← Case page: The Princes in the TowerHistorical MysteriesFull report · 14 September 2026Researched and written by Claude, an AI system, from the sources each case cites. Claude also drew the maps and charts marked "made for this report" from the data named in their captions.

Princes in the Tower

Did the Princes in the Tower Survive?

On 21 June 1483 a servant of the Lord Chancellor wrote to a friend that the nine-year-old Duke of York was in the Tower with his brother, “blessid be Jhesus, mery”. No one recorded a dated sighting of either boy again. In 2023 a research project led by the writer Philippa Langley said four documents show that they lived. Here are those documents, and the sources for the murder, in their own words.

The mystery

In the spring of 1483 England had a boy king. Edward IV, the first king of the house of York, died on 9 April, leaving two sons: Edward, Prince of Wales, who was twelve and became King Edward V, and Richard, Duke of York, who was nine. The boys’ uncle Richard, Duke of Gloucester, the dead king’s younger brother, was named Protector, to govern for the boy king. On 30 April, at Stony Stratford on the road to London, he took charge of the young king from the escort provided by the boy’s mother’s family and brought him to the capital, where he was lodged in the Tower of London, then a royal residence as well as a fortress and a prison, and the usual place for a king to wait for his coronation.

The boys’ mother, Queen Elizabeth Woodville, took her other children into sanctuary at Westminster Abbey, where by law no one could lay hands on them. On 16 June, with armed men filling Westminster, the younger boy was handed over to the Cardinal Archbishop of Canterbury and taken to join his brother in the Tower. Five days later Simon Stallworth, a servant of the Lord Chancellor, wrote to an Oxfordshire knight, Sir William Stonor, that the Duke of York was in the Tower with his brother, “blessid be Jhesus, mery”. In the same week Lord Hastings, the late king’s chamberlain, had been beheaded on 13 June; on 22 June the boys were publicly declared illegitimate, on the ground that their father’s marriage to their mother had not been valid, which put them out of the succession; and on 26 June Gloucester accepted the crown. He was crowned Richard III on 6 July.

From that point the boys were seen less often. Dominic Mancini, an Italian visitor who was in London until July and wrote up what he had seen for a French patron before the year was out, says that after Hastings’ death the boys’ attendants were dismissed, that the brothers were moved into the inner apartments of the Tower, and that they “began to be seen more rarely behind the bars and windows, till at length they ceased to appear altogether”. The young king’s physician, John Argentine, told Mancini that the boy went to confession every day because he believed he was going to die. A London chronicle records the boys shooting and playing in the Tower garden at some point in the mayoral year that ended in October 1483. No one recorded a dated sighting of either boy again. In the south of England men began to plot to free them. By October the plotters had heard that the boys were dead, and rose instead for Henry Tudor, an exile in Brittany with a distant claim to the throne, who swore on Christmas Day 1483 to marry the boys’ eldest sister, Elizabeth of York. In January 1484 the Chancellor of France told the Estates General, the assembly of the French realm, meeting at Tours, that Edward IV’s children had been murdered and the crown given to their murderer. Richard III never answered the charge and never showed the boys, and in March 1484 he gave their mother a public oath for the safety of her daughters that did not mention her sons. He was killed at Bosworth in August 1485 by Henry Tudor’s army. Henry became Henry VII and married Elizabeth of York.

Why it is a mystery

Nothing after the summer of 1483 is certain. No one who wrote at the time claims to know whether the boys died, or when, or how. Richard III never said. Henry VII’s first parliament condemned Richard by act of attainder, a parliamentary sentence, for “shedding of Infants blood”, but named no victims; Henry never produced bodies, held no inquest, and never published the confession that Sir James Tyrell, a knight who had served Richard III, is supposed to have made before his execution in May 1502. No text of any confession survives. The first full account of a murder was written about thirty years after the event by Thomas More, the lawyer who later became Henry VIII’s Lord Chancellor. It names Tyrell as the organiser and two men, Miles Forest and John Dighton, as the killers; it says the bodies were buried at the foot of a stair and then moved by a priest who died without telling anyone where; and it admits that “some remain yet in doubt” whether the boys died in Richard’s reign at all.

In 1674 workmen taking down a staircase in the White Tower, the central keep of the Tower of London, found the bones of two children about ten feet down. Charles II had them placed in an urn in Westminster Abbey. They were examined once, in 1933, by the Abbey’s archivist (its Keeper of the Muniments) and a professor of anatomy, who worked on the assumption that More’s story was true. They have never been dated, sexed or tested for DNA, and the Abbey has declined requests to test them in 1993, 1995, 2013 and 2018. There is therefore no body, no confession, no witness and no contemporary record of a death. The evidence consists of the disappearance and the conduct of those around the boys.

The two sides

Most historians of the period conclude that both boys died in the Tower in 1483, most probably on Richard’s orders, because that is what the conduct of everyone around them implies. This is the position of the standard academic studies of the reign, by Charles Ross (1981), Rosemary Horrox (1989), A. J. Pollard (1991), Michael Hicks (2003, and again in a 2024 article) and Sean Cunningham (2003), and of the recent biographies by David Horspool (2015) and Chris Skidmore (2017).

Against them stand Richard III’s defenders, known as Ricardians and organised since 1924 in the Richard III Society. They point out that no death is recorded, that the fullest accounts of a murder were written under the dynasty that replaced Richard, and that the two young men who claimed the Yorkist crown in Henry VII’s reign might have been the princes themselves. The first was a boy crowned “King Edward” at Christ Church, Dublin, on 24 May 1487, with an army of German mercenaries under Martin Schwartz sent by the boys’ aunt Margaret, Edward IV’s sister and the widowed Duchess of Burgundy. He was captured when that army was destroyed at Stoke, near Newark, on 16 June. Henry’s government said he was Lambert Simnel, an Oxford organ-maker’s son coached to pass as the boys’ cousin Edward, Earl of Warwick, the son of their uncle George, Duke of Clarence, whom Henry kept in the Tower. The second was a young man who appeared at Cork in 1491 as Richard, Duke of York, and was received under that name by Margaret in 1492, by Maximilian, the Habsburg King of the Romans (the elected German king) who governed the Low Countries, in 1493, and by James IV of Scotland in 1495. He was captured in 1497, confessed to being Piers Werbecque, called Perkin Warbeck, born at Tournai, a town in what is now Belgium, and was hanged in November 1499.

In 2023 Philippa Langley, the writer who led the search that found Richard III’s grave under a car park in Leicester in 2012, and her Missing Princes Project put four documents forward as evidence that both young men were who they said they were: a receipt in the accounts of Maximilian’s artillery, kept at Lille in northern France and dated December 1487, which calls the Dublin king “son of King Edward”; a first-person account of Richard of York’s escape from the Tower, in a manuscript of about 1500 in the provincial archive of Gelderland at Arnhem in the Netherlands; a pledge by “Richard of England” to Albert, Duke of Saxony, Maximilian’s commander in the Low Countries, dated at Dresden in October 1493; and a letter of Maximilian’s from the same winter naming three bodily marks by which the claimant could be known. Nobody on either side disputes that these documents are genuine. The dispute is over what the documents prove.

How the case was judged

Everything used here is in print or online. On the survival side: the four documents above and the records around them, that is the Irish grants and coins of 1487, the accounts of Margaret’s estates and towns, the chronicles of the Burgundian court, the claimant’s own letters of 1493 and 1497, the depositions about him taken at Setúbal, near Lisbon, in 1496, and the town records of Tournai. On the murder side: every source from Stallworth’s letter of June 1483 to More and the bones. Each document is quoted in its own language, Latin, French, Middle English, Irish or Dutch, with a translation, wherever an edition could be reached. Mancini’s Latin, the Dutch of the Arnhem manuscript and the Great Chronicle of London, a city chronicle written up about 1512, could not be, and are marked where they occur. Nothing was seen in manuscript.

One question was asked of each document. For the survival documents: does this show a prince alive, or only that a claimant, or someone paying for one, said so? For the murder sources: does the writer claim to know of a death, report a rumour, or draw an inference, and whose side was he on? Each carries a tag saying which. Modern scholarship is named as such, and partisanship, Ricardian or traditionalist (holding Richard guilty), is flagged where it affects reliability. The weighing at the end is written out as a ledger. The evidence is sorted into five clusters; each cluster is given a number for how likely that body of evidence would be if each of five outcomes were true; and the numbers are combined in the standard way for weighing evidence (Bayes’s rule) from a stated starting weight for each outcome. Each number is a judgment about the evidence, and another reader might set it differently. Their job is to make the weighing checkable and to show how far the verdict moves when any one of them changes. The script that computes them, ledger.py, and its output are published alongside this page with five research notes that record every text consulted.

The result is that both boys most probably died in the Tower in 1483, in the late summer or autumn, while in Richard III’s custody: about 85 chances in 100. The chance that one or both lived on, as the Dublin king, as the 1490s claimant or as someone unrecorded, comes out at about 6 in 100, and nearly all of that is an allowance for evidence not yet found. The rest, about 9 in 100, covers a death in 1484 or 1485 still in Richard’s keeping, a death under Henry VII, or a death abroad as children. The section headed “Weighing it” shows how far these figures move when any one judgment changes.

The reasons, in short, are these. The survival case rests on documents that record what a claimant and his backers said. Read in their own archives, they contain nothing a pretender’s court would not also have produced. The strongest of them, the Lille receipt of December 1487, is contradicted by every other contemporary record from Ireland, England and Burgundy that says whom the boy claimed to be (two later or ambiguous texts are discussed below). It is also contradicted by the dates on the boy’s own documents, an argument worked out by Randolph Jones, who has published on the Irish records in The Ricardian, the Richard III Society’s journal, and restated in 2024 by Chris Phillips, an independent researcher. The survival side has not answered it. The murder case is not proof either. No contemporary claims to know how the boys died, and no later narrative is more than report. But the boys vanished from view in the summer of 1483; London believed them dead by the autumn; their mother’s allies acted on that belief; and Richard, who could have ended the rumour by producing two living nephews, did not do so in the two years of his reign. His failure to produce the boys is the strongest evidence that they were dead.

Verdict

Figure 1. The three outcomes, weighed. Made for this report from the evidence ledger under “Weighing it”; the widths are its shares of 85, 6 and 9 per cent.

The evidence makes this verdict probable but does not prove it. It clearly favours death in 1483 over every other answer, and the percentage moves when the same evidence is weighed differently; the ledger below shows by how much. The case for murder is a case from how the people around the boys behaved. The verdict agrees with what the academic historians of the reign named above, Ross, Horrox, Pollard, Hicks and Cunningham, concluded, their books known here through summaries and Hicks’s 2024 note through its abstract; it changes nothing in their position. Every percentage on this page depends on the five outcomes and the weights set out in the ledger. Nothing here is a new document or a new fact. What is new is listed near the end, under “What is new here”.

What follows

The page begins with a short introduction to the witnesses whose names recur, and a dated table of events from April 1483 to 1933. Then comes the survival case, document by document. First the boy crowned at Dublin in 1487: the Lille receipt, every other record of who he was said to be, arranged by jurisdiction, and the test of the dates on his documents; then the conduct of Elizabeth Woodville and Henry VII. Next the claimant of 1491 to 1497: the Arnhem manuscript, his own letter of 1493 to Queen Isabella of Castile, the despatches of the courts that backed him, the records from Setúbal, Tournai and Exeter that place him, and the names his story shares with More’s. The murder case follows in the same way, source by source, from Stallworth’s letter of 21 June 1483 through Mancini, the anonymous Crowland chronicle of 1486, the French chancellor’s speech of January 1484, the rebels of that autumn, the short English and foreign notices, and the histories of Polydore Vergil and More, to the bones found in 1674. After that the evidence is weighed in the ledger, the case for survival is put at its strongest and answered, the evidence that would overturn the verdict is named, and what is new here is separated from what historians had already shown. The sources come last, with the research notes, the ledger script and its output.

The witnesses

A few sources recur in the tables and are cited there by short names. Mancini has been introduced. “Crowland” is the Crowland chronicle, an anonymous continuation of the history of Crowland Abbey in Lincolnshire, written in ten days in April 1486 by a man who describes himself as a doctor of canon law and a councillor of Edward IV; it is the best-placed English account of the reign. Polydore Vergil was an Italian scholar who wrote a history of England at Henry VII’s encouragement, drafted about 1512 and printed in 1534. The Great Chronicle of London and Robert Fabyan’s New Chronicles are London civic chronicles written up about 1512 and 1516. Titulus Regius is the act of Richard III’s parliament of 1484 that set out his title to the crown, and “Rot. Parl.” is the Rolls of Parliament, the official record, in the printing of 1767. Harleian MS 433 in the British Library is Richard III’s own register of grants and letters. Masselin’s Journal is the record kept by Jehan Masselin, one of the deputies, of the French Estates General at Tours in 1484. Morton’s register is the record of John Morton, Archbishop of Canterbury, and includes the proceedings of Convocation, the assembly of the clergy. The Heralds’ Memoir is an account of Henry VII’s early years kept by the royal officers of arms. Jean Molinet was the official historian of the Burgundian court. “Gairdner, Letters and Papers” is James Gairdner’s 1861 edition of the state letters of Richard III’s and Henry VII’s reigns; the Ormond Deeds are the charters of the Butler earls of Ormond in Ireland, edited by Edmund Curtis; “ADN Lille” is the departmental archive of the Nord at Lille; “CSP Spain”, “CSP Venice” and “CSP Milan” are the Calendars of State Papers, English summaries of the despatches of the Spanish, Venetian and Milanese envoys. Ann Wroe’s Perkin (2003) is the fullest modern study of the 1490s claimant. Tim Thornton is a historian whose articles of 2021 and 2024 trace More’s informants and the afterlife of Edward V’s belongings. “TNA PROB 11” is the register of wills at the National Archives. Tanner and Wright’s article of 1935 is the report of the 1933 examination of the bones, and P. W. Hammond’s of 1976, in The Ricardian, is a review of it.

Chronology

DateEventSource
9 Apr 1483Edward IV dies; Edward V, aged 12, is king.Crowland; Mancini
30 Apr 1483Gloucester takes custody of Edward V at Stony Stratford.Crowland; Mancini
16 Jun 1483Richard of York, aged 9, is taken from sanctuary at Westminster to join his brother in the Tower.Crowland; Mancini; Stallworth to Stonor, 21 June
22–26 Jun 1483The boys are declared bastards; Richard accepts the crown on 26 June and is crowned 6 July.Mancini; Crowland; Titulus Regius (1484)
Summer 1483Last sightings: the boys withdrawn to the inner Tower, servants dismissed, seen “more rarely” and then not at all; in the south, men begin to plot to free them.Mancini (by 1 Dec 1483); Great Chronicle of London; Crowland
Oct 1483Rising in the southern counties, first to free the boys, then, on the report that they were dead, for Henry Tudor.Crowland (Apr 1486)
1 Dec 1483Mancini finishes his account at Beaugency on the Loire: a suspicion that Edward V has been done away with, but no knowledge.Mancini
25 Dec 1483Henry Tudor swears at Rennes, in Brittany, to marry Elizabeth of York, with the backing of Elizabeth Woodville’s party.Vergil; Crowland
15 Jan 1484Chancellor Guillaume de Rochefort tells the Estates at Tours that Edward IV’s sons were murdered and the crown given to the murderer.Masselin’s Journal
1 Mar 1484Elizabeth Woodville leaves sanctuary under Richard’s public oath, which covers her daughters and says nothing of sons.BL Harleian MS 433
22 Aug 1485Bosworth. Henry VII’s first parliament attaints Richard for “shedding of Infants blood” (Nov 1485); Titulus Regius is repealed and Henry marries Elizabeth of York (Jan 1486).Rot. Parl. VI, 276, 289
Feb 1487The priest William Symonds confesses to taking an Oxford organ-maker’s son to Ireland, where he is taken for Warwick; the real Warwick is shown at St Paul’s.Morton’s register
24 May 1487A boy is crowned “King Edward” at Christ Church, Dublin. Stoke Field, 16 June: the Earl of Lincoln, the rising’s leader, killed, the boy captured.Rot. Parl. VI, 397; Heralds’ Memoir; Molinet
5 Jul 1487Henry VII asks the Pope to prosecute the Irish bishops for crowning a boy feigned to be Clarence’s son.Gairdner, Letters and Papers I, 94–96
13 Aug 1487The Earl of Kildare, governing Ireland, still issues letters patent (formal royal grants) in “King Edward’s” name, dated “the first year of our reign”.Ormond Deeds (Curtis III, 261–263)
16 Dec 1487The Lille receipt: 400 pikes for Schwartz’s men, sent to serve Margaret’s nephew, “son of King Edward”.ADN Lille B 3521/124564
1491–97A young man at Cork is put forward as Richard, Duke of York; received as Richard of York by Margaret of Burgundy (1492), Maximilian (1493), James IV (1495); captured 1497; confesses to Tournai birth.Warbeck dossier below
Apr 1496Depositions at Setúbal, Portugal, on the claimant’s origins, taken for Ferdinand and Isabella before Henry VII had him.Wroe (2003); CSP Spain I, no. 130
Nov 1499“Warbeck” hanged (23 Nov); Warwick beheaded (28 Nov).Great Chronicle
2–6 May 1502Sir James Tyrell tried at the Guildhall in London and executed; Henry VII is at the Tower 27 April–2 May. No confession text survives.Thornton 2024 (in the journal Historical Research)
c.1513–18Thomas More writes the History of King Richard III, naming Tyrell, Forest and Dighton.More
1516 / 1522Lady Margaret Capell, half-sister of Tyrell’s wife, bequeaths her husband’s chain “which was yonge kynge Edward the Vth”.TNA PROB 11/19
1674 / 1933Two children’s skeletons found under a White Tower stair; put in an urn in Westminster Abbey (1678); examined once (1933); never dated, sexed or DNA-tested.Tanner & Wright 1935; Hammond 1976

The survival case, document by document

Philippa Langley’s The Princes in the Tower: Solving History’s Greatest Cold Case (2023; revised 2025) and the Channel 4 television film of November 2023 rest on four named documents: a receipt for pikes at Lille (December 1487), a first-person narrative of Richard of York’s escape in the Gelderland provincial archive (Gelders Archief) at Arnhem, a pledge by “Richard of England” to Albert of Saxony at Dresden (October 1493), and a letter of Maximilian’s (November–December 1493) naming three bodily marks by which the claimant could be known. Around these sit Irish letters patent of 1487, a life of Henry VII written about 1500 by Bernard André, the blind French poet who served as the king’s court historian, and arguments from how Elizabeth Woodville and Henry VII behaved. Their authenticity is not in dispute. What follows asks what each of them proves.

The 1487 claimant and the Lille receipt

The Lille receiptBacker’s recordIn print since 1895

Archives départementales du Nord, Lille, B 3521 / 124564 · 16 December 1487 · French · artillery accounts of Maximilian, King of the Romans, who governed the Low Countries as regent for his young son. Abstracted, with the identifying phrase verbatim, by the archivist Jules Finot in the printed inventory of the Lille archives, Inventaire sommaire des Archives départementales … Nord, Série B, VIII (Lille, 1895), pp. 265–266 (read on archive.org); quoted for the pikes in Michael Bletzer’s 2013 study of the Habsburg captains at Bosworth and Stoke; presented as a May 2020 discovery in Langley (2023), pp. 176–177 and Appendix 2.

Jean de Finet, marchand fustailler à Malines, (fourniture, au prix de 120 livres, de 400 longues piques de bois de frêne, de 24 et 22 pieds de long, ferrées de bons fers à la nouvelle mode, acérés et émoulus, achetées par l’ordre du Roi au mois de juin 1487 et délivrées à Mgr. de Walhain, pour distribuer aux suisses allemands picquenars qu’était alors chargé de conduire par mer messire Martin de Zwarte, chevalier d’Allemagne, que Madame la douairière (Marguerite d’York) « envoya lors, avec aucuns capitaines gens de guerre d’Engleterre, pour servir son nepveu, filz du roy Édouart, jadiz son frère, que Dieu absoille, débouté de sa seignourie et obvier aux emprinses du Roy dudit Engleterre », 16 décembre 1487)

Jean de Finet, timber merchant of Mechelen: supply, at 120 livres, of 400 long ash pikes of 24 and 22 feet, shod with good heads in the new fashion, bought by the King’s order in June 1487 and delivered to the lord of Walhain to distribute to the German pikemen whom Sir Martin Schwartz was then charged to lead by sea, whom my lady the dowager “then sent, with certain captains and men of war of England, to serve her nephew, son of King Edward, formerly her brother, whom God absolve, ousted from his lordship, and to counter the enterprises of the King of the said England”, 16 December 1487. (Finot’s abstract; Langley’s translation of the full text reads the same at the decisive words.)

What it can bear. It shows that in December 1487 someone in Maximilian’s artillery office described Margaret of York’s protégé of that spring as a son of her brother King Edward. It names no Edward V, carries no regnal number and mentions no Ireland. Its own account of the pikes cannot stand as abstracted: bought “in June 1487” for a force that Henry VII already knew had landed in Ireland on 5 May (Henry VII’s letter to the Earl of Ormond, an Irish magnate loyal to him, printed in J. O. Halliwell’s Letters of the Kings of England, I, 171) and that was destroyed at Stoke, near Newark, on 16 June. The identifying words stand in the merchant’s receipt drafted by the secretary Florens Hauweel; the certificate signed by the two artillery officers attests the sale and delivery, not the description (Phillips 2024, pp. 13–14). It is evidence of what one clerk believed, not of the boy’s identity.

Every other record of who the boy was said to be

The receipt is the only contemporary text that unambiguously calls the Dublin king a son of Edward IV. The table collects the rest by jurisdiction, because the survival thesis explains the English evidence as Tudor propaganda: Ireland, where the Earl of Kildare, the Anglo-Irish magnate who governed the island as the king’s deputy and backed the boy, ruled in the boy’s name, and the Burgundian Netherlands, which paid for his army, were beyond Henry VII’s reach. Original words are given where the editions were checked. Dates in the form “1 Edward VI” mean the first year of that king’s reign; the point matters, and is taken up below the table.

RecordDateWhereWhom the boy was said to beWords
Confession of the priest William Symonds before Convocation, the assembly of the clergy (recorded in Archbishop Morton’s register)17 Feb 1487England, church courtAn Oxford organ-maker’s son, taken for Warwickfilium cujusdam [ ] orginmaker universitatis Oxon. … qui quidem filius pro comite Warwici ibidem reputabatur
Warwick exhibited at St Paul’s and before Convocationc.19–20 Feb 1487EnglandThe real Warwick, produced by the Earl of Derby (Thomas Stanley, Henry VII’s stepfather)Vergil; Morton’s register
York House Books (the city council’s minute books), heading to the rebels’ letter from Masham in Yorkshire8 Jun 1487England, civic“their king calling himself King Edward the sixth”in the name of ther king calling hymself King Edward the vjt (Attreed’s 1991 edition, II, 570; Raine’s 1941 edition misprinted “V”)
Heralds’ Memoir (kept by the royal officers of arms), account of Stoke1487England, officers of arms“King Edward”; real name John (manuscript) or Lambert (Hearne’s 1770 printing)the Lad that his Rebells callede King Edwarde, whos Name was indede …
Henry VII to Pope Innocent VIII, from Kenilworth Castle5 Jul 1487England to Rome (private diplomatic letter)A base-born boy feigned to be Clarence’s sonspurium quemdam puerum … confingentium puerum ipsum ducis quondam Clarentiae filium esse, in regem Angliae coronarunt
Innocent VIII’s mandate against the Irish bishops5 Jan 1488Rome“falsely alleging him to be a son of the late duke of Clarence”Calendar of Papal Letters XIV, 307
Attainder of the Earl of Lincoln (John de la Pole, Richard III’s nephew, who led the rising and was killed at Stoke)Nov 1487England, parliament“Lambert Symnell, a child of x yere of age, sonne to Thomas Symnell, late of Oxforde Joynoure”; no claimed identity statedRot. Parl. VI, 397
Notes of a London citizenc.1487–88England, private“saying he was the Duke of Clarence’s son, and was nat”ed. R. F. Green, English Historical Review 96 (1981)
Irish Exchequer memoranda roll, 2 Henry VII (destroyed in the 1922 fire at the Irish Public Record Office; known from two independent abstracts)May 1487Ireland, exchequerA writ dated “May, 1 Edward VI”Jones 2014
Letters patent (a formal royal grant) to Peter Butler as sheriff of Kilkenny, preserved among the Ormond Deeds13 Aug 1487Ireland, chancery (the royal writing office)“Edward … King of England, France and Ireland”, attested by Kildare as “our Lieutenant”, “in the first year of our reign”Curtis, Ormond Deeds III, 261–263
Irish coinage1487Ireland“Edwardus Rex Anglie Francie et Rex Hybernie”; no numberThe numismatists Carlyon-Britton (1941) and Dolley (1969)
Annals of Ulster (Gaelic annals kept in the north of Ireland)1487Ireland, Gaelic“the son of the Duke of York … proclaimed king … in Ath-cliath” (Dublin; ambiguous, see below)mic Diuíci Odhearc … rígh do gairm dhe … a m-Baile Atha Cliath
Domain accounts of Voorne (Margaret of York’s own estate in Holland)1487/88Burgundian NetherlandsMen to support “the Duke of Clarence”, her brother’s son, entitled to the thronePublished by Langley (2023), p. 424 n. 13
Accounts of the city of Mechelen (Malines), Margaret’s home town, for the procession of St Rumbold, the city’s patron saint18 Apr 1487Burgundian NetherlandsWine “to the son of Clarence from England”Redated by Langley’s team (2023), pp. 423–424
Jean Molinet, Chroniques, by the Burgundian court’s official historianc.1490Burgundian court“Edward, son of the Duke of Clarence”; the London cry “Vive Werwicq au roy Edouard!”Ce très noble rainceau est Edouard, fils du duc de Clarence
Adrien de But, a monk of the abbey of the Dunes in Flanders, writing to 14881486–88Flanders“the son of the Duke of Clarence … Duke of Warwick”; “the young Duke of Clarence”filium ducis Clarentiae … dux Verwecii; juvenis dux Clarentiae
The chronicle of Caspar Weinreich, a merchant of Danzig, a port of the Hanseatic trading league1487Prussia“George, the son of the Duke of Clarence”tr. Livia Visser-Fuchs, The Ricardian 7 (1986)
Bernard André, Vita Henrici Septimi (his Latin life of Henry VII)c.1500–02Tudor court, in retrospectA baker’s or cobbler’s boy put about as a son of Edward IV; “Edward’s second son” crowned in Irelandfilium Eduardi Quarti … Eduardi secundum filium in Hibernia regem coronatum fama retulit
The Lille receipt16 Dec 1487Habsburg artillery office“her nephew, son of King Edward, formerly her brother”son nepveu, filz du roy Édouart, jadiz son frère

Sources: Morton’s register as printed in A. F. Pollard’s source collection (a different Pollard from the historian A. J. Pollard cited above) The Reign of Henry VII from Contemporary Sources (1913–14), III, 246–249; the Heralds’ Memoir in Thomas Hearne’s Collectanea IV (1770), 209, 214 and in Emma Cavell’s edition (2009); Gairdner, Letters and Papers I, 95–96 (Latin read); Rotuli Parliamentorum VI, 397 (1767 printing read); Molinet in Buchon’s edition (1828), III, 152–156 (read); de But in Kervyn de Lettenhove’s edition (1870), I, 665–676 (read); André in Gairdner’s Memorials of King Henry VII (1858), 49–52 (Latin read); the Annals of Ulster in the CELT online text from University College Cork (read); Irish records via Randolph Jones and Chris Phillips. Full references are in the research notes published with this page.

The regnal year

The argument is Randolph Jones’s, worked out from the Irish records in three articles in The Ricardian (2014, 2021, 2023) and restated by Chris Phillips (2024, pp. 9–10). The printed records bear it out. Regnal years, the years of a king’s reign by which documents were dated, were counted in England and Ireland from accession, not coronation. Edward V’s first year ran from 9 April 1483 to 8 April 1484; his fifth would have run from 9 April 1487. A document issued on 13 August 1487 in Edward V’s name would be dated anno regni nostri quinto. The Butler patent (the grant of 13 August in the table) is dated in the first year, and the lost exchequer roll dated a writ of May 1487 to “1 Edward VI”. Langley’s answers are that the patent belongs to 1486, or that the count restarted at the Dublin coronation. The first is impossible: the Irish chancery was still issuing in Henry VII’s name on 1 May 1487 (Jones 2021, p. 49). The second has no precedent. Henry VI, the king of the rival house of Lancaster restored for six months in 1470, dated by “the forty-ninth year of our reign and the first of our readeption”; Edward IV, restored in 1471, continued his count; a regnal year that could be repeated would have been useless for dating (Jones 2023; Phillips 2024, pp. 9–10). Gordon Smith, the Richard III Society writer who first argued in 1996 that the Dublin king was Edward V, read the same patent and had to date it to 1486 to make it fit (Smith 1996, p. 516); that dating cannot stand.

The result is the same in all three jurisdictions. In England the boy was an organ-maker’s son taken for Warwick, and Henry VII said so in a private letter to the Pope whose purpose was to get four Irish bishops prosecuted, in a country where everyone knew whom they had crowned. In Ireland he was a new King Edward, in his first year, styling himself King rather than Lord of Ireland, the title English kings then used there. In the Burgundian Netherlands the duchess’s own estate paid for men “to support the Duke of Clarence, her brother’s son”, Mechelen gave him wine as “the son of Clarence”, and the court historiographer wrote him up as Edward, son of Clarence. Langley’s reading requires every one of these records to be mistaken and the artillery clerk alone to be correct.

André, the Annals, and the age of the boy

Two texts are usually cited beside the receipt. Bernard André, Henry VII’s blind court poet, writing around 1500 with blanks left where names should go, says the boy was put about as a son of Edward IV; but he says “Edward’s second son”, that is Richard of York, never Edward V, and his own Henry describes Margaret as one who “is not unaware that her blood was extinguished by her brother Richard” (sanguinem suum a Ricardo fratre suo extinctum non ignorat). Langley quotes the first phrase and omits the second (Phillips 2024, n. 39). The likeliest explanation is that André, thirteen years on, ran the 1487 boy together with the 1490s claimant, who did claim to be Richard of York and was likewise first recognised in Ireland and then taken up by Margaret. The Annals of Ulster call the Dublin king “the son of the Duke of York” and, under 1485, the one surviving “young man of the race of the blood royal”, exiled to Ireland the next year. Read literally the phrase fits no one: Richard, Duke of York (the boys’ grandfather, killed in 1460) had no son young enough, and Richard of Shrewsbury (the younger prince himself, named from his birthplace) had none; read as “grandson of the Duke of York” it fits Clarence’s son as well as Edward IV’s; and the same annals call Edward IV “king” under 1461 and 1503 (Phillips 2024, n. 47). The entry shows only that a Gaelic annalist far from Dublin described the boy loosely; it does not show that the boy was Edward V.

One point remains unresolved: the boy’s age. The attainder makes Lambert Simnel “a child of x yere of age”; Vergil makes him the same age as Warwick, twelve; Molinet, de But and the Annals describe a youth, and the person crowned and put at the head of an army was a youth. Ricardian writers infer that the crowned figure was older than the official ten-year-old, and therefore Edward V (sixteen). The inference does not follow. Ages in these texts are loose (Vergil misdates Warwick’s own age by five years), a twelve-year-old “Warwick” fits the Latin words used, puer (boy) and adolescens (youth), and if the government did understate the age of a substitute, that is a separate question from the identity the boy claimed, on which every record but one agrees.

Finding on 1487. The Dublin king claimed to be Edward, Earl of Warwick, or at any rate a new King Edward, not Edward V: about 95 per cent, as a judgment on the records above. The Lille receipt is best read as a clerk’s assumption that Margaret’s nephew was her brother the king’s son. Chris Phillips reached the same conclusion in 2024, in a note written with four members of the Dutch research group of Langley’s own project, and so did Michael Hicks in Historical Research the same year.

Elizabeth Woodville and Henry VII’s silence

Two arguments from conduct run through the survival literature. Elizabeth Woodville left sanctuary on 1 March 1484 under Richard’s public oath that her daughters would be kept “in surety of their lives” and married to gentlemen born (BL Harleian MS 433); a mother, it is said, would not have dealt with her sons’ murderer. Then on 12 February 1487, weeks before the Simnel plot surfaced, Henry VII moved her to Bermondsey Abbey, a religious house across the river from the Tower, and transferred her dower (the lands assigned to her as a widow) to her daughter, and her son by her first marriage, the Marquess of Dorset, was arrested; Vergil hints at punishment for her 1484 dealings with Richard, and Ricardians read it as punishment for backing a living son. Neither argument is strong. The 1484 oath is stronger evidence for the other side: it provides for daughters only, in March 1484, and a queen with living sons in the king’s custody would have asked for them. Her retirement in 1487 coincides with Henry’s suspicion of the whole Woodville connection at the moment a Yorkist rising was gathering, which would follow whether the figurehead was Warwick or anyone else; and in November 1487 Henry was still considering her as a bride for James III of Scotland, which he would not have done had he believed her to be plotting with a surviving rival king.

Henry VII’s first parliament attainted Richard for “shedding of Infants blood” (Rot. Parl. VI, 276) but named no victims; he produced no bodies, held no inquest, published no confession in 1502, and the first full account is More’s. The silence is consistent with a king who believed the boys dead but could not prove it, and Henry did say what he believed: to Sir Gilbert Talbot, one of his captains, in July 1493 the new claimant was a “feigned” boy who had tried three identities in turn, and André’s Henry treats Edward IV’s sons as extinguished. Under the hypothesis that Henry himself had them killed after 1485, the objection made in 1983 by Jeremy Potter, a Ricardian writer, stands: he would have produced the corpses and a story blaming Richard. The silence therefore counts a little against survival and a little for the outcome in which nobody under Henry knew where the bodies lay.

The claimant of 1491–97 and the Gelderland manuscript

The second claimant appeared at Cork in the autumn of 1491, was entertained by Charles VIII, King of France, in 1492, by Margaret of York from late 1492, by Maximilian from 1493 and by James IV from 1495, landed in Cornwall in September 1497, surrendered at Beaulieu Abbey in Hampshire, confessed, and was hanged at Tyburn, London’s gallows, on 23 November 1499. The survival thesis holds that he was Richard of York. Its documents are the Arnhem narrative, the claimant’s own letters and charters, and the acts of the courts that received him. Against them stand the records of three jurisdictions that had no stake in the answer.

The Gelderland narrativeClaimant’s narrativeFound 1951, re-found 2020

Gelders Archief, Arnhem, 0510 (Diverse charters / diverse aanwinsten), inv. 1549, catalogued as “unidentified author, story about Richard of York, circa 1500, with transcript, 1500s” · Middle Dutch, a single copy in one hand; third-person heading, first-person text · provenance unknown (a miscellaneous-acquisitions series) · found by the archivist A. P. van Schilfgaarde in 1951 and sent by his colleague P. J. Meij to D. Th. Enklaar, professor of medieval history at Utrecht; re-found in the archive’s online catalogue by Nathalie Nijman-Bliekendaal, a volunteer researcher with Langley’s project, in November 2020 · English translation in Langley (2023; revised 2025 with corrections by the historian Bart Lambert); no edition of the Dutch text has been published. Read here in Langley’s translation as quoted by the archive, by the Dutch broadcasters NOS and Omroep Gelderland, and by reviewers, and in the Dutch phrases quoted by Meij.

Het verhaal dat ik voor me heb vermeldt echter dat Richard uit de Tower weggevoerd is onder leiding van ‘Hinrijck und Thomas Parey’, te Parijs verblijft, later via Brabant naar Middelburg gaat, naar Lissabon, naar Ierland, Frankrijk en tenslotte bij de ‘herthoginne van Burghoendiën’ belandt.

The story I have before me states that Richard was taken away from the Tower under the guidance of “Hinrijck und Thomas Parey”, stays in Paris, later goes by way of Brabant to Middelburg, to Lisbon, to Ireland, France, and finally ends up with the “duchess of Burgundy”. P. J. Meij to D. Th. Enklaar, 1951, in the archive file; Omroep Gelderland, 21 Nov 2023.

Ik kan natuurlijk niet oordelen over uw archiefstuk zonder het gezien te hebben, maar ik vraag me af, of het een propagandastuk is om stemming te maken tegen Hendrik VII Tudor … en het hem lastig te maken door een valse Richard van York naar voren te schuiven.

I cannot of course judge your document without having seen it, but I wonder whether it is a propaganda piece to stir up feeling against Henry VII Tudor … and to make things difficult for him by putting forward a false Richard of York. Enklaar to Meij, 1951, same file.

“… we were delivered to Brackenbury, and then to sir James Tyrrell, and then to the Duke of Buckingham, by whose orders we were separated. And so they ordered my lord Foriest, Hamelett Maleven and Wylliam Puche by whom I was secretly taken into a room in a place where the lions are kept.” Lord Howard (John Howard, soon Duke of Norfolk) then comes to him, dismisses the guards, and brings “Henry and Thomas Parcij”, who swear to hide him “until a number of years had passed”; they shave his head, dress him in a drab shirt, take him to St Katharine’s, the precinct just east of the Tower, and by boat to Boulogne; then Paris “a long time”, Chartres, Rouen, Dieppe, Hainault, Brabant, Malines, Antwerp, Bergen, Middelburg; to Lisbon with the wife of Sir Edward Brampton, a Portuguese-born Yorkist knight, aged about fourteen; Henry Parcij dies of plague; Ireland, where the earls of Kildare and Desmond, the two great Anglo-Irish magnates, recognise him; Charles VIII, who promises help and withdraws after his treaty with Henry; and his aunt, who “recognised my rights and honesty”. Langley (2023) tr., as quoted by the historian Sylvia Barbara Soberton (Nov 2025), by NOS (Nov 2023) and by a Ricardian blog, sparkypus (Mar 2024).

What it can bear. It shows that a story of Richard of York’s escape circulated in Dutch around 1500, in a copy nobody can trace, with no author, date, seal or witness. What the archive’s staff and Andrew Dunning, curator of medieval manuscripts at the Bodleian Library in Oxford, have authenticated is the materials and the hand, that is, that the copy belongs to Warbeck’s own era, which no one disputes. The text’s outline (handed to a lord to be killed, spared on oath, sent abroad with two guardians, years in hiding, Portugal, Ireland, Charles VIII, the aunt) is that of the claimant’s Latin letter of 25 August 1493, below. That it tracks the claimant’s letter and the 1497 confession in outline has been said on both sides of the debate (Nathen Amin, a historian of the Tudor pretenders: it “adds nothing” to the claimant’s 1493 letter, dated at Dendermonde, below; the Richard III Society’s research pages, which read the match as corroboration; Hicks 2024, for whom the text is “propaganda as necessary for the imposter Perkin Warbeck as for the real prince”). One narrower point: its extra stations, Antwerp, Bergen op Zoom, Middelburg and the voyage to Lisbon in the company of Brampton’s wife, are, in the same order, the itinerary of Piers Werbecque of Tournai as deposed at Setúbal in April 1496 and confessed in 1497, and none of them is in the 1493 letter. What is new in it is a list of Tower officers and household men for 1483: Sir Robert Brackenbury, the Constable of the Tower; Tyrell; Buckingham; Miles Forest; Mauleverer; John Howard, soon Duke of Norfolk; Norris; Tyrwhitt. These fit the summer of 1483, and any of the Yorkist exiles who joined the claimant in 1492 could have supplied them. Enklaar in 1951 asked whether it was a propaganda piece for a false Richard; the archive still catalogues it as a “story” by an unidentified author; Hicks (2024) reads it as a document that either hypothesis needs, since a pretender required exactly such a text. It is a claimant’s narrative, the most detailed we have, and nothing in it is checkable except the names, which were public.
The claimant’s letter to Isabella of CastileClaimant’s narrativeIn print since 1838

British Library, Egerton MS 616, f. 3 · Latin · dated “Ex opido Andermunda … Octauo kallendas Septembris” 1493, that is Dendermonde, 25 August 1493 (the Calendar of State Papers, Spain, edited by Bergenroth in 1862, misread the date as 8 September) · printed in full by Sir Frederic Madden of the British Museum, “Documents relating to Perkin Warbeck”, Archaeologia 27 (1838), pp. 199–200; calendared CSP Spain I, no. 85.

Cum primogenitus quondam Regis Anglie Edwardi pie memorie, domini et parentis mei carissimi, Princeps Walie, miserando funere extinctus fuisset, egoque eciam nonum fere agens annum traditus essem cuidam domino extinguendus, placuit divine clementie ut ille dominus innocentie mee misertus me viuum et incolumem seruauerit, cum tamen primo supra sacrum corpus domini jurejurando me adegerit ne nomen, genus et familiam usque in annos certos cuippiam propalarem. Misit ergo me peregre cum duobus viris qui me custodirent ac gubernarent. … et fere annis octo per diversas provincias latitaui. Tandem altero eorum qui me gubernabat defuncto, altero in patriam misso et nunquam postea viso, vix puericiam egressus, solus utque egens, mansi ad tempus in Regno Portugallie; inde in Hyberniam nauigaui.

When the first-born of the late King Edward of England of pious memory, my dearest lord and father, the Prince of Wales, had been put to death by a pitiable end, and I too, going on nine years old, had been handed over to a certain lord to be killed, it pleased the divine mercy that that lord, taking pity on my innocence, kept me alive and unharmed, having first made me swear upon the Lord’s sacred body that I would reveal my name, birth and family to nobody until a fixed term of years. So he sent me abroad with two men to guard and govern me … and for almost eight years I lay hidden in various provinces. At last, one of my governors having died and the other having been sent home and never seen again, scarcely out of boyhood, alone and in want, I stayed for a time in the kingdom of Portugal; from there I sailed to Ireland.

What it can bear. This is the story in the claimant’s own name, six weeks before the Dresden pledge and before the Gelderland text, whose narrative ends at Margaret’s court and whose copy the archive dates about 1500. Two things follow. Richard of York was born on 17 August 1473 (his christening and his 1478 marriage are dated by it) and was ten when the boys vanished; “nonum fere agens annum” is a year short. James Gairdner, the Victorian historian of Richard III’s reign, saw this in 1898 (p. 268 n.): “Warbeck, however, supposed the character he was personating to have been not quite nine years old in 1483; which, we may presume, nearly tallied with his own age,” adding the Venetian envoy Trevisan’s report that he was 23 in November 1497 (CSP Venice I, no. 760), so born about 1474. The observation is Gairdner’s. Set beside it, the Gelderland text makes the same boy “about fourteen” at Lisbon. The letter also explains the Gelderland text: a claimant had to tell his story to courts, and the Arnhem copy is a fuller version of the same story into which the real Werbecque’s travels have been added. Note that the letter treats Edward V as dead by violence and says nothing about how.
The backers’ records, 1493–96Backers’ records

Dresden, Sächsisches Hauptstaatsarchiv: pledge by “Richard of England” to Albert, Duke of Saxony, for 30,000 florins, 4 October 1493, with monogram and seal, co-signed by two English supporters, Sir Robert Clifford and William Barley (found by Langley’s team, February 2019) · Maximilian to Henry VII, 27 November–12 December 1493, stating that Richard can be known by “several signs” including three marks that “cannot be counterfeited” (Austrian archives; found by Zoë Maula of Langley’s project, December 2020; read here only in Langley’s summary) · the Papal Penitentiary (the Roman office that granted exemptions from church law), indult (a grant of privilege) of 21 May 1494 to “the nobleman Richard, duke of York” (Clarke and Zutshi, Supplications, III, p. 127) · Nuremberg, Staatsarchiv, Ansbach foreign charters nos. 248–51 and 257: charters and letters of “Richard of England” to two German families, the Wolkensteins and Ludwig Wolfgang Tetzel, 1494–96, one sealed with the Prince of Wales’s ostrich feathers (found by Nijman-Bliekendaal, 2024) · West-Brabants Archief, Bergen op Zoom stewards’ accounts 1494/5, fol. 189v, expenses for the Duke of Saxony and the Duke of York, 6 January 1495 (the genealogist Yvette Hoitink, 2023) · Maximilian’s dealings as reported by the Venetian and Milanese envoys, CSP Venice I, nos. 641–665, CSP Milan I, no. 471.

The King of the Romans “rejoiced greatly, as he could dispose of this Duke of York ad libitum suum” (the Venetian envoys Contarini and Trevisano, Worms, 17 July 1495, no. 649). “Should he succeed in his attack on England, Maximilian would admit him … whilst on the other hand, should the Duke be worsted, the present King will be accepted in his stead” (16 August 1495, no. 652). To Lord Egremont, an English exile at his court, in a formal justification: the Duke of York, “whom he firmly believes to be the son of King Edward, came to him; and … he considered it his duty not to abandon the Duke” (6 January 1496, no. 665). Ferdinand and Isabella, on Margaret’s letter to them: “They had sent no answer to the pretended Duke of York, but only to the Duchess, showing her that the whole affair was an imposture. The Duchess made no reply” (to Puebla, their ambassador in London, 20 July 1495, CSP Spain I, no. 99). Cardinal Morton, Henry VII’s chancellor, to the Milanese envoy: Margaret “has at times written to the Cardinal recommending Perkin as the son of King Edward … The Cardinal replied: But indeed he is not reputed the son of King Edward in this kingdom” (Soncino, 30 September 1497, CSP Milan I, no. 545).

What it can bear. They show that from October 1493 the claimant was styled, sealed, lent money and entertained as Richard of York by Albert of Saxony, Maximilian, the papal Penitentiary, the Wolkensteins and Tetzel, and the town of Bergen op Zoom. Every one of these records is the claimant’s own act or a backer’s; none records a test of identity, and the “three marks” letter records that the marks were asserted, not that anyone who had known the prince checked them. His statements of belief (to Henry VII in 1493, to Egremont in 1496) are diplomatic documents; six months before the second, his ambassadors had reported him as saying that the boy could be disposed of at will and that if he lost, Henry would do. A pretender’s backers would have produced the same records as a real prince’s; the records therefore cannot decide the question.
The records that identify him: Setúbal, Tournai, ExeterIndependent records

Setúbal, April 1496: depositions of Sir Edward Brampton (Duarte Brandão), Rui de Sousa (Portuguese ambassador to England in 1489) and the herald Tanjar, taken at Spanish instance; printed by Ann Wroe, Perkin (2003), pp. 525–528, read here in summary · Ferdinand and Isabella to Puebla, 14 April 1496, CSP Spain I, no. 130 · Tournai, registers of burgesses and of the town court, 1460–76, printed by Gairdner, Richard the Third (1898), appendix, pp. 334–335 · the claimant’s letter to his mother, Exeter, 13 October 1497, French, Gairdner pp. 330–331 (from two Tournai copies) · the confession, as copied into a London chronicle (BL Cotton Vitellius A XVI), ed. Kingsford, Chronicles of London (1905), pp. 219–221 · Soncino to the Duke of Milan, CSP Milan I, no. 548; Trevisan, CSP Venice I, nos. 759–760.

“With respect to what you write, that the King of France has sent a paper, with the seal of his council, and a declaration from the King-at-arms of Portugal, stating that he of York is the son of a barber, and offering to send over his father and mother, &c., we have to observe, that if the King of England wish something of the kind we can do it much better than the King of France. We can send him the declarations of many persons who know him, amongst whom is a Portuguese knight, of the name of Ruy de Sosa. … Having been Portuguese ambassador in England, he knows the Duke of York very well, and has seen him there. Two years later he saw this other person in Portugal.” Ferdinand and Isabella to Puebla, 14 April 1496.

1474. 11 May. Jehan de Werbecque, pireman, fils de feu Diericq de Werbecque, a relevé sa bourgeoisie comme fil de bourgois né en bourgoisie, et en a fait le serment en tel cas introduit.  ·  1459 (1460 n.s.). 9 Feb. Pierart Flan, pireman, fils de feu Pierart, natif de Condé.

1474, 11 May. