Three Days That Lasted Years

The Morning of February 24

At approximately 4:00 a.m. Kyiv time on February 24, 2022, Russian forces crossed into Ukraine from the north, east, and south simultaneously, while cruise missiles struck airfields, command centers, and fuel depots in cities across the country. Vladimir Putin's television address, recorded and broadcast just before the strikes began, described the operation as a "special military operation" aimed at the "demilitarization and denazification" of Ukraine. Within hours, Russian armor was moving on Kyiv from Belarusian territory to the north, an axis of advance that would, within days, put Russian forces at the outskirts of the capital.

This much is settled fact, established by satellite imagery, open-source verification, and the reporting of journalists who were present. What makes February 24, 2022, the right place to open a course on international relations is not the invasion itself. Wars begin; it is, unfortunately, one of the most well-documented regularities of human political life. What makes this date pedagogically useful is what happened to the world's predictions in the seventy-two hours before and after it — because the predictions were made by serious people with serious information, and an unusually large share of them turned out to be wrong, in instructive and specific ways.

This course treats the Russian invasion of Ukraine as a standing laboratory: a single, extraordinarily well-documented case against which every theoretical tool the course introduces will be tested, not once, but repeatedly, from different angles, across ten units. That means starting with discipline about dates. Every claim about the war's course in this unit and in the units that follow will be marked with the point in time it was true as of — because the war did not end in three days, or in three months, and this course is being written while it is still being fought. As of early 2026, the war continues into its fifth year, and any account of "how it turned out" has to be read as a snapshot, not a verdict.

What the Forecasters Got Right, and What They Got Wrong

The single most consequential prediction going into February 2022 came from U.S. intelligence agencies, and it is worth being precise about exactly what they predicted, because the popular memory of this period tends to flatten a genuinely mixed record into either "the U.S. nailed it" or "the U.S. cried wolf." Neither is accurate.

Beginning in the autumn of 2021 and accelerating through the winter, the Biden administration made an unusual choice: it began publicly, repeatedly, and specifically warning that Russia was preparing a full-scale invasion of Ukraine, releasing detailed troop-movement assessments and even preempting anticipated Russian disinformation operations (a false-flag pretext involving a staged attack, which U.S. officials described publicly before it could be used) as a deliberate strategy to strip Moscow of the element of surprise. This was itself a break from precedent — governments do not typically publicize sensitive intelligence about an adversary's war plans, because doing so reveals sources and methods. The decision to do so anyway reflected a judgment, made in Washington, that public warning might complicate Russian planning even if it could not stop it outright.

On the narrow question of whether Russia would invade, and roughly when, American intelligence was right, and right early, when a great deal of expert opinion elsewhere doubted it. On the question of what would happen once the invasion began, American intelligence was substantially wrong, in the same direction as most other observers: it anticipated a rapid Russian victory, with assessments reportedly estimating that Kyiv could fall within days, that the Ukrainian government would be forced to flee or capitulate, and that organized Ukrainian resistance would collapse or fragment into an insurgency within a matter of weeks. This was not a fringe view. It was close to consensus among Western defense officials, and it shaped the initial offer, reportedly extended to President Zelensky in the war's first hours, of evacuation from Kyiv.

It is worth naming specific, dated predictions rather than gesturing vaguely at "the consensus," because a scorecard is only useful if it is actually a scorecard. U.S. Chairman of the Joint Chiefs of Staff General Mark Milley reportedly told members of Congress in a classified briefing in late January 2022 that Kyiv could fall within 72 hours of a Russian invasion — a specific, dated, falsifiable prediction that turned out to be wrong, and one whose existence became public only after the fact, through subsequent congressional and press reporting. Reporting from the war's opening days described broadly similar assessments circulating among European defense ministries, several of which reportedly anticipated a Ukrainian government collapse or exile within days rather than the sustained defense that followed. On the other end of the spectrum, a smaller number of specialists in irregular and urban warfare, along with some Ukrainian officials themselves, had argued publicly in the weeks before the invasion that Russia would face a much harder fight than conventional order-of-battle comparisons suggested, pointing to the post-2014 reconstruction of Ukraine's territorial defense forces and the country's dense urban terrain as reasons a rapid armored thrust toward Kyiv would be more exposed than it looked on paper. Their view was a minority position in Western defense commentary through January 2022, and it is precisely the kind of minority-but-correct prediction that a serious scorecard has to credit explicitly, rather than being reconstructed after the fact into a consensus that did not actually exist at the time.

