why France would not simply do what Washington wanted.

Odysseus, after all, did not sail in a straight line either. He kept getting blown off course toward islands that turned out, afterward, to have taught him something he needed for Ithaca. I have been blown toward MISR officers and drone wars and Marine Corps transformation and CH-53K lift requirements, and each time it has felt less like a detour and more like the sea doing what the sea does, carrying you toward the next necessary thing whether or not it was on your itinerary.

There is a cost to this, and I don’t want to romanticize it away. An odyssey does not offer the reassurances a career offers. There is no professional society that fully claims you, because you belong to several at once and fully to none. There is no single credentialing body that can certify what you do, because what you do keeps changing its shape. When you meet someone at a conference and they ask, “What do you do?”, there is no clean one-line answer, because the honest answer requires either five minutes or an evasion.

I have mostly stopped trying to compress it. I say I am a defense analyst, which is true, and then I let the actual texture of the work — kill webs, allied architecture, maritime autonomy, the history of Portuguese chokepoints, Marine Corps combat readiness — reveal itself only if the conversation goes on long enough to deserve it.

But there is also a freedom in it that I would not trade for the reassurances. Nobody can take away my box, because I never had one to lose. When institutions decide a subject is no longer fashionable, when “great power competition” replaces “counterinsurgency” as the phrase of the moment, or when a service branch’s budget priorities shift and strand an entire cohort of specialists, the people who built their careers around the old box are stranded with it. I have simply never been stranded in that way, because I was never standing on a single piece of ground.

When the Cold War ended and an entire generation of Soviet specialists found themselves professionally orphaned, the questions I had actually been pursuing, how do alliances hold together under strain, how does technology reorganize the categories of warfare, how do maritime chokepoints structure great-power behavior, turned out to still be exactly the right questions, just pointed at new geography. The box collapsed. The voyage did not, because it was never contained by the box in the first place.

I think what I am really trying to say is that “thinking outside the box” flatters an achievement that, for some of us, was never an achievement at all  for it was simply the absence of an alternative.

I did not choose to think this way as a strategy. I have never been able to think any other way, and I suspect that if I had tried to force myself into a single box for the sake of a tidier career, I would have done second-rate work in it, because my mind does not naturally stay where it’s put.

The people who do their best work inside well-defined boxes are not lesser thinkers than I am. They are simply built differently, and disciplines need them, because depth within a box is its own kind of rigor that breadth across boxes cannot replace. But I was not built that way, and pretending otherwise for the sake of a conventional career would have been its own kind of failure, a fish insisting it could learn to love the desert, if only it tried hard enough.

So when I look back now, at four decades that a résumé would struggle to summarize and a career counselor would have found alarming, I don’t see a lack of focus. I see a single sustained question, restated across geographies and technologies and eras: how do serious actors — nations, alliances, services, individuals — organize themselves to act coherently under conditions of danger and uncertainty?

That question does not respect academic departments. It does not respect service branch boundaries. It does not respect the line between history and current events, or between an analyst’s desk and a defense company’s engineering floor. It has taken me from Brzezinski’s seminar room to the flight line at Decimomannu, from French nuclear doctrine to Ghost Shark unmanned vessels off the Australian coast, from eighteenth-century Atlantic trade routes to the Luzon Economic Corridor, because that is where the question kept leading.

An odyssey does not end with a gold watch. It ends, if it ends at all, with the traveler recognizing the shore he has been making for all along and even then, I suspect, only provisionally, because the next question is usually already visible on the horizon.

I am content with that. I was never going to be the kind of person who thought inside a box, because I never quite understood what the box was for. I only ever understood the voyage, and the peculiar, unglamorous discipline of following a real question wherever it actually leads, institutional convenience be damned.

That is not a career. It is, I think, something better. messier, less secure, but truer to the actual shape of the problems I have spent my life trying to understand.

This article first appeared on Real Clear Defense.

https://www.realcleardefense.com/articles/2026/07/15/the_odyssey_not_the_career_1194566.html

Note: Next year Kenneth Maxwell and I will publish our co-autobigraphy which have been working on for several years.

Two 20th Century Men in the 21st Century: The Decades of Change

This book is a double memoir, a lived history, and a warning from two men who have spent seventy years watching the world change and now wonder whether the 21st century still remembers how it got here. It is written by a strategic analyst and a historian who turn their trained gaze on their own lives and times, decade by decade, from the 1950s to the pandemic years.​

Two 20th century men, born into a world before computers, mobile phones, and permanent connectivity, track how the social, political, and cultural ecosystems of the West have been dismantled and rebuilt across eight tumultuous decades. They are not celebrities and not victims, but witnesses and participants moving from postwar austerity and empire, through decolonization, Cold War showdowns, student revolts, Thatcherism, Reaganism, the end of the Soviet Union, and into the age of Brexit, Trump, and Covid-19.​

The book fuses personal narrative with professional analysis: a historian of Portugal, Latin America, and Britain and a defense and strategic affairs analyst mine their own archives, photos, journals, books, and publications, to reconstruct what it actually felt like to live these decades as they unfolded. Each chapter takes a single decade, from the 1950s through the second decade of the 21st century, interweaving family, careers, and friendships with the upheavals of war, revolution, technological disruption, and political crisis.​

The tone is unsentimental and unsparing: there is no royal-style grievance tour, no confessional therapy, and no attempt to inflate private dramas into world-historical events. Instead, the authors use their own trajectories from grammar school and provincial beginnings to Oxbridge, Harvard, policy think tanks, newsrooms, and defense ministries to test what the public record gets wrong, where memory fails, and how the narratives of the powerful erase the experiences of ordinary but engaged citizens.​

Running through the book is a stark question: what happens when a civilization forgets its own lived history and outsources memory to Wikipedia, feeds, and curated outrage? One author’s discovery that his online biography had casually relocated his education to another continent becomes a metaphor for a broader crisis of accuracy, authority, and truth in the public domain. The arrival of the Covid-19 pandemic at the start of the 2020s is treated not as an isolated shock, but as the hinge into a new historical period in which institutional trust, expertise, and shared narratives are visibly unraveling.​

For readers who came of age after the Cold War, this book offers an unvarnished guide to the world they inherited but never saw being built and to the freedoms, habits, and illusions now at risk. For those who lived through parts of this history, it offers a bracing reckoning: what was gained, what was squandered, and what it means to be “two 20th century men” still trying to make sense of a 21st century that often seems determined to amputate its own past.​