History is usually told as inevitability. Rome fell. Napoleon lost. The atomic bomb ended the Pacific War. The cold confidence of hindsight makes every outcome seem scripted in advance, as though the actors were simply playing out a story whose ending was already fixed. Counterfactual history challenges that confidence directly, and it does so not by spinning fantasies but by applying rigorous analytical pressure to the decision-points where outcomes genuinely could have differed. The field has a reputation problem. Popular treatments of historical what-ifs tend toward spectacle: cover art of Nazi swastikas over Washington, or Roman legions marching across North America. Serious historians have pushed back against this tradition with some force. Richard Evans’s Altered Pasts: Counterfactuals in History (2013) delivers a sustained skeptical argument that counterfactual exercises encourage overestimation of individual agency, produce untestable claims, and slide quickly into ideology-serving fantasy. Evans is right that the worst examples deserve...
Read the rest with bitcoin
One-time on-chain payment unlocks this article. No accounts, no email, no tracking.
History is usually told as inevitability. Rome fell. Napoleon lost. The atomic bomb ended the Pacific War. The cold confidence of hindsight makes every outcome seem scripted in advance, as though the actors were simply playing out a story whose ending was already fixed. Counterfactual history challenges that confidence directly, and it does so not by spinning fantasies but by applying rigorous analytical pressure to the decision-points where outcomes genuinely could have differed.
The field has a reputation problem. Popular treatments of historical what-ifs tend toward spectacle: cover art of Nazi swastikas over Washington, or Roman legions marching across North America. Serious historians have pushed back against this tradition with some force. Richard Evans’s Altered Pasts: Counterfactuals in History (2013) delivers a sustained skeptical argument that counterfactual exercises encourage overestimation of individual agency, produce untestable claims, and slide quickly into ideology-serving fantasy. Evans is right that the worst examples deserve the criticism he gives them. The problem is that his critique, applied broadly, would eliminate a legitimate methodological tool alongside the abuses of it.
Niall Ferguson’s edited collection Virtual History: Alternatives and Counterfactuals (1997) makes the constructive case. Ferguson argues that counterfactuals are not an indulgence but a necessity: historians who refuse to engage them are implicitly making causal claims about what was inevitable, and those claims go unexamined. Philip Tetlock, Richard Ned Lebow, and Geoffrey Parker’s Unmaking the West: “What-If?” Scenarios That Rewrite World History (2006) develops a methodological framework for distinguishing rigorous counterfactuals from idle speculation. The standard they articulate is demanding: a legitimate counterfactual must require the minimum number of initial changes to produce the alternate outcome, must be grounded in specific documented evidence about the decision-point in question, and must acknowledge the increasing uncertainty that accumulates as the chain of consequences lengthens.
Applying that standard to history’s biggest what-ifs is the task of this article. Ten scenarios are examined in chronological order, from the Greek Wars of the fifth century BCE through the Cuban Missile Crisis of 1962. Each scenario is assessed for its plausibility at the initial change-point, its methodological classification as structural or genuinely contingent, and the range of downstream consequences that follow with some analytical confidence from the initial change. The goal is not to produce a definitive alternative history but to understand what the scenarios reveal about how the actual history came to be.
The Method Before the Speculation
Before examining individual scenarios, the Ferguson-Tetlock methodological framework requires elaboration. The central distinction is between outcomes that were structurally overdetermined and outcomes that were genuinely contingent. A structurally overdetermined outcome is one that would have resulted from multiple independent causal pathways even if any single pathway had been blocked. A genuinely contingent outcome is one that depended critically on specific choices or events that could plausibly have gone another way.
The classification matters enormously for what a counterfactual can teach. If the fall of the Western Roman Empire was structurally overdetermined, then counterfactuals asking what would have happened had Stilicho not been executed in 408 CE or had the Rhine frozen less dramatically in 406 CE are interesting exercises but do not reveal much about causation. The Empire would have fallen through other pathways. If the fall was primarily contingent on specific events, then altering those events produces a genuine historical experiment. The truth, in most cases, is that specific large historical events contain both structural and contingent elements, and the analytical task is to identify which elements belong to which category.
The methodological standard also requires acknowledging ripple-effect limitations. Every change in the historical chain produces downstream changes that quickly become impossible to predict with any precision. A scenario that stops Archduke Franz Ferdinand’s assassination in 1914 can say something reliable about the immediate diplomatic crisis that did not occur. It cannot reliably predict whether World War One would have happened in 1917 instead, or whether Germany would have eventually produced a fascist movement even without the particular humiliation of the 1918 defeat. Responsible counterfactual analysis is explicit about where confidence ends and speculation begins.
Geoffrey Hawthorn’s Plausible Worlds (1991) provides a philosophical grounding for this kind of analysis. Hawthorn argues that plausible worlds - possible outcomes genuinely achievable from a documented historical moment - are analytically distinguishable from merely possible worlds, which require stacking implausibility on implausibility until the exercise bears no relationship to the actual historical record. The ten what-ifs that follow are analyzed as plausible worlds in Hawthorn’s sense, not as fantasy projections.
The distinction between minimum-change and maximum-change scenarios is among the most practically useful tools the Tetlock framework provides. A minimum-change scenario alters the smallest number of facts necessary to produce the alternate outcome. “What if Ogedei Khan had died in 1245 rather than 1241?” is a minimum-change scenario: it requires changing one date, leaving all surrounding circumstances intact. “What if the Mongol Empire had developed a democratic political tradition that made westward expansion impossible?” requires changing dozens of underlying conditions simultaneously and produces a scenario with no grounding in the actual Mongol political record. The minimum-change constraint keeps counterfactual scenarios analytically tractable because it limits the number of variables being simultaneously altered.
This constraint also reveals something important about historical explanation. When we say the Spanish Armada “failed because of storms,” we are implicitly identifying the storms as the proximate cause. Asking “what if the storms had not occurred?” tests that causal claim. If the analysis shows that the Armada’s fundamental planning problems would have produced failure even without the storms - as the evidence strongly suggests - then the causal claim needs revision. The storms did not cause the failure; they merely produced it more rapidly and more completely than alternative pathways would have. Counterfactual analysis of this kind is not speculation about what might have been; it is a disciplined test of what the actual historical explanation requires us to believe.
The scholarly debate between Ferguson and Evans partly runs on different intuitions about this test. Ferguson believes that making causal claims explicit and testable is always analytically productive, even when the tests reveal more uncertainty than conventional narratives suggest. Evans believes that the uncertainty revealed by counterfactual testing tends to be exploited ideologically rather than accepted honestly. The historical record supports both concerns: counterfactual history at its best does exactly what Ferguson describes; counterfactual history at its worst does exactly what Evans fears. The solution is not to eliminate the method but to apply it with the rigor that Tetlock’s framework demands.
These ten scenarios examined in this article are selected for a combination of methodological interest and historical significance. They span from the ancient world through the mid-twentieth century, cover military, political, and diplomatic decision-points, and range from scenarios where structural forces were dominant to scenarios where genuinely contingent individual choices appear to have been decisive. Together they constitute a kind of sample of the methodological diversity available within serious counterfactual historical analysis, demonstrating both the power of the approach and its inherent limitations.
What If Persia Had Won the Greek Wars?
The Persian Wars of 490-479 BCE represent one of history’s clearest cases of genuine contingency at the decisive engagement level combined with structural complexity at the civilizational level. The scenario requires the minimum change: a different outcome at Marathon in 490 BCE, or different outcomes at Salamis and Plataea in 480-479 BCE. The historical record shows that all three engagements were close-run. The outcome at Marathon depended on an Athenian decision to attack rather than wait, driven by Miltiades and opposed by other generals, that exploited a specific moment of Persian tactical overconfidence. The outcome at Salamis depended on Themistocles’s successful deception of the Persian command and specific navigational conditions in the narrows. The outcome at Plataea depended on coordination among Greek city-states that very nearly collapsed before the final engagement.
What Persia would have done with a conquered Greek mainland is knowable with some precision from the Persian record in Ionia and Egypt. The Achaemenid model was not elimination but absorption: local elites continued to function under Persian administrative authority, local religious practices were generally tolerated, and cultural production continued. Greek philosophy, mathematics, drama, and political theory did not emerge from conditions that could not have existed under Persian rule. What did depend on Greek independence was the specific political form of the polis and the particular practice of democratic self-governance that Athens had developed by 490 BCE.
Athenian democracy was not inevitable. It had developed through specific political reforms - Cleisthenes’s reorganization of the tribal structure in 508-507 BCE - that were themselves contingent responses to a particular political crisis. Whether democratic governance would have developed in Greek poleis under Persian suzerainty is genuinely unclear. What is clear is that the direct transmission of Athenian democratic practice to later republican and democratic traditions required Athens to survive as a functioning political entity, which a Persian victory at Marathon would have prevented.
The scenario is further complicated by what Persian victory would have meant for Sparta. The Spartans fought at Thermopylae and Plataea but had serious reservations about committing fully to a war they feared would extend beyond the Peloponnese. A different outcome at Marathon might have produced Spartan accommodation with Persia and the partitioning of the Greek world. The downstream consequences for Western European political development are genuine and substantial, but they stop well short of the scenario’s frequent framing as the near-death of all rationalism and democracy. Greek cultural influence had already spread through Magna Graecia in southern Italy and would have continued to exert influence even under Persian political control.