This matters for a reason beyond simple bookkeeping: the pattern of who got which piece right is itself evidence about which analytical inputs were doing the work. The actors who got the invasion timing right were drawing chiefly on capability and posture indicators — satellite imagery of fuel, ammunition, and blood-supply logistics being pre-positioned, patterns of unit movement, and diplomatic signals — a strength of the capability-counting method examined below. The actors who got the outcome right were drawing on a different kind of evidence entirely: qualitative judgments about Ukrainian institutional capacity, morale, and the lessons Ukraine's own military had drawn from its 2014-15 defeats in the Donbas, evidence that a purely quantitative order-of-battle comparison does not capture at all. The lesson worth taking from naming these specific, dated predictions individually, rather than flattening them into "the experts were wrong," is that different predictive methods have different, identifiable blind spots — and knowing which method produced a given forecast tells you something real about how much weight to put on it.

European governments, by contrast, were considerably more divided than the American assessment, and many were more skeptical that an invasion would happen at all. German officials in particular, along with a substantial body of opinion across Western European capitals, treated a full-scale invasion as one possibility among several rather than as the likely outcome — a skepticism rooted partly in the assumption that a war of this scale would be so economically and diplomatically costly to Russia that a rational Kremlin would not choose it, and partly in years of prior crises (2014's annexation of Crimea, repeated troop buildups on Ukraine's border in 2021 that did not result in invasion) that had trained observers to expect brinkmanship rather than full war. French President Emmanuel Macron traveled to Moscow in early February 2022 for direct talks with Putin, a diplomatic effort premised on the belief that war was still avertable through negotiation. It was not.

Why the Fall-of-Kyiv Prediction Failed

The gap between the accurate invasion forecast and the inaccurate outcome forecast is worth dwelling on, because it is not simply a story of some analysts being smarter than others. It is a story about two different kinds of military analysis, and the second kind failed in a specific, recurring way that shows up throughout this course.

The first kind of analysis is capability-counting, sometimes called order-of-battle analysis: tallying the tanks, aircraft, artillery pieces, and troop formations arrayed on each side, and inferring an outcome from the resulting balance of forces. By this measure, the mismatch between Russia and Ukraine in February 2022 was enormous. Russia fielded one of the largest standing militaries in the world, with a defense budget roughly ten times Ukraine's, a substantial air force, and a nuclear arsenal. Ukraine's military, though it had been rebuilding since 2014, remained a fraction of Russia's size on every conventional metric. A purely capability-counting analysis, and there were many produced in the war's opening days, pointed toward rapid Russian success.

What capability-counting leaves out is precisely what determined the war's actual opening weeks: logistics, command quality, and will. The historian Stephen Kotkin, whose register this unit borrows deliberately — evidence assembled first, generalization drawn cautiously and only afterward — has emphasized in his commentary on the war's opening phase that the Russian plan itself assumed a walkover: units moved toward Kyiv with parade uniforms and insufficient fuel and ammunition resupply, reflecting an assumption, baked into the invasion's very planning, that Ukrainian resistance would be minimal or would evaporate once decapitation strikes removed the government. That assumption was wrong, and its wrongness cascaded through the entire campaign. Russian columns advancing on Kyiv outran their own supply lines within days, producing the now-famous multi-mile traffic jam of stalled vehicles north of the capital — a logistics failure, not a firepower failure, and precisely the kind of variable that a raw count of tanks and troops does not capture.

