The Eclipse of 1133 — Death of Henry I and Saxon Chronicle

About The Eclipse of 1133 — Death of Henry I and Saxon Chronicle

On 2 August 1133, at approximately 14:00 local apparent time, the path of totality of a solar eclipse crossed northern England and southern Scotland, sweeping from the Atlantic across the Hebrides and the Scottish lowlands before continuing into northern Europe and the Levant. Modern calculation (NASA/Espenak, Five Millennium Canon of Solar Eclipses) places the duration at greatest eclipse at four minutes and thirty-eight seconds, with an eclipse magnitude of 1.0652 and the event listed as number 43 in Solar Saros series 102. The afternoon went dark across England; candles were lit; the disc of the sun, in the words of one chronicler, appeared as a moon three nights old, with stars visible around it at midday. King Henry I crossed from Portsmouth to Normandy on the previous day, the Feast of Lammas (1 August 1133). He never returned to England alive, dying twenty-eight months later, on 1 December 1135, at the castle of Lyons-la-Forêt in eastern Normandy. The chroniclers of twelfth-century England — the Peterborough continuator of the Anglo-Saxon Chronicle, John of Worcester, William of Malmesbury, Henry of Huntingdon — collapsed the gap. The eclipse, the king's crossing, the king's death, and the civil war that followed (the Anarchy of 1135–1154) were woven into a single sign.

The 1133 eclipse is the earliest English event for which the astronomical record, the textual record, and the political record can be cross-checked at fine resolution, and the cross-check reveals something more interesting than a simple omen. The Peterborough manuscript of the Anglo-Saxon Chronicle, written at Peterborough Abbey by a scribe who appears to have composed the continuation in or shortly after 1154, files the eclipse under the year 1135 — the year of Henry's death — not under 1133, the year it occurred. The slide of the event upward in the annals by two calendar years was not a calculation error in any astronomical sense. The scribe had no astronomical computation on which to base the date; he had a memory, a death, and a need to make the sky and the throne speak together. Eclipse and king were already, by 1154, a unit. Astronomy is recoverable from these texts only by working against the chroniclers' own framing.

The umbral track of the 1133 August 2 eclipse approached from the western Atlantic, made landfall across the southern Hebrides, and crossed the British Isles diagonally on a southeast vector. Maximum totality at greatest eclipse — 4 minutes 38 seconds — fell over the open ocean east of Newfoundland; the British leg of the path saw shorter totalities of approximately three to four minutes. The central line ran across the southern Outer Hebrides, the Firth of Lorne, the Scottish lowlands south of the Highland Boundary Fault, and exited the British coastline in the vicinity of the modern English-Scottish border, before continuing across the North Sea, northern Germany, central Europe, and ending near the Persian Gulf. Locations south of this path — including the Anglo-Saxon Chronicle scriptoria at Peterborough, Worcester, and Malmesbury — experienced a deep partial eclipse rather than totality proper.

Espenak's Five Millennium Canon places greatest eclipse at approximately 12:09 UTC, which corresponds to about 14:02 local mean time in central England (the modern timezone offset is not strictly applicable to medieval reckoning, but the noon-to-mid-afternoon window is approximately right). The Peterborough Chronicle's report that the darkness fell when men were "eating their meat" — the principal midday meal in twelfth-century English monastic and lay practice was taken between the canonical hours of Sext (noon) and Nones (mid-afternoon) — independently confirms the timing. The convergence of the chronicler's domestic detail with NASA's modern ephemeris is one of the cleanest such matches in the medieval English record.

The eclipse belongs to Saros 102, a long-running series of eclipses spaced by 6,585.32 days (the saros period). The series began in 792 CE and ended in 2090 CE, producing 73 eclipses across roughly 1,298 years. The 1133 eclipse sat near the middle of the series, when the geometry produces deep total eclipses on long durations. The previous Saros 102 eclipse was on 23 July 1115 (visible across central Asia); the next, 13 August 1151, would be visible across the southern Indian Ocean. The English chroniclers had no concept of the saros — the periodicity was Babylonian, transmitted to medieval Latin Europe only through translations of Arabic astronomical works in the twelfth and thirteenth centuries — and so the eclipse appeared, to them, as a singular event rather than as the 43rd recurrence in an 18-year cycle.

