The Year Without Summer Nobody Remembered: How AD 536 Became the Worst Year in Human History and Why It Took 1,500 Years for Anyone to Notice

The anomaly

In approximately AD 536, something happened to the sun.

The Byzantine historian Procopius of Caesarea, attached to the staff of the general Belisarius and stationed in Italy, recorded that "the sun gave forth its light without brightness, like the moon, during this whole year, and it seemed exceedingly like the sun in eclipse." He noted, as a separate matter, that the years following were marked by famine and unprecedented pestilence.

The Italian Roman statesman Cassiodorus, writing to a provincial prefect, described the same phenomenon in a different vocabulary. The sun, he wrote, was "of a bluish colour." Shadows had lost their definition at noon. The seasons had become "jumbled up together" and harvests had failed. He framed the letter as a practical request — the empire needed to reorganize tax and grain distribution because the agricultural calendar had stopped working.

The Syriac chronicler John of Ephesus reported, independently, that the sun "became dark and its darkness lasted for eighteen months," shining for only a few hours per day and producing fruit that would not ripen. The Chinese Bei Shi chronicle, written from a completely different cultural tradition with no contact with the Mediterranean sources, recorded yellow dust falling like snow, summer frosts killing the crops, and famines lasting three successive years. The Annals of Ulster in Ireland recorded, with characteristic brevity, "a failure of bread."

For most of the past fifteen hundred years, mainstream Western historians treated these accounts as rhetorical embellishment. Procopius had a reputation for dramatization. Cassiodorus was understood as a polished stylist whose descriptions of natural phenomena were taken to be figurative. The Chinese, Irish, and Syriac sources were studied largely in isolation from one another and never assembled into a unified account. The dominant assumption was that the chroniclers were exaggerating.

The chroniclers were not exaggerating.

Background context

The first hard scientific evidence that the literary accounts described a real event came from dendrochronology — the study of tree rings.

Beginning in the 1990s, multiple research groups produced cross-referenced tree-ring records from oaks in Ireland, Scots pines in Finland, and bristlecone pines in the American Southwest. The records, when aligned, showed an unmistakable signature: the worst sustained climatic disruption in the past two thousand years of dendrochronological data, beginning in approximately AD 536 and continuing, with secondary pulses, for over a decade.

The disruption was global. It was simultaneous across the Northern Hemisphere. It was severe enough to suppress tree growth across continents.

The orthodox explanation, by the early 2000s, was that some kind of "atmospheric dust veil" had blocked solar radiation. The most plausible candidate was a major volcanic eruption — large enough to inject sulfur aerosols into the stratosphere, where they would scatter incoming sunlight for years before settling out.

The candidate eruption, however, was not identified.

That identification came in 2015. Michael Sigl and an international team of ice-core researchers, working with newly-refined chronologies of ice cores from both Greenland and Antarctica, published a comprehensive timeline of volcanic activity over the past 2,500 years in Sigl et al., Nature (2015). Their data identified not one but a cluster of large eruptions in the relevant period: a major eruption in approximately 536 — most likely high-latitude Northern Hemisphere, with leading candidates including an Icelandic volcano — followed by a second, even larger eruption in approximately 540, of uncertain location but with chemical fingerprints consistent with a low-latitude tropical source.

Markus Toohey and colleagues subsequently modeled the climatic consequences in Toohey et al., Climatic Change (2016), demonstrating that a "double event" of the magnitude indicated by the ice cores would have produced exactly the climatic signature observed in the dendrochronological record — sustained cooling, suppressed sunlight, agricultural failure across multiple latitude bands.

Ulf Büntgen and a separate consortium then formalized the broader concept in Büntgen et al., Nature Geoscience (2016), proposing that the AD 536 disruption marked the onset of a sustained cool period — the Late Antique Little Ice Age — lasting from approximately 536 to around 660 CE, during which average temperatures in much of the Northern Hemisphere remained measurably depressed.

The chroniclers had been telling the truth. It had simply taken fifteen centuries for Western science to develop the tools to confirm it.

Evidence presentation

The case that AD 536 represents one of the most consequential single years in human history rests on three convergent lines of evidence: contemporary written accounts, paleoclimatic data, and archaeological evidence of the downstream cascade.

The written record is, by the standards of late-antique sources, unusually robust. Procopius is one source. Cassiodorus is a second. John of Ephesus, a third — writing in Syriac from a completely different theological and political tradition. The Chinese Bei Shi adds a fourth, geographically remote witness. The Irish annalistic tradition adds a fifth. Pseudo-Zacharias of Mytilene, a sixth. Michael the Syrian, working later but drawing on earlier sources, a seventh. The convergence of these independent sources — across cultures and languages with limited contact — is the kind of cross-confirmation that historians of the period rarely possess for any single event.

The paleoclimatic evidence is, by 2020s standards, well-developed:

• Greenland ice cores show a sulfate spike consistent with a major high-northern-latitude eruption in 536, and a second larger spike in 540 (Sigl et al., 2015).
• Antarctic ice cores show a corresponding spike for the 540 event, consistent with a low-latitude tropical eruption that distributed sulfate to both hemispheres.
• Alpine ice cores drilled from the Colle Gnifetti glacier on the Italian-Swiss border preserve a year-by-year layer record that, in Loveluck, McCormick et al., Antiquity (2018), was used to date the 536 event with unprecedented precision and to trace its consequences across European economic activity.
• Tree-ring records from at least three continents show suppressed growth beginning in 536 and continuing for over a decade.

The downstream cascade is, perhaps, the most consequential element of the evidence. The Plague of Justinian — the first documented pandemic of Yersinia pestis, the same bacterium responsible for the later Black Death — emerged in the Mediterranean basin in 541, just five years after the initial climatic disruption. Ancient genomic studies of plague-victim skeletal remains by Wagner et al., The Lancet Infectious Diseases (2014) and Keller et al., PNAS (2019) have now confirmed that the bacterium recovered from sixth-century burials across Europe is genetically continuous with the Black Death lineage centuries later.

The plague killed, by conservative estimates, between twenty-five and fifty million people in successive waves over the following two centuries. The combination of climatic disruption, plague mortality, and accompanying agricultural and demographic collapse coincides with — and almost certainly contributed to — the political reorganization of three continents: the weakening of the Sasanian Persian Empire, the failure of Justinian's western reconquest, the depopulation of the Roman successor states, and the demographic and economic conditions that preceded the rise of Islam in the following century.

The mainstream historiographical consensus has now shifted significantly toward treating these events as a connected cascade. Kyle Harper's The Fate of Rome (2017, Princeton University Press) presents the synthesis at book-length, drawing the climatic, epidemiological, and political evidence into a single argument about the end of antiquity.

Alternative perspectives

The volcanic explanation for AD 536 is now the dominant scientific consensus, and the evidentiary base supporting it is, by the standards of historical climatology, unusually strong. The Sigl, Toohey, and Büntgen analyses converge on a coherent picture: a sequence of one to three major volcanic eruptions between 536 and 547 produced sufficient stratospheric sulfate loading to suppress solar radiation for years, with downstream effects on agriculture, plague ecology, and political stability.

Several elements of the case, however, remain genuinely unresolved.

The source volcanoes have not been physically identified. The leading candidate for the 536 eruption is an Icelandic source, but no specific Icelandic volcano has been definitively associated with the geochemical signature in the ice cores. The 540 event is generally attributed to a tropical source, with Ilopango in El Salvador being one frequently-discussed candidate — but the dating of the Ilopango "Tierra Blanca Joven" eruption has been revised multiple times and the correlation with the ice-core signature remains contested.

The blue-sun observation in Cassiodorus is unusual. Volcanic dust veils typically produce reddened sunsets, since fine ash and sulfate aerosols scatter short-wavelength (blue) light preferentially. A blue sun is rare and is generally associated with a specific narrow particle-size distribution — roughly one micron — observed historically only during a small number of major forest-fire and volcanic events. The chemistry that produced a blue sun across multiple Mediterranean witnesses in 536 has not been satisfactorily modeled.

The global simultaneity of the disruption, including documented effects in equatorial latitudes and the Southern Hemisphere, remains slightly uncomfortable for a model based primarily on high-latitude Northern Hemisphere volcanism. A tropical eruption can explain inter-hemispheric distribution, but the timing of the proposed tropical event in 540 postdates the initial onset of the 536 disruption.

A minority of researchers have raised the possibility of an extraterrestrial component — either a contemporaneous comet or fragmentary impact event, or sustained influx of cometary dust from a denser-than-usual passage through a debris stream. This hypothesis has the advantage of accounting for the blue-sun observation (cometary dust falls in the relevant size range) and the global simultaneity, but it has the substantial disadvantage that no physical evidence of a major impact event in this period has been identified.

The mainstream position is that the volcanic explanation is sufficient and that the unresolved details will be clarified by further high-resolution ice-core and chronological work. The minority position is that the unresolved details are meaningful and that the case should remain genuinely open.

Open questions

Several specific questions would, if answered, substantially settle the remaining debates:

• Which Icelandic volcano (if any) erupted in 536? Tephrochronology — the identification of volcanic ash layers by their geochemical fingerprint — has the technical capacity to answer this if a matching tephra layer can be identified in a peat bog or ice core stratigraphy and traced back to a specific Icelandic source.
• Is the proposed 540 tropical eruption Ilopango, or something else? Refined argon-argon dating and geochemical fingerprinting of candidate tropical eruptions could resolve this.
• Was there a cosmic component? A targeted search for impact-related proxies — iridium, platinum, nanodiamonds, microspherules — in precisely-dated 536–540 sediment layers would constrain the question. To date no such systematic study has been published.
• Why a blue sun? The atmospheric chemistry that would produce a sustained blue-shifted solar appearance from volcanic aerosols, rather than the expected reddening, has not been adequately modeled.

The broader question — whether the AD 536–660 Late Antique Little Ice Age should be understood as a primarily geophysical event with social consequences, or as a connected geophysical-epidemiological-political cascade that genuinely terminated antiquity — remains an active area of research.

Conclusion

The most striking feature of the AD 536 case is not the catastrophe itself. It is the fifteen-hundred-year delay between the recording of the event by multiple independent contemporary witnesses and the moment mainstream science finally confirmed they had been telling the truth.

Procopius said the sun dimmed. Cassiodorus said it was blue. The Chinese said dust fell like snow. The Irish said the bread failed. They wrote it down. The records survived. And until tree rings and ice cores forced the issue, the records were treated as exaggeration.

The question worth keeping in mind is what contemporary records — what patterns currently being observed and recorded by witnesses we may be inclined to dismiss — will turn out, fifteen hundred years from now, to have been describing reality with embarrassing accuracy all along.

The chroniclers of 536 were not wrong.

They were merely fifteen centuries early.