Ukrainian resistance, meanwhile, was not a passive given; it was actively organized, in real time, out of a population and a government that had spent the preceding eight years, since the loss of Crimea and the outbreak of war in the Donbas in 2014, rebuilding both military capacity and a sense of what was at stake. Territorial defense units were mobilized within days. Javelin anti-tank missiles and Stinger anti-aircraft missiles, supplied by the United States and allies in the months before the invasion — itself a form of deterrence-adjacent signaling worth returning to — proved lethal against Russian armor moving without adequate infantry escort or air cover. None of this was visible to an analysis that only counted hardware on each side's inventory list.

It is worth being precise about why capability-counting fails in this specific way, rather than simply noting that it failed, because the failure mode recurs throughout this course and is worth naming as a general pattern rather than a one-off surprise. Order-of-battle analysis is, at its core, a snapshot method: it measures what forces exist and where they are positioned at a given moment, treating a tank, an aircraft, or a battalion as a roughly fungible unit of combat power that will perform close to its rated capability whenever it is employed. What this method structurally cannot measure, because it is not a property of the equipment at all, is whether the humans operating that equipment are well-led, adequately supplied at the point of contact, and motivated to fight rather than to surrender or retreat — variables that live in command culture, logistics execution, and political will, none of which show up on an inventory list no matter how detailed. A tank with a full fuel tank and a trained, motivated crew and a supply line that will still be intact in a week is not the same military asset as a nominally identical tank without those things, but a count of "tanks" treats them as identical, because counting is what the method does well and assessing organizational quality is what it does not do at all.

Kotkin's evidence-first register is worth dwelling on here specifically because it supplies the corrective discipline this failure mode calls for. Kotkin's method, observable across his public commentary on the war's opening phase, is to resist generalizing about "Russian military weakness" or "Ukrainian resolve" as abstract character traits, and instead to trace the concrete, documentable mechanics that produced a specific outcome: which units moved on which roads, with what documented fuel and ammunition resupply (or lack of it), under what command structure, facing what specific Ukrainian units equipped with what specific weapons at what specific chokepoints. This is a deliberately unglamorous method — it produces a account built out of logistics reports and unit-level detail rather than sweeping civilizational or cultural claims — and its payoff is precision: it lets you say exactly which link in the causal chain broke (fuel resupply north of Kyiv; overextended columns without infantry screening; a Ukrainian anti-armor ambush at a specific chokepoint) rather than reaching for an explanation pitched at the level of national character, an explanation that would be nearly impossible to falsify and would generalize badly to the next case. The discipline this course borrows from Kotkin throughout is exactly this: state what the evidence documents before reaching for what it might mean.

❓

Concept Check

Order-of-battle analysis (counting tanks, troops, and aircraft) accurately predicted the invasion would be a mismatch on paper but failed to predict Kyiv holding. What two categories of variable does capability-counting systematically underweight, and how did each one specifically alter the outcome in the war's first weeks?

▸

Capability-counting underweights logistics/sustainment and will/organization. Logistics: the Russian advance on Kyiv assumed a rapid walkover and was not resupplied adequately with fuel and ammunition, producing stalled, exposed columns vulnerable to ambush — a planning failure invisible to a static inventory count. Will and organization: Ukrainian territorial defense mobilized rapidly and used Western-supplied anti-tank and anti-aircraft missiles effectively against exposed Russian armor, reflecting years of post-2014 institutional rebuilding and a level of motivated, coordinated resistance that a raw troop-and-equipment tally does not measure. Both variables mattered more than the raw force ratio in determining what happened in the war's opening weeks, though the raw force ratio remained highly consequential over the following years of attritional fighting.

Deterrence Failed. What Does That Even Mean?

It is common to say, in retrospect, that "deterrence failed" in Ukraine before February 2022. The phrase is accurate, but it hides an important distinction that this course will return to at length in Unit 4: the difference between deterrence and assurance, and the related question of what, exactly, Western signaling in 2021 was actually trying to accomplish.