The Peterborough manuscript (Bodleian MS Laud Misc. 636, the "E" recension of the Anglo-Saxon Chronicle and the only manuscript to continue past the Norman Conquest, terminating in 1154) entered the eclipse under the annal for 1135 in a passage now standardly rendered: "In this year went the King Henry over sea at the Lammas; and the next day, as he lay asleep on ship, the day darkened over all lands, and the sun was all as it were a three-night-old moon, and the stars about him at midday." The translation is from Garmonsway's 1953 edition (The Anglo-Saxon Chronicle, J.M. Dent, London) and is followed in Swanton's 1996 revision (The Anglo-Saxon Chronicles, J.M. Dent, London).

The passage is a textual problem before it is a historical one. The internal evidence — "the next day" after the king's Lammas crossing — fixes the eclipse to 2 August, which matches the 1133 astronomical event exactly and matches no astronomical event in 1135. The Peterborough scribe, writing in or after 1154 (a continuation produced by a single hand covering 1132–1154 according to the analysis of Cecily Clark in her 1958 edition for Oxford University Press, The Peterborough Chronicle 1070–1154), conflated the eclipse with the king's death because the two events had, by the time of writing, become a unit in monastic memory. The scribe was working at a distance of approximately twenty-one years from the eclipse and nineteen years from the death; the conflation is a memory effect, not a fabrication.

The Peterborough entry is part of a sequence of marvels in the same annal — earthquakes, atmospheric darkness, the king's death, and the political collapse that followed — presented as a continuous chain of signs. The chronicler's prose, increasingly close to early Middle English in this final continuation, is among the most-studied passages in the history of the English language: it documents the linguistic transition out of late Old English under the pressure of Anglo-Norman administrative and ecclesiastical Latin. The 1135 annal is one of the great set pieces of that transition, the eclipse passage embedded within it.

John of Worcester, a Benedictine monk and chronicler at the cathedral priory of Worcester, recorded the same eclipse in his Chronicon ex Chronicis under the correct year — 1133 — and with a notable precision of astronomical vocabulary. The entry, preserved in Oxford, Corpus Christi College MS 157 (an autograph or near-autograph copy of the chronicle, completed in stages between approximately 1124 and 1140), reports that "in 1133 a darkness appeared in the sky throughout England. In some places it was only a little dark; in others, candles were needed. The sun looked like a new moon, though its shape constantly changed. Some said that this was an eclipse of the sun. If so, then the sun was at the Head of the Dragon and the moon at its Tail, or vice versa." The translation is paraphrased from R.R. Darlington and P. McGurk's three-volume Oxford Medieval Texts edition (1995–1998), The Chronicle of John of Worcester.

The "Head of the Dragon" and "Tail of the Dragon" are technical terms — caput draconis and cauda draconis in medieval Latin astronomy — for the ascending and descending nodes of the lunar orbit, the two points where the Moon's path crosses the ecliptic. Solar eclipses are possible only when a new moon occurs near one of these nodes. The vocabulary is a calque on the Arabic al-jawzahr (from Persian gawzahr, "node-snake"), and arrived in medieval Latin Europe through the translations of Arabic astronomical works undertaken at Toledo and elsewhere in the twelfth century. John of Worcester's casual deployment of the terminology in 1133 indicates that the Arabic-Latin transmission had reached the Worcester cathedral library within decades of its arrival in Christian Spain, and that the monastic computists of western England possessed a working theoretical model of eclipse causation — not merely a record of darkness.

John's report distinguishes itself from the Peterborough entry by retaining the technical hedging. "Some said that this was an eclipse of the sun." The phrasing leaves room for an alternative — atmospheric darkness, divine intervention, a non-eclipse cause — and inserts the eclipse interpretation as one position among possible readings. The chronicler is doing what twelfth-century English monastic intellectuals were beginning to do across the disciplines: separating mechanism from meaning, recording the natural cause and the providential reading on parallel tracks. The eclipse can be both an astronomical event at the lunar node and a sign concerning the king. The two readings are not mutually exclusive in John's framework.