The analysis of why Athenian military power was so decisive in this period connects directly to the civilizational comparison in our examination of how Athens and Sparta built fundamentally different power models in the same Greek world. The Persian Wars exposed those differences as sharply as any other event in the classical period.
The Persian empire’s administrative geography also matters for assessing this scenario. By 490 BCE, the Achaemenid empire stretched from the Indus to the Aegean, incorporating dozens of distinct linguistic and cultural communities under a tiered administrative system of satrapies. Darius I had demonstrated administrative competence of a high order in managing this diversity without requiring cultural homogenization. The Greek poleis would have entered that system as one satrapy among many - important for their commercial networks and their silver mining in Attica and Laurion, but not uniquely significant in the broader imperial calculation.
What would have been lost is specifically the competitive political experimentation that the independent polis system enabled. The rivalry between Athens and Sparta - economically and culturally destructive as it ultimately proved in the Peloponnesian War - was also intellectually generative. The contrast between Athenian democracy and Spartan oligarchy as alternative political models drove political theory from Thucydides through Plato and Aristotle. A Persian administrative framework would have reduced that competitive experimentation without necessarily eliminating the intellectual traditions that the competition stimulated. Persian court culture supported significant intellectual activity, and the major Greek thinkers of the fifth and fourth centuries BCE - Socrates, Plato, Aristotle - operated in a context of patronage and philosophical debate that had Persian analogues.
The philosophical tradition that ultimately matters most for European intellectual history - Platonic philosophy and its Aristotelian development - is the element of the Greek legacy most difficult to predict under Persian administration. Socrates’s specific mode of philosophical practice, his public questioning of political authority and religious convention, was eventually prosecuted and executed by Athenian democracy itself. Whether a Persian administrative context would have been more or less tolerant of that specific form of philosophical provocation is genuinely unclear. Persian royal courts patronized poets and astronomers; their treatment of philosophers with political implications for governance is less documented.
What If Alexander Had Lived Longer?
Alexander III of Macedon died in Babylon in June 323 BCE at the age of thirty-two, almost certainly of typhoid fever complicated by heavy alcohol consumption, though the specific medical cause remains debated. His death produced the Wars of the Diadochi - the succession conflicts among his generals that fragmented his empire into the Hellenistic kingdoms within forty years of his death. The question of what Alexander would have done next is not idle: he had specific documented plans.
The ancient sources - Arrian, Diodorus Siculus, and Plutarch drawing on earlier accounts - record that Alexander had ordered preparations for a westward campaign against Carthage and the western Mediterranean, a southward campaign into Arabia, and further consolidation of his eastern territories. These were not vague ambitions but operational planning documents. Diodorus preserves what appear to be Alexander’s actual hypomnemata (memoranda) for these campaigns, though scholars debate their authenticity.
The most significant consequence of Alexander’s survival would have been the political unity of his empire for at least another decade or two. The Wars of the Diadochi consumed an enormous amount of the military talent, economic surplus, and political energy that the Hellenistic world might otherwise have directed elsewhere. A unified Alexandrian empire in 300 BCE would have possessed resources sufficient to challenge both the rising Roman power in the west and the Mauryan Empire in the east.
Rome’s conquest of the Mediterranean world was not structurally inevitable. The Romans were consistently skilled at adapting to military challenges, but they faced several moments of genuine crisis - Pyrrhus of Epirus very nearly inflicted decisive defeats in the 280s BCE, and the Macedonian phalanx was a formidable tactical problem even for Roman legions. An Alexandrian empire with decades of organized military and administrative development behind it would have presented a qualitatively different challenge than the Hellenistic kingdoms Rome eventually absorbed.
The scenario’s confidence limits appear quite early, however. Alexander’s own empire showed signs of administrative stress before his death: the satrapal revolts, the Macedonian officers’ resistance to Persian court customs, the murder of several key subordinates on Alexander’s orders. Whether Alexander would have managed the political contradictions of ruling a multiethnic empire across three continents is genuinely unclear. The scenario reveals something important: the fragmentation of the Hellenistic world was not entirely the product of Alexander’s death. It also reflected the structural tensions of an empire assembled through conquest and held together primarily by a single extraordinary personality.
The question of succession is critical to understanding why Alexander’s early death had such decisive consequences. Unlike the Persian Achaemenid emperors, who maintained elaborate court rituals and administrative networks designed to survive the death of individual rulers, Alexander had not created comparable institutional structures for transmitting authority. His empire was in significant part a personal achievement sustained by his presence. The Diadochi - Antigonus, Ptolemy, Seleucus, Cassander, Lysimachus, and others - each commanded regions where they had built local legitimacy through years of military administration. Without Alexander’s authority to hold these competing power centers together, fragmentation was not merely possible but probable.
A longer-lived Alexander - one who reached fifty or sixty - would have had time to develop the succession institutions that his early death prevented. The Persian model was available: an elaborate royal court, a system of satrapal administration with central oversight, and a tradition of hereditary succession backed by court ritual and dynastic ideology. Alexander had been in the process of incorporating Persian administrative practices at the time of his death, dressing in Persian royal costume and requiring Persian-style proskynesis from his court. Whether these attempts at cultural integration would have succeeded politically - whether Macedonian officers and Greek administrators would have accepted them - is among the most interesting open questions in ancient history.
The intellectual consequences of a unified Alexandrian empire would have been significant in their own right. The Hellenistic kingdoms that actually emerged from the Diadochi wars became major centers of Greek intellectual life: Ptolemaic Alexandria with its Library and Museum, the Seleucid patronage of Babylonian and Greek scientific exchange, Attalid Pergamon with its own library and cultural program. These were competitive projects driven partly by the kingdoms’ efforts to claim the cultural legitimacy that Alexander had embodied. A unified empire would have concentrated patronage differently, and the specific geography of Hellenistic intellectual culture - the Library at Alexandria, the Stoic school at Athens, the Epicurean garden - might have developed in significantly different form.
What If the Western Roman Empire Had Survived?
The conventional date for the fall of the Western Roman Empire is 476 CE, when the Germanic chieftain Odoacer deposed the last emperor Romulus Augustulus and sent the imperial regalia to Constantinople. The problem with this framing is that contemporaries did not experience 476 as a decisive rupture. The Eastern Roman Empire continued for nearly another thousand years. Roman law, Latin language, and Catholic Christianity all persisted across the former Western territories. The question “what if Rome had never fallen” partly rests on a historical category error.
A more precisely formulated version asks: what if the political authority of the Western Roman court had survived through the fifth century and into the sixth, preventing the political fragmentation of the western Mediterranean into competing Germanic kingdoms? This is a plausible counterfactual because several specific fifth-century crises had contingent elements. The crossing of the Rhine by Vandals, Suebi, and Alans in 406 CE exploited an unusually hard freeze that does not recur frequently in the historical record. The assassination of the effective western general Aetius in 454 CE by the emperor Valentinian III - who thereby destroyed his own best military resource, dying himself the following year - was a specific political miscalculation that does not appear structurally inevitable.
The problem with the survival scenario is the structural pressure that was generating those contingent crises. The third and fourth centuries had seen significant demographic pressure on Roman frontiers from populations displaced by events further east. The Huns’ westward movement beginning around 370 CE created refugee pressure from Gothic populations that the Roman state lacked the administrative and military resources to absorb without disruption. Whether different individual decisions at specific moments would have resolved that structural pressure or merely postponed the crisis is genuinely unclear.
What is clear is that the Eastern Empire’s survival demonstrates that the civilizational contents of Roman culture were not dependent on Western political continuity. Greek philosophy, Roman law, and Christian theology all survived and developed through the Byzantine tradition. The more productive version of the question may be: what specific losses resulted from the Western fragmentation? The answer centers on administrative continuity, literacy rates, trade networks, and urban density - all of which declined sharply in the former Western provinces in the fifth and sixth centuries. A surviving Western imperial administration would not have guaranteed these outcomes, but it would have maintained the institutional infrastructure that supported them.
Our analysis of why the Roman Empire fell examines the structural forces that made Western fragmentation increasingly likely and identifies which of those forces were genuinely structural versus which were amplified by specific contingent decisions. That analysis provides essential background for assessing what a survival scenario would have required.
What If the Mongols Had Continued Into Western Europe?
The Mongol campaign into Central Europe in 1241 is one of history’s most striking examples of external contingency redirecting a military campaign. Batu Khan’s forces had crushed the Polish forces at the Battle of Legnica and the Hungarian forces at the Battle of Mohi in April 1241, and were wintering in Hungary preparing for a westward advance. Then Ogedei Khan, Genghis’s son and successor, died in December 1241. The Mongol tradition of kurultai - the assembly of all senior Mongol leaders to select a new Great Khan - required Batu’s presence in Mongolia. The western campaign was suspended and never renewed at the same scale.
Genuine contingency is present here: Ogedei was forty-five years old at his death, and his excessive drinking was documented in the sources, but his specific death in December 1241 rather than several years later was not structurally predetermined. The counterfactual requires only that Ogedei die in 1245 rather than 1241, which is entirely within the range of plausible variation.