Deterrence, in the standard usage this course will build out formally later, means convincing an adversary that the costs of taking an action will outweigh the benefits, so the adversary chooses not to act. Assurance is a different project entirely: convincing an ally or partner that you will stand behind them, so that they do not feel compelled to take destabilizing action of their own (arming unilaterally, for instance, or seeking security guarantees elsewhere). The two goals can pull in opposite directions. A signal strong enough to deter an adversary may look, to that same adversary, like an alliance closing in; a signal reassuring enough to calm a partner may look, to the adversary watching, like weakness worth testing.

What did the United States and its NATO allies actually signal to Russia in the months before February 2022? The record is genuinely mixed, and this course will not pretend otherwise. On one hand, the Biden administration stated explicitly and repeatedly, well before the invasion, that U.S. troops would not fight directly in Ukraine — a statement made partly to reassure the American public and partly to avoid any accidental military escalation with a nuclear-armed power, but one that also removed, in advance, the single most credible deterrent threat available: the prospect of direct NATO involvement. On the other hand, the United States and European allies did supply Ukraine with lethal weapons in the run-up to the invasion (a form of costly signaling that a purely status-quo, non-provocative posture would have avoided), and the public intelligence disclosures described above were themselves intended as a deterrent of sorts — an attempt to convince Putin that invasion would not achieve strategic surprise and that the world would see it coming and hold Russia accountable.

Whether a differently calibrated signal — clearer red lines, an earlier and larger weapons pipeline, an explicit (if still short of NATO membership) security guarantee — would have changed Putin's decision calculus is a counterfactual this course cannot settle, and it should be treated with appropriate humility rather than false certainty. What can be said with more confidence is that the signal actually sent did not deter the invasion, and that the mix of "we will help you fight" and "we will not fight alongside you" is a genuinely difficult signal for an adversary to read, which is itself a lesson about the limits of deterrence theory in practice — a theme Unit 4 develops at much greater length with formal tools.

It is worth being precise about the mechanism by which a mixed deterrence-and-assurance signal actually breaks down, rather than simply asserting that it does. Deterrence, to function, requires an adversary to believe two things simultaneously: that the threatened costs are real, and that the threatener will actually impose them if the line is crossed. Assurance, aimed at a partner rather than an adversary, requires a different pair of beliefs: that support will actually arrive, and that it will arrive regardless of the risk to the guarantor. The trouble is that the same statement can supply one of these beliefs while undermining the other, depending on who is listening. When the Biden administration stated publicly that U.S. troops would not fight directly in Ukraine, that statement was, from Kyiv's vantage point, a genuine assurance-relevant fact — it clarified what kind of support was and was not coming, reducing Ukrainian uncertainty about the scope of Western backing. From Moscow's vantage point, the identical statement removed the single costliest and most credible deterrent option from the table before the invasion even began, because it told Russian planners, in advance and on the record, exactly which response they did not have to prepare for. A signal aimed at reassuring a partner without alarming an adversary into overreaction can, in the same breath, hand that adversary a foreclosed option it would otherwise have had to price into its own risk calculus — which is precisely why deterrence theorists have long treated deterrence and assurance as being in tension rather than as two goals a single sentence can costlessly serve at once.

Compellence theory, developed formally by Thomas Schelling and built out at length in Unit 4, supplies one more piece of the puzzle worth previewing here: deterrence asks an adversary to refrain from an action not yet taken, while compellence asks an adversary to reverse or stop an action already underway — a harder, more visible, and more escalation-prone demand. Some analysts have argued, in retrospect, that Western pre-invasion signaling was caught awkwardly between the two: not quite a credible deterrent threat (because the costliest option was ruled out in advance), and not yet a compellence campaign (because the war had not yet started, so there was nothing to compel Russia to stop). Only after February 24 did the Western response fully shift into compellence mode — sanctions and weapons transfers aimed at reversing an ongoing invasion rather than preventing one that had not yet happened — a shift in strategic category, not merely intensity, that this course asks you to notice precisely because it is easy to describe both phases loosely as "deterrence" when only one of them actually was.

❓

Concept Check

Explain, using the mechanism this unit describes, how the same public statement (that U.S. troops would not fight directly in Ukraine) could function as assurance toward Kyiv while undermining deterrence toward Moscow.