Henry I, fourth and youngest son of William the Conqueror, born in or around 1068 and king of England since the death of his elder brother William II in a hunting accident on 2 August 1100, had ruled England for thirty-three years by the time of the 1133 eclipse. His reign had been consolidative and administrative: the establishment of the Exchequer, the imposition of strong royal justice (the epithet Leo Iustitiae, Lion of Justice, attached to him in subsequent chronicle tradition), and the eventual reunification of England and Normandy under a single ruler after his victory over his elder brother Robert Curthose at the Battle of Tinchebrai in 1106. By 1133 he was the most powerful Anglo-Norman ruler of his line.

The succession was the dynastic disaster of his reign. His only legitimate son, William Adelin, drowned in the wreck of the Blanche-Nef (the White Ship) on the night of 25 November 1120, an event Orderic Vitalis treats at length in Book XII of his Historia Ecclesiastica (completed by 1141). Henry's second marriage to Adeliza of Louvain in 1121 produced no further legitimate children. By 1126 Henry had designated his daughter Matilda — widow of the Emperor Henry V, married in 1128 to Geoffrey of Anjou — as his heir, and had compelled the English and Norman magnates to swear oaths of fealty to her. The eclipse of 1133 occurred during what was, by Anglo-Norman dynastic standards, a precarious succession arrangement: an aging king, no legitimate son, a daughter pledged to an Angevin foreigner, and a circle of cousins (including Stephen of Blois, the king's nephew) with credible counter-claims.

Henry crossed to Normandy on 1 August 1133. He spent the remaining twenty-eight months of his life on the continent, attempting to stabilize Norman politics and to secure the loyalty of the Anglo-Norman baronage to Matilda's succession. He fell ill in late November 1135 at the castle of Lyons-la-Forêt, in the forest of Lyons east of Rouen, after a hunting expedition. Henry of Huntingdon's Historia Anglorum (Greenway's 1996 Oxford Medieval Texts edition) records that the king dined on lampreys against the advice of his physician and worsened over the following week. He died on the night of 1 December 1135, aged approximately sixty-seven. Modern paleopathological analysis (cited in the 2023 review by Liu et al., Clinical Infectious Diseases, and the 2023 Cambridge volume Forbidden Fish) raises listeriosis as a plausible alternative diagnosis given the cold, damp November conditions at the castle and the king's advanced age, but the historical record gives only the surfeit of lampreys.

The body was carried to Rouen, embalmed, and eventually transported to England, where it was interred at Reading Abbey — the foundation Henry himself had endowed in 1121. The crown did not pass to Matilda. Within three weeks of Henry's death, his nephew Stephen of Blois had crossed to England, secured the royal treasury at Winchester, and was crowned at Westminster on 22 December 1135. The Anarchy — nineteen years of intermittent civil war between Stephen's and Matilda's factions — followed.

The omen reading was inherited from two parallel traditions and reinforced in the chronicle scriptoria of the twelfth century. The first tradition was patristic and biblical: the Gospel of Matthew (27:45), the Gospel of Mark (15:33), and the Gospel of Luke (23:44) all record a darkness over the land at the crucifixion of Christ, traditionally read in the Western Church as a portent of cosmic significance. Eusebius of Caesarea, in his Chronicon (early fourth century, preserved in Jerome's Latin translation), associated the crucifixion darkness with a historical eclipse — a claim later defended by Humphreys and Waddington in Nature (vol. 306, 1983, pp. 743–746) on the grounds of a 3 April 33 CE lunar (not solar) eclipse, a proposal that has been received skeptically by most New Testament historians and most astronomers, given that the gospel timing falls during Passover full moon (precluding a solar eclipse) and the Humphreys-Waddington lunar eclipse would have been only marginally visible from Jerusalem.

The second tradition was classical-historical. Herodotus, in Histories I.74, records that an eclipse predicted by Thales of Miletus interrupted a battle between the Medes and the Lydians and led to peace — the eclipse traditionally identified with the event of 28 May 585 BCE. The Herodotean precedent, transmitted to medieval Latin Europe through encyclopedic compilations and through the works of Macrobius and Isidore of Seville (Etymologiae, completed c. 625 CE), supplied the model of eclipse-as-political-marker. The conjunction of the patristic darkness-at-crucifixion and the Herodotean eclipse-as-political-omen produced, in medieval Latin Christendom, an interpretive frame in which any unusual celestial event could be read against the ruling dynasty.

The Venerable Bede's De Temporum Ratione (725 CE), the foundational textbook of medieval Western computus, supplied the technical apparatus. Chapter 27 (De magnitudine vel defectu solis et lunae) sets out the geometry of solar and lunar eclipses — the sun obscured by the intervening Moon, the Moon by the interposed Earth — in terms drawn from Pliny the Elder's Natural History (Book II) and from Macrobius. The treatise was the standard reference for monastic computists for four centuries; John of Worcester, Henry of Huntingdon, and the Peterborough scribe had all been trained on it. Bede's framing was naturalistic in its physical mechanics and providential in its temporal framework: eclipses were predictable celestial events occurring within a divinely ordered time. The two readings ran in parallel, and the chroniclers of 1133 inherited both at once.

The recovery of the correct 1133 dating from the Peterborough Chronicle's 1135 annal is a methodological case study in the use of astronomical computation to discipline textual history. The pioneering modern treatment is Robert R. Newton's Medieval Chronicles and the Rotation of the Earth (Johns Hopkins University Press, 1972), an 848-page assembly of medieval eclipse records collated against retrocalculated astronomical events. Newton's project was not primarily historical: he was attempting to extract observational data from chronicles in order to constrain models of long-term changes in Earth's rotation rate (the parameter ΔT, the difference between Terrestrial Time and Universal Time, which has accumulated to approximately +69 seconds over the past millennium due to tidal braking of Earth's spin). The 1133 eclipse is a useful Newton-style data point precisely because the chroniclers anchored it to identifiable local events (the king's crossing, the lighting of candles at midday, the timing of meals) that constrain when and where totality was observed.

The chronological disciplining runs in both directions. Astronomical retrocalculation tells the historian that no solar eclipse over England occurred in 1135; the Peterborough Chronicle's "1135 eclipse" must therefore be a misdated 1133 event. The Chronicle, in turn, tells the astronomer something about ΔT: if the umbral path reached southern England rather than passing further north, then the assumed Earth rotation rate is incorrect by a quantifiable amount. Newton, and subsequent investigators including F. Richard Stephenson (Historical Eclipses and Earth's Rotation, Cambridge University Press, 1997), have used hundreds of such cross-checked records to map ΔT across two millennia. The 1133 eclipse is a contributing record in this much larger reconstruction.

Modern paleographical analysis of the Peterborough manuscript itself — particularly the work of Cecily Clark and, more recently, Susan Irvine in her 2004 edition of MS E (The Anglo-Saxon Chronicle: A Collaborative Edition, Volume 7: MS E, D.S. Brewer) — confirms that the final continuation was composed in a single hand in the 1150s. The eclipse was therefore being recorded in or around 1154, twenty-one years after the event, by a scribe who had personal recollection or close hearsay access to it. The "memory window" between event and chronicle is short enough to preserve the perceptual detail (candles at midday, the moon-shaped sun, the visible stars) while being long enough to slip the date by two years.

The 1133 eclipse was the most spectacular but not the only celestial or meteorological anomaly recorded in English chronicles for that year. William of Malmesbury, in the Historia Novella (the chronicle of contemporary events he began around 1140, edited by Edmund King and translated by K.R. Potter in the Oxford Medieval Texts edition of 1998), recorded an earthquake that struck England in the same year, providing one of the earliest eyewitness earthquake reports in the English documentary record: he wrote that "the wall of the house in which I was sitting was lifted up by two shocks, and settled again with a third." The British Geological Survey's historical seismicity catalogue (Musson, BGS Report MUSS008, 2008) accepts the 1133 event as a real seismic occurrence in East Anglia, with damage including ringing church bells.

The chronicle entries clustered these signs. The earthquake, the eclipse, the king's departure to Normandy, and (retrospectively, after 1135) the king's death and the descent into civil war were grouped in the chronicles into a single proleptic chain. Henry of Huntingdon, writing his successive recensions of the Historia Anglorum through the 1130s, 1140s, and 1150s, increasingly arranged the events of 1133–1135 as the prelude to the disorder of Stephen's reign. The chronicler's hindsight is the structuring principle: events that, prospectively in 1133, had no necessary connection were retrospectively, by 1154, woven together as the demonstration of providential pattern.

The cluster effect is itself a documentable phenomenon in medieval chronicle production. The Worcester scriptorium's earlier scribal generation — John of Worcester's predecessor Florence (whose chronicle John extended) — had similarly bundled the eclipse of 1110 with that year's political upheavals. The eclipse of 5 April 1131 had been bundled with the death of Bishop Robert of Lincoln. The eclipse of 26 January 1153 would be bundled with the negotiations leading to the Treaty of Wallingford. The bundling is a stable pattern in twelfth-century English chronicle method, and the 1133 eclipse/1135 death pairing fits the pattern exactly.

The arc from the omen reading of 1133 to the modern astronomical treatment of eclipses concludes, in one familiar telling, with the eclipse of 29 May 1919. The Greenwich Royal Observatory expedition led by Frank Watson Dyson, with observing teams headed by Arthur Eddington (Príncipe, Gulf of Guinea) and Andrew Crommelin and Charles Davidson (Sobral, Brazil), measured the apparent positions of stars near the obscured solar disc during totality, comparing them to their positions away from the sun. The displacement matched the prediction of Einstein's 1915 general theory of relativity (1.75 arcseconds at the solar limb) and ruled out the Newtonian prediction of half that value. Dyson, Eddington, and Davidson's paper — "A Determination of the Deflection of Light by the Sun's Gravitational Field, from Observations Made at the Total Eclipse of May 29, 1919," Philosophical Transactions of the Royal Society A, vol. 220 (1920), pp. 291–333 — became the canonical demonstration that eclipses had been recruited from omen to physics.

The 21 August 2017 total solar eclipse, whose path of totality crossed the contiguous United States from Oregon to South Carolina with a maximum totality of 2 minutes 40 seconds at the central point near Hopkinsville, Kentucky, is the closest modern analogue to the 1133 English event in cultural visibility — a continent-wide darkening of the daytime sky observed by approximately 215 million people. The 2017 eclipse was predicted to the second decades in advance, observed by tens of millions, and produced no political reading in any major news cycle. The 1133 eclipse, equally precise in its astronomical parameters and only marginally shorter in totality, was read by twelfth-century English monastic intellectuals as the sign of a king's coming death — twenty-eight months in advance.

The cultural distance between 1133 and 2017 is the recoverable substance of the 1133 record. The astronomy is identical; the interpretive frame is not. The chroniclers of Peterborough, Worcester, Malmesbury, and Huntingdon were not reading the eclipse incorrectly in any astronomical sense — John of Worcester's reference to the lunar nodes is technically accurate — but they were reading it within a frame in which celestial mechanics and political fate were intelligible only together. The frame is what changed.

The gap between the eclipse of 2 August 1133 and the death of Henry I on 1 December 1135 is twenty-eight months, not the "immediate" or "shortly after" framing that the Peterborough Chronicle's filing of the eclipse under 1135 suggests, and not the "two years later" approximation that some modern popularizations have repeated. The chronological precision is important. The eclipse was not a same-day, same-week, or same-month omen; it was a sign read retrospectively, two and a third years after the fact, as the chronicler reached for an opening event to frame the king's death.

The retrospective construction is the historically interesting fact. Twelfth-century English chroniclers were not predicting the king's death at the time of the eclipse — there is no contemporary record from 1133 that interprets the eclipse as a portent of imminent royal demise. The interpretation appears only after 1135, in entries written down at varying distances from the event (John of Worcester's near-contemporary entry; the Peterborough Chronicle's continuation written around 1154; Henry of Huntingdon's successive recensions). What looks, in retrospect, like a tight omen-and-fulfillment couplet is a twenty-eight-month gap collapsed by chronicle method. The omen reading is a historical artifact of the chronicler's hindsight, not a contemporary perception of 1133.

The honest description of the 1133 eclipse is therefore double: it was an astronomical event of measurable parameters (Saros 102 #43, 2 August 1133, 4m 38s totality, magnitude 1.0652), observed widely across northern Europe and the Middle East, and it was — separately, and only after the fact — recruited as the founding sign of the Anarchy by chroniclers writing between 1135 and 1154. The two facts do not contradict; they describe distinct phenomena. The astronomical record belongs to the history of celestial mechanics; the omen record belongs to the history of twelfth-century English political theology. The 1133 eclipse is consequential in both registers, and the registers must be kept distinct to read either of them accurately.

Precision

Date / time / duration verified against NASA Espenak Five Millennium Canon (Solar Saros 102, event 43; greatest duration 4m 38s; magnitude 1.0652)

Modern Verification

Espenak & Meeus 2006 NASA TP-2006-214141; Newton 1972; Stephenson 1997

Significance

The 1133 eclipse is one of the cleanest cross-checked datapoints in the medieval English documentary record: an astronomical event whose date, duration, magnitude, and path are independently recoverable from modern retrocalculation (NASA/Espenak's Five Millennium Canon, Solar Saros 102 #43, 4m 38s greatest duration, magnitude 1.0652), and whose perception is independently recorded by at least four twelfth-century English chroniclers — the Peterborough manuscript of the Anglo-Saxon Chronicle, John of Worcester's Chronicon ex Chronicis, William of Malmesbury's Historia Novella (which records the contemporaneous earthquake), and Henry of Huntingdon's Historia Anglorum. The astronomical and the textual records can be put in dialogue at fine resolution.

The dialogue reveals something more interesting than a confirmed omen. The Peterborough Chronicle files the eclipse under 1135 — the year of Henry I's death at Lyons-la-Forêt — not under 1133, the year the eclipse occurred. The twenty-eight month gap between portent and fulfilment was collapsed by a Peterborough scribe writing in or around 1154, twenty-one years after the eclipse and nineteen years after the king's death. The chronicle's omen reading is therefore not a contemporary 1133 perception but a retrospective construction, produced when the eclipse and the death had already become a unit in monastic memory. This is itself a historically substantive finding: it documents how twelfth-century English chronicle method produced proleptic signs by collapsing chronology in hindsight, rather than by recording prospective omens.

The eclipse is also a transmission marker. John of Worcester's reference to the "Head of the Dragon" (caput draconis) and "Tail of the Dragon" (cauda draconis) — the ascending and descending nodes of the lunar orbit, calqued from Arabic al-jawzahr — places the technical apparatus of Arabic eclipse theory inside an English cathedral scriptorium within decades of its arrival in Christian Spain via the Toledo translation movement. The Worcester chronicle thus dates the speed of the twelfth-century Arabic-to-Latin scientific transmission with unusual precision: by 1133, the Worcester cathedral library possessed at least the vocabulary, and probably some of the underlying ephemeris-based theory, of Arabic astronomical computation.

For the political history of England, the eclipse stands at the threshold of the Anarchy (1135–1154), the first English civil war and the most prolonged dynastic crisis of the Anglo-Norman period. Henry I's death without a recognized male heir, his daughter Matilda's contested claim, and his nephew Stephen of Blois's coronation on 22 December 1135 together produced nineteen years of intermittent warfare resolved only by the Treaty of Wallingford (1153) and Henry II's accession (1154). The chroniclers' decision to make the 1133 eclipse the opening sign of this nineteen-year crisis is itself a piece of the historiographical infrastructure of the Anarchy: the eclipse becomes the literary device that opens the civil war's narrative arc.

Connections

The 1133 eclipse can be situated within the broader corpus of historical eclipses recorded across civilizations and within the medieval English engagement with astronomical knowledge. For the comparative tradition of solar eclipse omenology, see Venus across civilizations, which traces the parallel role of Venus heliacal phenomena as political markers in Babylonian, Mesoamerican, and Greco-Roman astrology, and Orion across cultures, which examines a different mode of stellar political signification.

The mechanics of eclipse prediction in medieval Europe drew on cycles developed centuries earlier. The precession of the equinoxes documents one of these long-period cycles; the saros cycle — the 18-year, 11-day, 8-hour period to which the 1133 eclipse (Saros 102 #43) belongs — was a Babylonian discovery transmitted to medieval Europe through Arabic intermediaries. For the broader context of heliacal rising and other periodic stellar phenomena used in calendrical and political prediction, see the dedicated page.

Mesoamerican eclipse-tracking offers a productive comparative case: the Dresden Codex contains an eclipse table covering 405 lunations and approximately thirty-three years, demonstrating that the Maya tracked eclipses with predictive intent — a level of computational sophistication the twelfth-century English chroniclers did not match. The Maya Long Count and tzolk'in 260-day cycle provide the calendrical scaffolding within which Maya eclipse prediction operated.

For other politically interpreted solar events, see the Chichen Itza equinox phenomenon and the comparable solstice alignments at Stonehenge and the Giza pyramid complex. These sites engage with the predictable rather than the anomalous behavior of the sun, providing the structural counterpoint to the eclipse omen tradition that the 1133 event represents.

Further Reading

• Clark, Cecily (ed.). The Peterborough Chronicle 1070–1154. Oxford: Clarendon Press, 1958 (2nd ed. 1970). The standard scholarly edition of the Peterborough continuation, with detailed analysis of the two scribal hands and the chronology of composition. The eclipse passage and its 1135 dating are treated in the textual commentary.

• Irvine, Susan (ed.). The Anglo-Saxon Chronicle: A Collaborative Edition, Volume 7: MS E. Cambridge: D.S. Brewer, 2004. The current authoritative edition of the Peterborough manuscript within the multi-volume collaborative edition. Contains the Old English text with full critical apparatus.

• Swanton, Michael (trans.). The Anglo-Saxon Chronicles. London: J.M. Dent, 1996. The most widely cited modern English translation, presenting parallel translations of all surviving manuscripts. Useful for the comparison of the Peterborough 1135 entry with the absence of equivalent entries in the other (earlier-terminating) manuscripts.

• Darlington, R.R., McGurk, P., and Bray, J. (eds.). The Chronicle of John of Worcester. 3 vols. Oxford: Clarendon Press, 1995–1998 (Oxford Medieval Texts). The standard modern edition of the Worcester chronicle, with Latin text, English translation, and extensive commentary on John's astronomical knowledge and his use of Arabic-derived terminology.

• Greenway, Diana (ed. and trans.). Henry, Archdeacon of Huntingdon: Historia Anglorum. Oxford: Clarendon Press, 1996 (Oxford Medieval Texts). The standard edition of Henry of Huntingdon's chronicle, including his account of Henry I's death and the surfeit-of-lampreys tradition.

• King, Edmund (ed.) and Potter, K.R. (trans.). William of Malmesbury: Historia Novella. Oxford: Clarendon Press, 1998 (Oxford Medieval Texts). The contemporary chronicle of the succession crisis and the descent into the Anarchy, including William's eyewitness account of the 1133 earthquake.

• Newton, Robert R. Medieval Chronicles and the Rotation of the Earth. Baltimore: Johns Hopkins University Press, 1972. The pioneering modern treatment of medieval European eclipse records as constraints on Earth-rotation parameters; the 1133 eclipse is among the records analysed.

• Stephenson, F. Richard. Historical Eclipses and Earth's Rotation. Cambridge: Cambridge University Press, 1997. The current standard reference on the use of historical eclipse records to reconstruct long-term changes in Earth's rotation rate, updating Newton's 1972 study.

• Hollister, C. Warren. Henry I. New Haven: Yale University Press, 2001 (Yale English Monarchs). The standard modern political biography, with full treatment of the 1133 Normandy departure and the king's death.

• King, Edmund. King Stephen. New Haven: Yale University Press, 2010. The companion volume on Stephen's reign and the Anarchy — the political context within which the 1133 eclipse was retrospectively read as an omen.

• Espenak, Fred and Meeus, Jean. Five Millennium Canon of Solar Eclipses. NASA TP–2006–214141, 2006. The standard reference catalogue of all solar eclipses across five millennia, supplying the astronomical parameters for the 2 August 1133 event.