The analytical challenge is assessing what a continued western campaign would have encountered. Western European military forces had performed poorly against the Mongols in 1241 - the fragmented command structure, the emphasis on heavy cavalry charges, and the reliance on castle sieges all played poorly against Mongol combined-arms tactics. The Mongol track record against similar military formations in Persia, the Caucasus, and Russia was consistent. The most honest assessment is that the forces of the Holy Roman Empire and France would have suffered similar defeats to those sustained by Poland and Hungary.
What would have followed military conquest is less clear. The Mongol empire in Central Asia and Persia showed significant variation in administrative approach. Some conquered territories were managed extractively; others, particularly regions with established urban and commercial networks, were administered more carefully because they generated tax revenue. Western Europe in 1241, with its fragmented political structure and predominantly agricultural economy, may have attracted less intensive administration than the Mongols gave to Persia or China.
The scenario’s most interesting dimension is what it reveals about the subsequent century. The Black Death of 1347-1351, which devastated European population by roughly one-third, traveled the trade routes of the Mongol empire. A Mongol-administered western Europe would have been connected to those trade routes far more directly than the actual western Europe was. The timing and severity of disease transmission are genuinely uncertain, but the probability of earlier and more severe epidemic disease in a Mongol-connected western Europe is analytically supportable.
The Catholic Church’s institutional role in medieval western Europe also deserves consideration in this scenario. The Church in 1241 represented the primary supranational administrative structure in western Europe, commanding the loyalty of literate administrative elites across multiple kingdoms. The Mongols’ approach to religious institutions varied significantly. In Persia, where the Mongols initially supported Nestorian Christian communities against Sunni Muslim populations, religious institutions were sometimes deliberately cultivated as administrative resources. In areas with more entrenched religious hierarchies, the Mongols were more likely to extract tribute than to engage in systematic destruction. The Mongol court in this period included Buddhist, Nestorian Christian, Shamanist, and Muslim advisers in varying combinations, and the Great Khan’s religious policy was primarily pragmatic rather than ideological.
Whether a western European Mongol administration would have suppressed or exploited the Catholic Church’s administrative infrastructure is genuinely uncertain, but the analogy with other conquered regions suggests exploitation was more likely than systematic suppression. The doctrinal and institutional development of medieval Catholicism in the thirteenth century - Aquinas’s synthesis, the mendicant orders, the development of university education - might have continued in a significantly modified form under Mongol oversight. Alternatively, the disruption of ecclesiastical networks and the loss of revenues that supported scholastic intellectual life could have interrupted those developments with consequences for the fourteenth and fifteenth century intellectual transformations that eventually produced the Renaissance and the Reformation.
What the counterfactual most clearly reveals is the degree to which medieval western European development was shaped by its relative isolation from the Eurasian trade and disease networks that connected Central Asia, Persia, and China. That isolation was not purely geographic - western Europe was reachable overland and by sea from the eastern Mediterranean - but partly political, reflecting the fragmented political structure that limited large-scale integration into long-distance commercial networks. A Mongol-administered western Europe would have been far more thoroughly integrated into those networks, with consequences for both commercial development and epidemic disease that are impossible to model with precision but are likely to have been transformative in both directions.
What If the Spanish Armada Had Succeeded?
The Spanish Armada of 1588 is among the most thoroughly analyzed near-misses in military history. Philip II of Spain assembled 130 ships and approximately 27,000 men for an amphibious operation intended to land a Spanish army from the Netherlands on the English coast, depose Elizabeth I, and restore Catholic governance. The plan failed through a combination of English defensive action, fireships at Gravelines that disrupted the Armada’s formation, and then storms in the North Sea that wrecked over half the Spanish fleet during the withdrawal around the north of the British Isles.
The contingency analysis, as this site’s detailed examination of the Armada campaign and Philip II’s planning errors establishes, shows that the operation faced severe structural problems that the storms merely finalized. The fundamental flaw was that the Armada could not provide offshore fire support for the army’s landing because the rendezvous with Parma’s Netherlands forces required specific coastal geography that was not available. The Spanish ships were too deep-drafted to anchor in the shallow Flemish waters where Parma’s barges waited. A plan that required a condition it could not provide was not undone by storms; it was undone by Philip’s insistence on proceeding despite documented planning problems.
A version of the scenario in which the logistical problems were resolved - a different rendezvous plan, a different landing strategy - is methodologically weaker than a simple-change scenario because it requires multiple modifications to the original plan. The more interesting question is what Spanish success would have meant for England’s subsequent development. Elizabeth’s England of 1588 was not yet the dominant maritime power it would become. The East India Company was not founded until 1600. English colonization of North America had barely begun. A Spanish-controlled England in 1590 would have changed the configuration of Atlantic colonization, probably to the benefit of Iberian colonial expansion and the detriment of the later Dutch and English trading empires.
The longer-term scenario is extremely uncertain. English Protestantism was not dependent on the English monarchy - it had deep roots in the population and would likely have produced significant resistance to Spanish Catholic restoration. The scenario of a successfully conquered England transforming seamlessly into a Spanish Catholic province underestimates the ferocity of English Protestant identity by 1588. A more plausible version involves a prolonged military occupation contested by sustained English resistance, consuming the Spanish resources that went instead into the Eighty Years’ War in the Netherlands.
What If the American Revolution Had Failed?
The American Revolution of 1776-1783 presents a genuinely interesting counterfactual because the structural case for American independence was strong while the military case was, for much of the war, precarious. The Continental Army suffered severe manpower and supply problems throughout the conflict. Washington’s crossing of the Delaware and the Trenton engagement on December 26, 1776 - which rescued a campaign on the verge of complete collapse - was both a genuine turning point and a genuinely contingent event that exploited a specific moment of Hessian overconfidence.
British victory would have required something that Britain consistently failed to achieve: a strategy for pacifying a colonial population hostile to occupation spread across enormous territory. The American military position was genuinely weak; the American political position was structurally strong because Britain could not afford the administrative costs of permanent military occupation. A British military victory in 1778 or 1779 - perhaps preventing the French alliance of 1778 that tipped the strategic balance definitively - would likely have produced a negotiated settlement rather than permanent reintegration.
The scenario’s most consequential dimension concerns the French Revolution of 1789. The French intervention in the American war contributed substantially to the financial crisis that destabilized the French monarchy in the late 1780s. Without the American war’s costs, French finances would have been in significantly better condition in 1787-1789. Whether the French Revolution would have occurred without the fiscal crisis is uncertain, but the specific sequence of events that produced 1789 - the calling of the Estates-General, the Third Estate’s assertion of sovereignty, the fall of the Bastille - was shaped by the financial emergency that the American intervention helped create.
The democratic-republican ideology of the American Revolution also had independent influence on European political movements. That influence would not have disappeared with a British military victory; the political ideas of Locke, Montesquieu, and Rousseau that shaped American political thought were European in origin and would have continued to circulate regardless of American military outcomes.
What the counterfactual analysis clarifies is the difference between ideological influence and institutional precedent. The American Constitution of 1787 - its system of separated powers, its enumerated rights, its federal structure balancing central and state authority - was not simply a philosophical document. It was a practical institutional design that influenced constitutional drafters across Latin America, France, and eventually much of the world over the following century. That specific institutional model required American independence and the specific political negotiations of 1787 to exist. European liberals drawing on Locke and Montesquieu without an American institutional model would have been drawing on philosophy without a functioning demonstration. Whether the constitutional arrangements that emerged across Europe and Latin America in the nineteenth century would have been different, or more unstable, without that American precedent is a genuine analytical question.
The scenario also opens a question about the subsequent development of the transatlantic slave trade. American independence produced a United States that abolished the international slave trade in 1808 and eventually abolished slavery itself in 1865. A British-controlled North America would have followed the British timeline: Britain abolished its slave trade in 1807 and abolished slavery across the Empire in 1833. In the short term, British abolition timelines were faster than American ones. Whether a British North America would have abolished slavery in 1833 across its North American territories - which were significantly more agricultural and slavery-dependent than the Caribbean colonies that defined the British abolitionist political economy - is genuinely uncertain. The American Civil War as a specific event would not have occurred; whether a comparable political conflict over slavery in a British colonial context would have emerged is an open question.
Geographic and commercial development of North America also takes a substantially different shape under continued British administration. The Louisiana Purchase of 1803, which Jefferson negotiated from Napoleon, would not have occurred - Napoleon was negotiating with a sovereign United States, not with a British colonial administration. The westward expansion that defined nineteenth-century American development, with its specific combination of land grants, settler migration, and indigenous dispossession, would have proceeded under a different legal and political framework with significantly different outcomes for indigenous populations and for the eventual configuration of North American political geography.
The intellectual and cultural tradition that American independence produced - the specific combination of Protestant religious culture, frontier experience, commercial republicanism, and Enlightenment political philosophy that Alexis de Tocqueville analyzed in the 1830s - would not have developed in the same form under British colonial administration. Whether that specific intellectual formation was necessary for the development of liberal democratic politics more broadly, or whether comparable formations would have emerged through other pathways, is among the most interesting questions the scenario opens. The answer has implications well beyond the specific case of American history.