▸

Deterrence requires an adversary to believe both that threatened costs are real and that they will actually be imposed; assurance requires a partner to believe that support will arrive regardless of risk to the guarantor. The same statement can supply one belief while undermining the other because it is heard differently by its two audiences. Ukraine heard a clarification of what support was and was not coming, which reduced uncertainty and functioned as assurance. Russia heard, in advance and on the record, that the single costliest and most credible deterrent option — direct NATO military involvement — had already been removed from the table, which let Russian planners exclude that risk from their invasion calculus entirely. A statement designed to reassure a partner without provoking an adversary can therefore simultaneously hand that adversary a foreclosed risk it no longer needs to weigh, which is why deterrence and assurance are treated as being in genuine tension rather than two goals easily served by the same sentence.

"I Need Ammunition, Not a Ride"

On February 25 and 26, 2022, as Russian forces continued to advance toward Kyiv and reports circulated that the capital might fall within days, the United States reportedly offered President Volodymyr Zelensky evacuation — an offer consistent with the intelligence community's assessment that Kyiv's fall was likely imminent, and consistent with standard practice for extracting an endangered head of state before a capital falls to an invading force.

❝

"I need ammunition, not a ride."

Whatever the precise circumstances of the exchange, the statement's effect was immediate and enormous: it reframed, almost overnight, how much of the Western public understood the war, from a story about an inevitable, fast-moving Russian conquest to a story about a government and a population choosing to fight rather than flee. That reframing mattered strategically, not just emotionally. Zelensky's decision to remain in Kyiv, visible in nightly video addresses filmed on the city's streets, functioned as a costly signal of Ukrainian resolve that shaped Western publics' willingness to support continued military aid — a soft-power and information-war dimension of the conflict this course will examine in detail in Unit 8.

Snyder's Argument: The Stake Was Statehood Itself

The historian Timothy Snyder, whose work on Ukraine (most fully developed in his 2022 book The Making of Modern Ukraine and the accompanying Yale lecture course) this unit draws on directly, has argued that the war cannot be fully understood as a conventional territorial or geopolitical dispute alone. Snyder's central claim is that Ukrainian statehood and national identity were themselves among the primary things being contested by the invasion — not simply a backdrop against which a war over territory or alliances was fought, but a central stake of the war itself.

This claim rests on a specific reading of the public arguments Russian officials, including Putin himself, made both before and during the invasion. In a lengthy essay published in July 2021, months before the war began, Putin argued at length that Russians and Ukrainians are "one people," artificially divided by historical accident and Soviet-era administrative boundary-drawing, and that Ukraine's independent statehood as a distinct nation was, in essence, a historical mistake. Snyder's argument is that this is not incidental rhetoric surrounding the invasion but close to its animating premise: if Ukraine is not a genuine nation with its own history and the right to determine its own future, then an invasion aimed at reversing its independence is not, in the eyes of its planners, really a war of conquest against a sovereign neighbor at all — it is closer, in that framing, to a restoration.

Snyder's response, elaborated across his broader scholarship, is that Ukraine's national identity has a documentable history considerably older and more independent than this framing allows — one that runs through medieval Kyivan Rus' (a state whose legacy both Russia and Ukraine claim, in different and competing ways), through a distinct early modern Cossack political tradition, through a twentieth century marked by a Ukrainian famine (the Holodomor of 1932–33) that Snyder and other historians treat as a formative, though contested, moment in the consolidation of a specifically Ukrainian national memory distinct from a Russian one, and through the post-1991 experience of independent statehood itself, including the 2004 Orange Revolution and the 2014 Euromaidan protests — both mass mobilizations, Snyder argues, in which Ukrainians asserted a political identity oriented toward European integration and against Russian-aligned governance, well before this war began. On Snyder's reading, the intensity and durability of Ukrainian resistance after February 2022 is itself evidence for this deeper argument: a population is unlikely to organize the sustained, costly resistance Ukraine mounted for a state it did not consider genuinely its own.

This is a strong claim, stated here as fairly as the space allows, and it is not universally accepted in its strongest form; some historians and political scientists would grant the reality and depth of Ukrainian national identity while still analyzing the war's causes chiefly through the lens of great-power security competition (a realist account this unit introduces below and Unit 2 develops formally), treating identity as a contributing factor rather than the central stake. Both kinds of claims can be, and in this course will be, evaluated against the same evidentiary record rather than asserted by authority.

It is worth dwelling further on the specific mechanics of Snyder's deep-history argument, because its force depends on details that a summary tends to compress away. Snyder's treatment of Kyivan Rus' is not simply an assertion that Ukraine has an old history; it is an argument about competing inheritance claims over the same medieval polity. Kyivan Rus', the loose federation of East Slavic principalities centered on Kyiv from roughly the ninth through the thirteenth centuries, dissolved under Mongol invasion long before either a Russian or a Ukrainian state existed in anything like their modern form — which means later claims by both Moscow and Kyiv to be Kyivan Rus's rightful heir are themselves retrospective national narratives constructed centuries afterward, not settled facts inherited automatically by either side. Snyder's point in tracing this out in detail is that treating Kyivan Rus' as simply "early Russian history," the framing Putin's essay relies on, forecloses by assumption a question that the actual medieval record leaves genuinely open — which successor political tradition, if any, has the stronger claim to that inheritance is a question historians dispute, not a settled premise available for a modern territorial argument to build on without defending it.

The Cossack political tradition Snyder emphasizes supplies a second, independent thread, chronologically and institutionally distinct from the Kyivan Rus' inheritance question. The Zaporozhian Cossacks, a self-governing military and political community centered on the Dnipro River from roughly the sixteenth through the eighteenth centuries, developed their own elective leadership structures, treaties, and a documented sense of distinct political identity — most famously formalized in the 1654 Treaty of Pereyaslav with Muscovy, an agreement whose precise terms (a protectorate arrangement, a full incorporation, or something contested between the two) has been argued over by historians on both the Russian and Ukrainian sides ever since, which is itself Snyder's point: even a foundational document establishing a Russian-Ukrainian political relationship has never commanded a single agreed reading, undermining any claim that the two peoples' unity has been an uncontested historical fact rather than a live, centuries-old argument.

The Holodomor, the 1932-33 famine that killed millions in Soviet Ukraine under policies historians continue to debate as to how deliberately they targeted Ukrainians specifically as opposed to peasants generally across the wider Soviet famine, occupies a different kind of place in Snyder's argument: not a claim about pre-modern political inheritance, but a claim about how a modern, twentieth-century experience of catastrophic loss, suffered collectively and remembered collectively, can itself forge a national identity independent of any older lineage — a mechanism closer to what constructivist IR theory (introduced below and developed formally in Unit 2) describes as identity formation through shared historical trauma and commemoration, rather than through any claim about deep medieval descent. Snyder's full argument, then, braids three distinct kinds of evidence — contested medieval inheritance, an early modern tradition of self-governance, and a twentieth-century trauma consolidated into national memory — into a case that Ukrainian nationhood rests on multiple independent historical foundations rather than any single, easily dismissed one.

Three Lenses, One Question

Everything examined so far in this unit — the failed predictions, the ambiguous deterrence signal, Zelensky's defiance, Snyder's identity argument — can be organized as competing candidate answers to a single overarching question that will structure the rest of this course: why did this war happen, and why did it unfold the way it did?

This course will not answer that question with a single theory declared the winner. Instead, it introduces, and will repeatedly apply across the next nine units, three major theoretical traditions in the study of international relations, each of which offers a genuinely different account of the war's causes and its trajectory. Realism, developed formally in Unit 2 through the work of Kenneth Waltz, explains the war chiefly through the logic of power and security in an anarchic international system: Russia, as a great power bordering an expanding Western alliance system, acted to prevent what it perceived as an intolerable shift in the regional balance of power, regardless of Ukraine's own preferences or identity. Liberalism, developed through Joseph Nye's work on institutions and interdependence, explains the West's response chiefly through the strength (and limits) of international institutions, alliance coordination, and economic interdependence — NATO's unusually unified sanctions and weapons-supply response is, on this account, evidence for how much institutions matter, even as the war itself is evidence for their limits in restraining a determined aggressor. Constructivism, developed through Alexander Wendt's work on identity and shared meaning, explains the war chiefly through the clash of national identities and historical narratives examined above in Snyder's argument — the war, on this account, is in significant part a war over what Ukraine and Russia each mean, to themselves and to each other, as nations.

None of these three lenses is being declared correct here, and none should be treated as mutually exclusive; a realist account of Russian security anxiety about NATO's eastward reach and a constructivist account of contested national identity are not necessarily in competition — a single war can have multiple, simultaneously true causal stories operating at different levels of analysis, a point Unit 2 will formalize through Waltz's three images of international politics. What this course asks of you is not to pick a favorite lens and defend it, but to hold all three in view at once, and to get specific and disciplined about what evidence would make you weight one explanation more heavily than another for a given event. That is the actual skill this course is teaching, and Unit 2 begins building the theoretical vocabulary to do it properly.

❓

Concept Check

Why does this course introduce three competing theoretical lenses rather than teaching one 'correct' theory of international relations?

▸

Because a single war typically has multiple, simultaneously true causal stories operating at different levels — a great power's security anxiety about alliance expansion (realism), the strength or weakness of international institutions and interdependence in shaping the response (liberalism), and a contested clash of national identities and historical narratives (constructivism) can all be operating in the same conflict at once, rather than being mutually exclusive rival explanations where only one can be right. The course's goal is not to select a winning theory but to teach students to specify what evidence would lead them to weight one explanatory lens more heavily than another for a specific event — framework pluralism as a disciplined analytical practice, not framework shopping for whichever theory is convenient.

What Comes Next

The chapters that follow will not proceed chronologically forward from some tidy starting point in the deep past. This course works the way historical understanding actually tends to work when it starts from a live, urgent case: backward, from the war as it stands today into the theoretical and historical material that makes sense of it. Unit 2 builds the three lenses introduced above into full theoretical frameworks, tested against Thucydides's Melian Dialogue — a conversation roughly 2,400 years old about exactly the kind of power asymmetry Ukraine faced in February 2022. Unit 3 tests the most famous prediction about the post-Cold War world, Francis Fukuyama's "end of history" thesis, against the war's inconvenient evidence. And the laboratory stays open for the rest of the course: every unit that follows returns to February 2022 and its aftermath, from a different analytical angle, because the war did not end when this unit went to press, and the point of studying international relations is to build tools durable enough to still be useful when the next crisis, whatever and wherever it is, begins.

Key Debates: Engaging the Scholarship

Was the intelligence "failure" actually a failure? One view, common in retrospective commentary, treats the widespread expectation of a rapid Russian victory as an intelligence failure comparable to other famous forecasting misses. A more precise view, and the one this unit has tried to model, separates the question into two genuinely different predictions with genuinely different track records: the invasion-will-happen prediction (largely vindicated, and vindicated unusually early by U.S. intelligence specifically) and the invasion-will-succeed-quickly prediction (largely wrong, across nearly every Western government and independent analyst). Collapsing these into a single "the experts were wrong" narrative obscures which analytical tools worked and which did not — precisely the distinction Kotkin's evidence-first register insists on holding open rather than flattening into a single verdict.

How much should identity explain? Snyder's identity-centered argument is influential and well-evidenced, but it is not the only defensible reading of the war's causes, and a fair treatment of the debate has to state the realist alternative on its own terms rather than as a strawman. Scholars in the realist tradition, most prominently John Mearsheimer (whose position this course states fairly and scores against evidence in Unit 2), have argued that NATO's post-Cold War expansion created a security dilemma serious enough to explain Russian behavior largely without reference to contested claims about Ukrainian national identity at all — on this account, a large power's response to what it perceives as an encircling alliance system is a recurring, structurally predictable pattern, observable across many historical cases, that does not require any particular theory of Ukrainian nationhood to explain. The identity-centered and security-centered accounts are not fully reconcilable at the level of "which one is the primary cause," and this course does not resolve that disagreement by fiat; it asks you to weigh both against the documentary record as later units accumulate more evidence.

Does the deterrence-failure framing let anyone off the hook, or does it fairly distribute responsibility? Some commentators resist the phrase "deterrence failed" because it can imply that Western policy choices, rather than the Russian decision to invade, are the primary object of analysis — a framing critics worry shifts moral responsibility away from the actor who chose war. Others argue that analyzing why deterrence failed is a separate project from assigning moral responsibility for the invasion, and that a course teaching international relations has an obligation to examine the strategic logic of deterrence and signaling honestly, including its failures, without that analysis functioning as an excuse for the invasion itself. This course treats the two projects — strategic analysis of what happened, and moral judgment of who is responsible for it — as related but distinct, consistent with the methodological/moral distinction this kind of analysis requires.

Assessment Suggestions

• Prediction autopsy: Students select one specific pre-invasion or early-invasion prediction (from a named government, agency, or analyst) and write a short analysis identifying exactly what was predicted, what actually happened, and which category of variable (capability, will, logistics, political signaling) best explains the gap, if any.
• Deterrence vs. assurance memo: In the role of a national security advisor in December 2021, students draft a one-page memo recommending a specific signaling strategy toward Russia, explicitly weighing the deterrence benefits against the assurance costs (or vice versa) of the options considered, using the vocabulary this unit introduces.
• Snyder vs. Mearsheimer structured debate: Students are assigned (not self-selected) to argue either the identity-centered account of the war's causes (Snyder) or the security-dilemma account (Mearsheimer, previewed here and developed fully in Unit 2), each side required to state the opposing position accurately before rebutting it.
• Three-lenses preview exercise: Given a short list of documented facts about the war's opening week, students sort each fact according to which of the three lenses (realism, liberalism, constructivism) it most directly supports, and identify at least one fact that could plausibly support more than one lens simultaneously.

Vocabulary

• Sovereignty: A state's supreme authority over its own territory and population, free from external control — the norm the invasion is most directly charged with violating.
• Deterrence: Convincing an adversary that the costs of an action will outweigh its benefits, so the adversary chooses not to act — distinct from assurance, which is convincing a partner that you will stand behind them.
• Realism: The IR tradition explaining state behavior chiefly through power, security, and the absence of a higher authority above states (anarchy); developed formally in Unit 2.
• Liberalism (IR): The IR tradition emphasizing institutions, interdependence, and cooperation as forces that shape and constrain state behavior beyond raw power; developed formally in Unit 2.
• Constructivism: The IR tradition emphasizing that state interests and behavior are shaped by shared ideas, identities, and historical narratives, not fixed by material power alone; developed formally in Unit 2.
• Order-of-battle analysis: Military assessment based on counting and comparing forces, equipment, and formations — useful but incomplete, since it underweights logistics, command quality, and will.
• Capability-counting: See order-of-battle analysis; used in this unit to describe the analytical habit of predicting outcomes primarily from material force ratios.

Recommended Resources

Lectures

• Timothy Snyder, The Making of Modern Ukraine, Yale University — a free 23-lecture course available on YouTube, covering the deep history this unit only previews; start with the lectures on the Holodomor and the 2014 Euromaidan for material most directly relevant to this unit's argument.

Books

• Timothy Snyder, The Making of Modern Ukraine (2022) — the companion volume to the Yale lecture series.
• Serhii Plokhy, The Gates of Europe: A History of Ukraine (2015, revised 2021) — a complementary, slightly more traditional narrative history, useful for cross-checking Snyder's interpretive framing against another leading historian's account.

Video / Ongoing Analysis

• Perun (YouTube) — defense-economics-focused analysis of the war's material and logistical dimensions; useful for testing the capability-counting versus will/logistics distinction this unit introduces against ongoing developments.