What If Napoleon Had Won at Waterloo?
Waterloo on June 18, 1815 represents one of history’s closest near-misses. The battle was decided by the timing of Prussian reinforcement - if Blücher’s forces had arrived two hours later, Wellington’s position at Mont Saint-Jean would likely have been overrun. The timing depended on the Prussian withdrawal route after the Battle of Ligny two days earlier. An alternate withdrawal route would have delayed Prussian arrival at Waterloo by the margin that changed the outcome.
Niall Ferguson has given the Waterloo counterfactual sustained analysis in Virtual History, arguing that Napoleon’s position in 1815 was structurally weakened regardless of the specific battle outcome. The Seventh Coalition had committed enormous resources to Napoleon’s permanent removal from French power. A Waterloo victory would have been followed by new Coalition campaigns with fresh Austrian, Russian, and Prussian forces that Napoleon’s depleted France could not have matched for long. The scenario of permanent Napoleonic victory across Europe requires not just a Waterloo win but a subsequent diplomatic settlement that the Coalition powers had explicitly committed to refuse.
The more limited scenario - Napoleon wins at Waterloo and secures better peace terms than the Treaty of Paris of 1815 - is more analytically tractable. Better peace terms for France would likely have meant retention of France’s 1792 borders rather than the 1789 borders imposed by the actual settlement, and potentially a Napoleon who remained as French emperor under constraint rather than a returned Bourbon monarchy. The subsequent European political development - the Congress System, the Revolutions of 1830 and 1848, the eventual unification movements in Italy and Germany - would all have played out against a different baseline.
The analysis of Napoleon’s wars as both military campaigns and political transformations provides the necessary context for understanding why a Waterloo victory would not have resolved the structural tensions that the Napoleonic period had created throughout Europe. The spread of legal codification, administrative rationalization, and nationalist political movements was already underway across the territories Napoleon had controlled. Those political forces would have continued regardless of the specific outcome at Waterloo.
What the scenario clarifies about structural versus contingent causation is particularly instructive. Napoleon’s legal and administrative reforms - the Napoleonic Code, the prefectural system, the reorganization of the university and the judiciary - were implemented across territories that Napoleon no longer controlled by 1815. The Congress of Vienna restored political boundaries but could not restore the pre-Napoleonic administrative arrangements. The administrative and legal modernization that Napoleon had introduced across Europe persisted because it was functionally superior to what it had replaced, and because the local elites who administered it had built careers around it. A Waterloo victory would have preserved Napoleon’s personal power in France while changing the diplomatic settlement; it would not have changed the deeper political economy of European administrative modernization that the Napoleonic era had produced.
Prussia’s post-1806 military reforms illustrate this most clearly. Scharnhorst’s reorganization - the Scharnhorst-Gneisenau military reorganization, the beginning of conscription, the professionalization of the officer corps - were responses to Napoleonic military dominance. Those reforms made Prussia the dominant German military power by 1866, enabling the unification of Germany under Bismarckian leadership. A Napoleon who survived at Waterloo would have continued to face a Prussian military that had restructured itself explicitly to defeat him. The scenario’s downstream consequences for German unification are genuinely uncertain in timing but the structural forces driving eventual German state-formation under Prussian leadership were already substantially in place by 1815.
Italy presents a comparable case. Napoleon’s reorganization of Italian political geography had created the administrative experience of larger political units and the intellectual formation of Italian nationalism as a political program. The moderate liberals and radical republicans who organized the Risorgimento in the 1840s and 1850s drew on that Napoleonic administrative legacy even as they rejected Napoleonic political dominance. A continuing Napoleonic French empire in 1820 and 1830 would have presented Italian nationalism with a different set of opportunities and obstacles than the Austrian-dominated system that actually existed. The specific sequence of the Risorgimento - Cavour’s diplomatic maneuvering, Garibaldi’s expedition, the eventual Piedmontese monarchy - would have been different. Whether Italian unification in some form would have occurred is less certain to be different.
In analytical terms, the most significant contribution of this counterfactual is not the question of what France or Europe would have looked like in 1820 under a surviving Napoleon. It is the demonstration that the structural forces Napoleon had set in motion - legal modernization, nationalist politics, administrative rationalization, conscript armies - were already substantially independent of Napoleon’s personal survival by 1815. The contingency in the Waterloo scenario is real at the level of immediate political outcomes; it is much weaker at the level of the deeper structural transformations that define the Napoleonic period’s historical significance.
What If Archduke Franz Ferdinand Had Survived?
The assassination of Archduke Franz Ferdinand in Sarajevo on June 28, 1914 sits at the most analyzed decision-point in modern historical writing. The literature on the July Crisis that followed - the sequence of ultimatums, mobilizations, and declarations that converted a Balkan political murder into a continental war - is enormous. Christopher Clark’s The Sleepwalkers (2012), John Keegan’s The First World War (1998), and Margaret MacMillan’s The War That Ended Peace (2013) all engage the contingency question.
Initial circumstances were genuinely contingent: Franz Ferdinand was nearly not killed. The assassin Gavrilo Princip encountered him by chance after the first assassination attempt had failed, when Franz Ferdinand’s driver took a wrong turn. The assassination was both organized and lucky. Eliminating the luck - the driver takes the correct route back to the Konakstrasse - removes the spark but leaves all the combustible material in place.
Whether World War One would have occurred anyway is genuinely difficult to assess. The structural conditions for great-power conflict were present in 1914: the Anglo-German naval rivalry, the Austro-Hungarian confrontation with Serbian nationalism, the Franco-German tension over Alsace-Lorraine, the Russian need to maintain credibility as a Slavic great power after the humiliation of 1905. Several historians - Christopher Clark prominent among them - argue that the assassination was a trigger for pressures that would have found another outlet within a few years. Barbara Tuchman’s The Guns of August is more contingency-oriented, emphasizing how specific military planning commitments - particularly the Schlieffen Plan’s logic of attacking France through Belgium - locked the powers into a cascade that rational decision-making might have avoided.
Analytical weight of the evidence suggests that European war in some form before 1920 was likely regardless of June 28, 1914. The Austro-Hungarian Empire was in genuine internal crisis over South Slav nationalism and would have required some external assertion of authority. The question is whether a different trigger would have produced a differently configured conflict. A war that started in 1916 or 1917 rather than 1914 would have developed against different military-technological conditions - the tank, the aircraft, the submarine were all advancing rapidly in those years - and might have been shorter and less catastrophic than the actual four-year attrition.
Consequences of an avoided or significantly different World War One extend to the Russian Revolution of 1917, the rise of fascism in Italy and Germany, the Holocaust, and the entire political configuration of the twentieth century. This is precisely where the methodological caution about ripple effects becomes essential. The scenario supports the claim that a different Balkan crisis would have been likely and that some form of European great-power conflict was structurally probable. It cannot support confident claims about whether Lenin would have come to power, whether Hitler would have risen, or whether the specific catastrophes of 1939-1945 would have occurred.
The Russian dimension is where the downstream uncertainty is most acute. The February Revolution of 1917 that brought down the Romanov dynasty was driven by military exhaustion, food shortages, and the accumulated political failures of three years of catastrophic war. Without a war of that scale and duration, the structural preconditions for February 1917 simply do not exist in the same configuration. Russia in the early twentieth century had serious internal tensions - the 1905 revolution demonstrated that clearly enough - but the Romanov regime had survived 1905 with the constitutional concessions of the October Manifesto and had subsequently operated a functioning parliament, however imperfect. Whether those tensions would have produced a revolutionary rupture in the absence of total war is a genuine historical question. Historians of Russia as diverse as Orlando Figes and Richard Pipes have noted the degree to which the specific collapse of 1917 required the particular pressures of the war rather than being an inevitable outcome of pre-existing social contradictions. A plausible-world scenario in which the 1914 war is avoided, or in which a shorter, less devastating war ends before 1917, is a world in which Lenin’s particular historical window may never have opened. The Bolshevik seizure of power in October 1917 was itself highly contingent, dependent on a specific configuration of political weakness following the February Revolution and the Provisional Government’s decision to continue the war. That specific configuration required the war in the first place.
German fascism presents a similarly complex question that demands intellectual honesty about the limits of the counterfactual. Adolf Hitler’s political career was shaped at every point by the particular circumstances of German defeat in 1918 and the peace settlement of 1919. His early political formation in Munich in 1919 and 1920 was directly rooted in the stab-in-the-back mythology that arose from the unexpected surrender, the shock of which was amplified by the fact that German armies still occupied foreign territory when the armistice came. The Versailles settlement’s territorial provisions, reparations clauses, and the war-guilt article created the specific grievances that the National Socialist movement exploited most effectively. A Germany that did not fight the war, or that fought a different war with a different outcome, would not have had those specific grievances to exploit. Whether that means fascism of some variety could not have arisen in Germany - given the country’s other political tensions, anti-Semitic traditions, and economic vulnerabilities - is a different and harder question. What the counterfactual supports reliably is that the specific form of National Socialism that came to power in 1933, organized precisely around the humiliation of 1918 and 1919 and the promise to reverse it, required the actual history of the war as its generative material. The Holocaust, which required the particular institutional framework of the Nazi state, the specific conditions of a second world war, and the operational decisions of 1941 and 1942, represents exactly the kind of deeply downstream contingent outcome that responsible counterfactual analysis treats with great caution. The most that can be said is that an avoided or radically different World War One is a world in which the specific causal chain leading to the Holocaust does not exist. Whether other chains would have produced comparable atrocities is beyond the reach of any methodologically disciplined answer.
What If Hitler Had Been Killed Before 1944?
The multiple assassination attempts against Hitler represent a striking case of documented contingency. Between 1939 and 1944, there were at least fifteen serious attempts on his life that came close enough to leave historical records. The July 20, 1944 plot - Operation Valkyrie, led by Colonel Claus von Stauffenberg - came closest. Stauffenberg placed a briefcase containing two blocks of plastic explosive under the conference table at the Wolf’s Lair in Rastenburg. One block failed to detonate because time constraints prevented Stauffenberg from arming it. The other was shifted away from Hitler’s position when another officer moved the briefcase to avoid a table leg. Hitler survived with burst eardrums and minor burns.
Unusually well-documented in the historical record is what would have followed Stauffenberg’s success. The conspirators had a detailed plan - Operation Valkyrie itself - for seizing administrative control of Germany in the hours following the assassination. General Ludwig Beck was designated as head of state, Carl Friedrich Goerdeler as chancellor. The conspirators had established contact with Allied representatives about potential armistice terms and anticipated withdrawing Germany from the war.
The probability that Operation Valkyrie would have succeeded completely even with Hitler’s death is debated. The plan depended on the conspirators successfully activating Reserve Army units in Berlin before the SS could organize a loyalist response. The first communications went out on the assumption of Hitler’s death, and several regional commanders began responding before confirmation that Hitler had survived came through. Whether the conspirators would have prevailed against SS and Hitler Youth units in Berlin is genuinely uncertain.
What is clear is that any successful assassination in 1944 would have occurred when Germany’s military position was already deteriorating seriously. The Wehrmacht had been driven back from the Volga after Stalingrad, the Allies had landed in France, and synthetic fuel production was increasingly disrupted by Allied bombing. A conspiracy government seeking armistice in August 1944 would have faced the Allied demand for unconditional surrender, which had been declared at Casablanca in January 1943 and was not negotiable.
Most important among the scenario’s consequences is the potential avoidance of the final year of the Holocaust. By July 1944, the deportation of Hungarian Jews had been underway for less than two months. Approximately 800,000 Hungarian Jews were alive when Stauffenberg’s bomb went off. The murder of those people - along with the continued killing at Auschwitz-Birkenau through January 1945 - depended on the continuation of Hitler’s direct authority over the SS system. A successor government seeking armistice would have had strong pragmatic reasons to halt the deportations as a negotiating gesture, regardless of any moral calculation.
The detailed analysis of the Holocaust as a planned, industrial system of murder explains why the continuation of that system depended so specifically on Hitler’s continued authority and the preservation of the SS command structure through January 1945. The counterfactual reveals how much specific harm depended on the survival of a specific individual through a specific period.
Earlier assassination scenarios - Georg Elser’s November 1939 bombing of the Munich Bürgerbräukeller, which missed Hitler by thirteen minutes because Hitler left the beer hall earlier than planned - would have occurred before the Holocaust’s systematic implementation. The Einsatzgruppen mass shootings began in June 1941 with the invasion of the Soviet Union. The Wannsee Conference, which coordinated the “Final Solution,” took place in January 1942. An assassination in 1939 or 1940 would have found a Germany at war but without the systematic extermination program that defined Hitler’s specific ideological ambition. Whether an alternative German leadership in 1940 would have pursued the war to different ends, or sought an early negotiated settlement, depends on which conspirators and which political groupings would have taken power - a question without a confident answer, since the conspirators of 1939 were a more diffuse group than the July 1944 network.
Postwar Europe also raises a significant question in this scenario: would have looked like without the specific horrors of 1944-1945. The death marches from the concentration camps in early 1945, the final months of Allied combat operations through Germany, the Dresden bombing, the Berlin Battle - all of these resulted from the war’s continuation past the point where any rational assessment would have found further resistance futile. A Germany that negotiated armistice in August 1944 would have avoided those specific events while still facing occupation, reconstruction, and reckoning with the crimes of the preceding years. Whether a negotiated end rather than unconditional surrender would have produced a more stable German political reconstruction - as some historians have argued, citing the Versailles comparison - is genuinely uncertain, but the question is analytically tractable in ways that purely speculative alternate histories are not.
What If the Cuban Missile Crisis Had Gone Nuclear?
The Cuban Missile Crisis of October 1962 is the twentieth century’s most intensively studied case of crisis management and near-catastrophe. The thirteen days between October 16 and October 28, 1962, when the Kennedy administration and the Khrushchev government navigated the confrontation over Soviet missiles in Cuba, produced multiple documented near-misses that deserve individual analysis.
Most significant of these involved a Soviet submarine, B-59, operating in the quarantine zone around Cuba on October 27, 1962. The submarine had lost contact with Moscow for several days and its crew was experiencing severe heat and oxygen depletion. American destroyers had been dropping practice depth charges to force the submarine to surface - the crew had no way to know these were non-lethal signals rather than actual attacks. The submarine’s commander, Valentin Savitsky, ordered the submarine’s nuclear torpedo armed. Under Soviet operational protocols, launching required authorization from three officers: Savitsky, the political officer Ivan Maslennikov, and the flotilla commander Vasili Arkhipov. Savitsky and Maslennikov both authorized. Arkhipov refused. His name was largely unknown to the public for decades after the crisis.
Minimum-change logic is satisfied here: Arkhipov agrees with his two colleagues rather than refusing. The B-59 launches its nuclear torpedo against the USS Randolph. The American naval response - and the subsequent escalation through the entire theater of the confrontation - follows from that single decision by a single man. What makes this scenario particularly significant from a methodological perspective is that the conditions aboard B-59 on October 27 were almost perfectly designed to produce the worst outcome. The crew had been submerged without radio contact for days. They were physically weakened by heat and oxygen depletion. Their instruments indicated that depth charges were striking nearby. From inside the sealed hull of the submarine, the operational situation was indistinguishable from being under genuine attack in the opening hours of a war that had already begun on the surface without their knowledge. Arkhipov’s refusal was not merely a brave decision; it was a decision made under conditions that systematically argued for the opposite conclusion. The scenario thus illuminates not only the contingency of the crisis’s resolution but the specific psychological and situational pressures that drove the near-miss. Nuclear deterrence in its theoretical form assumed rational actors with good information; Arkhipov’s situation was designed by circumstance to produce irrational actors with the worst possible information.
American nuclear targeting plans for the Soviet Union in 1962 called for a massive first strike in response to any Soviet nuclear use. The Single Integrated Operational Plan for 1962 designated over 1,000 targets across the Soviet Union and China. The American response to a Soviet nuclear torpedo would almost certainly have triggered the full implementation of that plan. The Soviet Union maintained comparable retaliatory capabilities. The result, modeled by various analyses since the crisis’s documents were declassified, would have been somewhere between 100 million and 600 million immediate deaths across the Northern Hemisphere, with subsequent famine and disease killing hundreds of millions more in what we now call nuclear winter conditions.
Analytical value of this scenario lies not in the speculation about casualties, which is genuinely uncertain at every level. Its value is in what it reveals about how the actual crisis was resolved. The detailed reconstruction of the Cuban Missile Crisis decision-making demonstrates that the resolution depended on the combination of private diplomatic communication, Kennedy’s willingness to accept a face-saving formula for Khrushchev, and the specific procedural structure of Soviet naval command that required three authorizations for nuclear torpedo use. Remove any one of those factors and the outcome changes decisively.
The October 1962 crisis also demonstrates the limits of structural deterrence arguments. The doctrine of mutual assured destruction held that rational actors would not initiate nuclear exchange because the consequences were unacceptable. Arkhipov, Savitsky, and Maslennikov were not facing a rational-actor calculation; they were facing genuine uncertainty about whether they were under attack, severe physical stress from oxygen depletion, and the possibility that their country was already at war. Deterrence theory assumed away precisely the conditions under which the actual near-miss occurred.
The B-59 episode was not the only near-miss of the thirteen days. On October 27 - the same day as the submarine crisis - an American U-2 reconnaissance aircraft strayed into Soviet airspace over the Chukotka Peninsula due to a navigational error. Soviet MiG interceptors scrambled; American F-102 fighters in Alaska, armed with nuclear-tipped Falcon missiles, were sent to escort the U-2 back to American airspace. The F-102s were equipped with nuclear air-to-air missiles because their conventional armament had been removed earlier that year. Had the Soviet fighters engaged the American escort aircraft, the first nuclear weapon use of the crisis might have come from an American defensive engagement over Soviet territory rather than from a Soviet submarine in the Caribbean.
Simultaneously, a different U-2 was shot down over Cuba on October 27, killing the pilot, Major Rudolf Anderson. Kennedy was under pressure from military advisers to respond immediately with airstrikes against Cuban surface-to-air missile batteries. He chose not to. A different president - or the same president under slightly more acute pressure - might have ordered those airstrikes, beginning the conventional military conflict that both sides had agreed would likely escalate to nuclear use.
What the concentration of near-misses on October 27, 1962 reveals is that the crisis’s peaceful resolution required multiple independent decisions to go the cautious way rather than the escalatory way simultaneously. Arkhipov’s refusal, Kennedy’s non-retaliation for the U-2 shootdown, the U-2 navigational error not triggering a confrontation, and the continued functioning of the back-channel communication between Robert Kennedy and Soviet Ambassador Anatoly Dobrynin - all of these had to hold at once. The probability that all of them would hold was not high. The probability that the crisis would be resolved peacefully in the form it actually took was substantially lower than the probability that some form of escalation would occur. The world of October 1962 was genuinely and substantially more contingent than post-hoc accounts of crisis management tend to suggest.
The comparative examination of how revolutionary transformations interact with military confrontation provides useful context for understanding why the Cuban revolutionary government created the political conditions that made the missile crisis possible. Castro’s Cuba was not simply a pawn in Soviet-American confrontation; it was an active actor whose willingness to accept Soviet missiles, despite understanding the risks they created, reflected the specific political logic of a revolutionary government seeking security guarantees against American intervention.
The Structure-Contingency Distinction
Surveying the ten scenarios together, a pattern emerges that the Ferguson-Tetlock framework predicts: the distribution between structural and genuinely contingent outcomes is uneven. Some of history’s most dramatic apparent near-misses turn out to be predominantly structural: the Armada’s failure, for example, reflects deep planning problems that the storms finalized rather than created. Napoleon’s long-term defeat looks increasingly structural the more carefully the Coalition’s resource advantage is examined. Other scenarios cluster more clearly toward genuine contingency: the Cuban Missile Crisis’s resolution, the July 20 plot’s failure, and the Mongol western campaign’s suspension all depended on single decision-points with minimal structural overdetermination.
The practical implication is that different kinds of historical analysis are appropriate for different scenarios. Structurally overdetermined outcomes call for analysis of the forces that made them likely rather than the specific events that produced them. Genuinely contingent outcomes call for attention to the specific decision-makers, information environments, and pressures that shaped the choices made. Counterfactual analysis helps identify which type of analysis a given case requires.
Robert Cowley’s edited volume What If? The World’s Foremost Military Historians Imagine What Might Have Been (1999) collects case studies from professional historians across many of these scenarios, and its most consistent finding is that professional historians are substantially more cautious than popular counterfactual writers about projecting confident alternative timelines more than a decade or two beyond the initial change point. The accumulated uncertainty compounds too quickly for analytical rigor to survive.
The comparative approach that drives this article - examining ten scenarios against a common methodological standard - also reveals something about the nature of historical causation that individual scenario analysis tends to obscure. The clustering of genuinely contingent moments in certain periods (the fifth century BCE, the sixteenth century, the twentieth century’s crisis years) reflects the structural conditions that made those periods generative of large-scale change. The contingency was real; but the conditions that made contingent choices so consequential were structural.
Readers interested in how the revolutionary dynamics explored across these scenarios connect to the comparative analysis of great transformations will find a useful complement to the counterfactual framework. Revolutionary moments are precisely the occasions when contingent choices interact with structural pressures to produce outcomes that neither force alone would have generated.
The relationship between individual leadership and structural conditions also deserves elaboration. The comparative examination of history’s greatest military commanders addresses the question of how much individual military genius contributes to outcomes versus structural advantages in resources and geography. That analysis bears directly on several of these counterfactuals, particularly the scenarios involving Alexander, Napoleon, and the Cuban Missile Crisis commanders.
A matrix of the ten scenarios by methodological classification illuminates the pattern further. At the high-contingency end - where the specific outcome depended primarily on individual decisions with minimal structural overdetermination - sit the Cuban Missile Crisis (Arkhipov’s refusal), the Mongol halt (Ogedei’s death), and the July 20 plot (the briefcase position). At the high-structure end - where structural forces were dominant and the specific events were more triggers than causes - sit the Armada, the American Revolution (whose ideological consequences would have emerged through other channels), and Napoleon’s defeat (whose structural basis in Coalition resources was decisive). In between, most scenarios show genuine mixture: the Persian Wars involved close-run contingent battles in the context of structurally significant civilizational competition; the Roman Empire combined structurally generated pressure with specific contingent administrative failures; the Franz Ferdinand assassination triggered a crisis whose structural preconditions made some form of major European conflict likely.
This distribution has an important implication for how historians should approach major historical transitions. The romantic view of history - that great individuals and decisive battles are the primary movers of civilizational change - overstates contingency at the expense of structural analysis. The determinist view - that structural forces make individual decisions largely irrelevant - understates the genuine contingency that appears repeatedly across the historical record. The counterfactual method, applied with discipline, helps calibrate between these extremes by forcing explicit evidence-based judgments about where specific outcomes fell on the spectrum.
The analysis connects directly to the larger questions about what history can and cannot tell us about the future. Structurally determined outcomes in the past are not necessarily indicators of structurally determined outcomes in the present, because structural conditions change. The conditions that made European great-power conflict highly probable in 1900-1914 are not present in the same form today. The conditions that made Cuban Missile Crisis management so contingent - nuclear arsenals under fragmented command structures, limited real-time communication between adversaries, intense time pressure and information scarcity - have changed in some ways and persisted in others. Understanding the structure-contingency balance in historical cases is not a direct guide to present policy, but it is an essential component of realistic thinking about how historical change occurs.
Counterfactual History and Its Critics
Evans’s critique deserves a fair engagement rather than dismissal. His core concern is that counterfactuals are too easily bent toward confirming the values their authors brought to them. A German nationalist writing about Stauffenberg naturally reaches different conclusions than a Pacific War historian writing about the same event. The ideological valence of alternate histories - which tend to moralize about how much better or worse things might have been - is a genuine methodological vulnerability.
Evans’s concern, however, applies equally to conventional history and is not unique to counterfactuals. to conventional history. Historians writing about what actually happened are not free of ideological valence; the selection of events to emphasize, the causal explanations preferred, and the moral frameworks applied all reflect the historian’s values. The solution is not to avoid counterfactuals but to subject them to the same critical scrutiny applied to conventional historical claims.
Ripple-effect concerns that Evans emphasizes are real in a different sense. A counterfactual that changes 1914 and then confidently describes 1960 has accumulated so much inferential uncertainty that its specific claims deserve almost no confidence. But this is a problem of scope rather than of method. A counterfactual that changes June 28, 1914 and then describes the July Crisis that did not occur is within the range of evidence-supported inference. The method is not illegitimate; its application must be constrained by honest acknowledgment of confidence limits.
The larger point that counterfactual analysis serves is one that Evans himself partly endorses in practice: understanding what was structurally determined versus what was genuinely contingent is essential for understanding historical causation. Refusing to ask counterfactual questions does not make causation clearer; it leaves causal claims about inevitability unexamined. The historian who says the Roman Empire “had to” fall at roughly the time it did is making an implicit counterfactual claim - that no plausible alternative intervention would have prevented the outcome. Making that claim explicit and testing it against the evidence is what rigorous counterfactual history does.
Tracking all ten scenarios across their respective decision-points, and placing them within the longer sweep of causation, is exactly the kind of chronological mapping that structured tools like the interactive World History Timeline on ReportMedic make accessible. Seeing where these moments cluster and what structural conditions surrounded them changes the analytical relationship to each individual case.
A secondary critique of counterfactual history comes not from Evans but from a different methodological tradition: the social-scientific concern that counterfactuals in history cannot be validated against evidence in the way that scientific counterfactuals can. A chemist who asks “what if we had added reagent X instead of Y?” can run the experiment. A historian who asks “what if Ogedei had died five years later?” cannot. The counterfactual is necessarily speculative in a way that laboratory experiments are not.
This objection is genuine but proves too much. Most of the important questions in the social sciences cannot be settled by controlled experiment either. Economists cannot run the Great Depression again with different monetary policy to test the Friedman-Schwartz thesis. Political scientists cannot assign countries randomly to democratic and authoritarian governance to measure the effects on growth. The entire field of comparative historical analysis depends on using variation across cases to approximate experimental conditions without actually producing them. Counterfactual history is doing the same thing within a single case rather than across multiple cases, and the methodological standards that apply are analogous rather than uniquely demanding.
The more defensible version of the methodological objection is not that counterfactuals are impossible but that they are harder to discipline than comparative analysis. When comparing the American and French Revolutions, the comparison is grounded in documented evidence about both cases. When asking what would have happened if the French Revolution had not occurred, the analyst is reasoning from evidence about one case into a hypothetical that has no documentary record. The discipline must come from the minimum-change constraint, from explicit acknowledgment of confidence limits, and from the willingness to say “I do not know” at the point where inference outstrips evidence.
That discipline is achievable, and the scholars who have developed serious counterfactual methodology - Ferguson, Tetlock, Hawthorn, Lebow - have been explicit about when they are confident and when they are speculating. The problem in the popular market for alternate history, where confidence is commercially rewarded and uncertainty is commercially punished, is precisely that this discipline is abandoned. The analytical response is not to abandon the method along with its abuses but to enforce the standards that serious scholarship has developed.
Why This Methodology Belongs in the Classroom
Counterfactual history’s educational value extends beyond the interesting scenarios it generates. Teaching students to ask “was this outcome inevitable?” develops the same analytical muscle that distinguishes historical thinking from historical memorization. A student who can identify that the Cuban Missile Crisis’s resolution depended on Arkhipov’s specific decision has understood something about historical causation that a student who simply memorizes the crisis’s dates and outcomes has not.
Methodological discipline that counterfactual exercises impose is also pedagogically valuable. Distinguishing plausible from merely possible counterfactuals requires evidence engagement. Students who want to argue that the Roman Empire would have survived if it had simply been better governed must engage the evidence about what kinds of governance alternatives were actually available within the Roman political tradition. The constraint of plausibility keeps counterfactual exercises tethered to the historical record rather than drifting into pure invention.
The scenarios examined here also illustrate a point about historical contingency that has significant implications beyond the classroom. The view that history moves toward inevitable outcomes - that modernity, liberal democracy, and technological progress are the necessary destinations of the human story - is undercut by taking seriously the moments where different choices would have produced genuinely different trajectories. The Persian Wars did not have to end as they did. The Mongols’ western advance was halted by a death that could easily have come later. The Cuban Missile Crisis was resolved by a procedural rule and one man’s willingness to invoke it.
The structural conditions of any historical moment constrain but do not determine its outcomes. Good counterfactual analysis helps identify where constraint ends and genuine choice begins. That identification is not merely an academic exercise. It is an argument that the choices being made in any present moment are genuinely consequential in ways that deterministic readings of history deny.
Fate and necessity are recurring themes in the classic literary tradition precisely because the question of what is determined versus what is genuinely chosen has fascinated human thought across every period of recorded culture. The literary exploration of those themes - from Greek tragedy through Victorian naturalism through modernist fiction - provides an illuminating parallel framework to the historical analysis in this article. The comparative examination of how major literary works test different philosophical frameworks about fate and free will reveals that the greatest literary minds have consistently refused to settle the question in either direction, which is itself an important finding about the nature of historical agency.
The full chronological context for all ten scenarios examined here - from the Persian Wars through the Cold War - is best understood within the long sweep of recorded human history. Exploring how these moments fit within the broader patterns of world history through the interactive timeline tool provides the comparative depth that individual scenario analysis cannot replicate.
Practical pedagogy of counterfactual analysis also addresses one of the chronic failures of historical education: the tendency to treat historical figures as inevitable successes or failures rather than as decision-makers operating under uncertainty. Xerxes at Salamis, Batu Khan in Hungary, Philip II planning the Armada, Kennedy and Khrushchev in October 1962 - none of these figures knew how their decisions would turn out. Teaching counterfactual history means recovering the genuine uncertainty that all historical actors faced and that retrospective narrative tends to eliminate. A student who understands that Arkhipov did not know in October 1962 whether his country was at war understands something about decision-making under uncertainty that applies to politics, business, and personal life well beyond the historical case. Counterfactual history at its best is not only an analytical tool for understanding the past; it is an intellectual discipline that trains the capacity to think carefully about causation, contingency, and the relationship between individual choice and structural constraint that defines every significant human situation.
The methodological case for counterfactual history in the classroom ultimately rests on the same foundation as the methodological case for teaching scientific experimentation rather than scientific results. The what-if question is history’s equivalent of the experimental variable: it isolates specific causal factors by asking what would have happened if they had been different. Understanding how to formulate that question rigorously, how to assess what evidence can and cannot support, and how to acknowledge the limits of confident inference - these are intellectual skills that extend far beyond the specific historical cases in which they are developed. A student who has worked through the structure-contingency analysis of the Persian Wars is better equipped to analyze any complex causal situation, historical or otherwise, than a student who has memorized the dates of Marathon and Salamis without ever asking why those battles mattered or whether they had to turn out as they did.
Frequently Asked Questions
Q: What is counterfactual history?
Counterfactual history is a methodological approach that examines how historical outcomes might have differed if specific decision-points, events, or circumstances had been different. It is not science fiction or idle speculation, but a disciplined form of causal analysis. The key methodological constraint, articulated by Niall Ferguson, Philip Tetlock, and Richard Ned Lebow, is that rigorous counterfactuals must require the minimum number of changes from the documented historical record, must be grounded in specific evidence about the decision-point in question, and must acknowledge the rapidly increasing uncertainty that accumulates as the chain of consequences lengthens. The method’s value is that it forces explicit engagement with questions of historical causation that conventional history often leaves implicit.
Q: Are some historical outcomes actually inevitable?
Some outcomes are structurally overdetermined, meaning they would have resulted from multiple independent causal pathways even if any single pathway had been blocked. The fall of the Western Roman Empire is often cited as an example: the demographic, fiscal, and military pressures on the Empire in the fifth century were severe enough that blocking any single contingent event - any specific battle lost or any specific assassination - would likely have redirected rather than prevented the eventual fragmentation. But even heavily structural outcomes usually involve contingent elements that affected their timing, their specific form, and their immediate human costs. Very few historical events are so structurally determined that no plausible intervention could have changed any aspect of their occurrence.
Q: What if Rome had never fallen?
The scenario requires precision about what “Rome falling” means. The Eastern Roman Empire continued until 1453 CE. What ended in 476 CE was Western Roman political authority over the former western Mediterranean provinces. A more useful question asks what would have been required to maintain that political authority through the fifth century. The answer involves both structural conditions - fiscal reform, a more sustainable frontier policy, different administrative approaches to Germanic peoples - and specific contingent events, particularly the Rhine crossing of 406 CE and the assassination of Aetius in 454 CE. Even with a surviving Western imperial administration, the cultural and religious contents of Roman civilization would have continued to develop, since they did so even without it. The specific loss from Western political fragmentation was administrative continuity, urban density, and literacy rates, all of which declined significantly in the former western provinces.
Q: What if Hitler had been assassinated earlier?
The most thoroughly analyzed version of this scenario involves the July 20, 1944 Operation Valkyrie plot, which failed because Claus von Stauffenberg could not arm both explosive charges and because another officer shifted the briefcase away from Hitler’s position. An earlier assassination - say the 1939 Georg Elser bombing of the Munich beer hall, which missed Hitler by thirteen minutes - would have occurred before the systematic murder of European Jews had been fully organized. Einsatzgruppen mass shootings had begun in the east by mid-1941 and the death camps were operational by late 1942. An assassination in 1943 or 1944 would still have found the Holocaust extensively underway, but approximately 800,000 Hungarian Jews were still alive in July 1944 when the July 20 plot failed. A successor government seeking armistice would have had pragmatic reasons to halt deportations as a negotiating gesture. The war in Europe would likely have ended months earlier, saving hundreds of thousands of additional military casualties on both sides.
Q: What if the Spanish Armada had won?
A successful Armada landing would have required solving a logistical problem that Philip II’s planning never resolved: the Armada’s deep-drafted ships could not rendezvous with Parma’s army in the shallow Flemish waters. A hypothetical version of the scenario in which this problem was solved produces a Spanish army landing in Kent against Elizabethan defensive forces that were substantially weaker than the Continental forces Spain regularly faced. An initial Spanish military success is plausible. The longer-term scenario is far less certain. English Protestantism had deep popular roots by 1588, and a Spanish Catholic occupation would have faced sustained resistance. The scenario’s most significant consequence concerns Atlantic colonization: English colonization of North America was still in its earliest stages in 1588, and Spanish control of England would have redirected the resources that produced the Virginia settlements and eventually the English colonial presence in North America.
Q: Was World War One inevitable?
The structural conditions for great-power conflict in 1914 were significant: the Anglo-German naval rivalry, Austro-Hungarian internal crisis, Franco-German tension, and Russian need to maintain credibility as a Slavic power after the humiliation of 1905. Several historians argue these pressures would have produced a major European war within a decade regardless of the June 28, 1914 assassination. The counterfactual value of examining Franz Ferdinand’s survival is less about whether any European war occurred and more about whether the specific war that did occur would have had the same configuration. A war beginning in 1916 or 1917 rather than 1914 would have developed against different military-technological conditions. A war beginning over a different pretext might not have involved the same alliance configurations. The specific shape of the 1914-1918 conflict - particularly the Western Front’s static attrition - was not structurally predetermined even if some form of European great-power war was probable.
Q: What if JFK had not been assassinated?
This scenario falls outside this article’s primary chronological focus but merits brief engagement because it is among the most frequently asked counterfactual questions. Kennedy’s assassination on November 22, 1963 presents a genuine contingency problem: Lee Harvey Oswald’s success depended on a specific motorcade route change made the morning of the Dallas visit. The policy question most often attached to the scenario concerns American escalation in Vietnam. Kennedy had authorized 16,000 military advisers in Vietnam by late 1963 and had signed National Security Action Memorandum 263 calling for withdrawal of 1,000 advisers by year’s end. Whether Kennedy would have escalated to the 500,000-troop commitment that Lyndon Johnson eventually made is genuinely uncertain. Kennedy’s record on foreign policy shows both hawkish and cautious impulses in roughly equal measure. The scenario cannot support confident claims about Vietnam; it can support the claim that American policy in Vietnam was not structurally determined to escalate as it did.
Q: What if the Cuban Missile Crisis had gone nuclear?
Vasili Arkhipov’s refusal to authorize the nuclear torpedo on Soviet submarine B-59 on October 27, 1962, is the most precisely documented near-miss in nuclear history. American depth charges had been falling around the submarine for hours; the crew had no way to know these were non-lethal signals. If Arkhipov had agreed with his two colleagues who wanted to fire, the USS Randolph would have been hit by a nuclear torpedo. American nuclear war plans for 1962 called for massive retaliation against Soviet targets in response to any nuclear use. The immediate casualties from full nuclear exchange, based on the weapons deployed and targeting plans declassified since, would have been in the range of 100-600 million deaths across the Northern Hemisphere, with subsequent nuclear winter conditions producing famine and disease deaths that would have substantially depopulated both superpowers and caused significant mortality across the rest of the Northern Hemisphere. The scenario demonstrates the specific limitation of deterrence theory: the doctrine assumes rational actors with accurate information about each other’s capabilities and intentions, precisely the conditions that did not exist on B-59 in October 1962.
Q: Is alternate history useful for historians?
Alternate history in popular fiction - novels, films, television series - serves different purposes than counterfactual analysis in historical scholarship. Popular alternate history uses changed historical premises as narrative devices to explore cultural anxieties, examine the nature of contingency, and generate thought experiments. It does not claim to be analytically rigorous and should not be evaluated by that standard. Counterfactual analysis in historical scholarship is constrained by the requirement of plausibility - scenarios must be grounded in documented evidence, must require minimum changes from the historical record, and must be explicit about confidence limits. The two practices share the basic question (what if?) but diverge sharply in their methodology and their standards for a good answer.
Q: What if the Mongols had not been stopped at Legnica and Mohi?
The Mongol advance into Central Europe in 1241 was halted not by military defeat but by Ogedei Khan’s death in December 1241, which required Batu Khan’s forces to withdraw for the kurultai succession assembly. Had Ogedei died in 1245 rather than 1241, the western campaign would have continued through the spring of 1242. European military forces had performed poorly against Mongol combined-arms tactics in Poland and Hungary, and there is no strong evidence that French or German forces would have performed significantly better. A Mongol advance to the Rhine was plausible militarily. The long-term scenario is much less certain: the Mongol administrative model in conquered territories varied significantly, and western Europe’s value to Mongol empire-builders as an extractable revenue source was lower than the richer urban economies of Persia and China that the Mongols administered more carefully.
Q: Did the assassination of Franz Ferdinand cause World War One?
The assassination was the trigger for the July Crisis but not its sole cause. The structural conditions for great-power conflict were extensively present before June 28, 1914. The assassination provided Austria-Hungary with a pretext for action against Serbian nationalism that Austro-Hungarian leaders had been seeking. The German blank check of support to Austria-Hungary, the Russian decision to mobilize in support of Serbia, and the German implementation of the Schlieffen Plan’s attack on France through Belgium all followed from pre-existing plans and commitments that the assassination activated rather than created. The most accurate formulation is that the assassination was a necessary but not sufficient cause of the war in its specific 1914 configuration. A different trigger at a different moment might have produced a differently configured conflict; it would probably have produced some form of European great-power confrontation within the following decade.
Q: What is the difference between structural and contingent causes?
Structural causes are conditions that make certain outcomes likely or nearly inevitable regardless of specific individual choices or events. Contingent causes are specific events, decisions, or circumstances that could plausibly have been otherwise. The distinction matters because it determines what kind of historical analysis is appropriate. Structurally determined outcomes call for analysis of the forces that made them likely. Genuinely contingent outcomes call for close analysis of specific decision-makers, information environments, and the pressures they faced. Most significant historical events involve both types of cause, and the analytical task is identifying which elements belong to which category. The periodic confusion of structural with contingent causes produces both the deterministic fallacy (everything that happened had to happen) and the voluntarist fallacy (outcomes are always the product of individual choices unmediated by structural conditions).
Q: Are there outcomes that depend on a single person?
The historical record contains several genuine cases where specific individuals appear to have been decisive rather than substitutable. Arkhipov’s refusal on B-59 is the clearest twentieth-century case: no structural force produced his decision, and no comparable figure was available to make the same decision if he had not. Stauffenberg’s role in the July 20 plot is another: no other German officer with comparable access and motivation was positioned to make the same attempt. Alexander the Great presents a different version of the question: Alexander’s specific military genius was not structurally produced by Macedonian conditions, but the geopolitical opportunity he exploited was. Whether another Macedonian commander would have produced comparable conquests is unknowable, but the structural opportunity existed independently of Alexander’s particular capacities. The general finding is that individuals are most consequential at decisive moments when structural conditions have produced a situation with multiple plausible outcomes and where specific individual choices determine which outcome occurs.
Q: How should students approach counterfactual history?
The most productive approach is to start with minimum-change questions rather than maximum-change questions. “What if Hitler had not been born?” requires changing so many prior circumstances that the question generates no analytical traction. “What if Stauffenberg had armed both explosive charges on July 20, 1944?” changes one documented fact that could easily have been different and generates specific analytical questions about the July 20 conspirators’ plan, its likelihood of success, and its likely consequences for the war. Starting from documented decision-points, using evidence about what the decision-makers actually knew and faced, and being explicit about where confidence ends - these are the practices that make counterfactual analysis intellectually rigorous rather than purely speculative.
Q: What if Alexander had not conquered Persia?
Alexander’s campaign against the Achaemenid Persian Empire, launched in 334 BCE, depended on specific Macedonian political conditions: Philip II’s reorganization of the Macedonian army, the financial reserves Philip had accumulated, and the political consolidation of the Greek city-states under Macedonian hegemony after the Battle of Chaeronea in 338 BCE. Without Alexander specifically, a differently led Macedonian force might have made more limited gains in Anatolia while preserving the Persian heartland. The Achaemenid Empire under Darius III was experiencing internal stresses - satrapal revolts, succession tensions - that would have continued regardless of Macedonian pressure. The scenario of a surviving Achaemenid Empire through the third and second centuries BCE would have significantly altered the context for Roman expansion eastward, since the Romans eventually absorbed the former Hellenistic kingdoms that the Wars of the Diadochi had created from Alexander’s empire.
Q: What would have happened if nuclear weapons had not been developed in World War Two?
This scenario lies at the boundary of the plausible counterfactual method because it requires changing not a specific decision-point but the outcome of a broad scientific and engineering program involving thousands of researchers across multiple countries. The more tractable version asks what would have happened if the Manhattan Project had been slower by three years and the first nuclear tests had occurred in 1948 rather than 1945. The Pacific War scenario changes substantially: the planned invasion of Japan’s home islands (Operation Downfall, planned for November 1945) would likely have proceeded, with casualty estimates ranging from hundreds of thousands to several million depending on the source and methodology. The Cold War’s technological competition would have been delayed but not prevented. The scenario illustrates the limitation of individual-focused counterfactuals: when an outcome depends on the collective scientific progress of thousands of people, no single minimum change produces the alternate outcome.
Q: How does the contingency-structure debate apply to contemporary events?
The analytical framework developed by Tetlock and Ferguson applies to any historical moment, including those close to the present. Understanding whether a contemporary crisis is structurally determined or genuinely contingent - whether the outcome depends on large-scale forces or on the specific choices of specific decision-makers - has direct implications for how effective different kinds of interventions can be. A crisis that is primarily structural calls for engagement with the conditions producing it. A crisis that is primarily contingent calls for attention to the decision-makers and the pressures they face. The counterfactual method is not merely an exercise in historical nostalgia; it is an analytical tool for understanding causation in any time period, including the present.
Q: Is it disrespectful to speculate about events like the Holocaust?
Concerns about disrespect in counterfactual engagement with atrocity are understandable but misplaced when the engagement is rigorous and purposeful. The question “what if Hitler had been assassinated in 1944 and approximately 800,000 Hungarian Jews had survived?” is not speculation about the Holocaust for entertainment. It is an analytical exercise that reveals how specifically the continuation of that murder system depended on specific political and military conditions. That revelation serves the purpose of historical understanding rather than undermining it. The disrespectful counterfactual is the one that treats atrocity as a narrative device or that normalizes it by placing it within a speculative adventure story. Rigorous counterfactual engagement with atrocity insists on its contingency - insists that it did not have to happen - in ways that support rather than diminish serious historical reckoning.
Q: What is the most important lesson from studying history’s biggest what-ifs?
Most important of all is a lesson not about any specific scenario but about the nature of historical causation itself. The persistent human tendency to read the past as a sequence of inevitable developments - to assume that what happened had to happen - is both psychologically understandable and analytically misleading. The ten scenarios examined here demonstrate that even the most consequential events in world history contained specific contingent elements where different choices would have produced genuinely different outcomes. This finding does not dissolve into relativism or into the claim that everything is random. Structural conditions are real, and they constrain the range of plausible outcomes at any given moment. The lesson is that constraint is not determination, and that historical analysis requires distinguishing carefully between the two.
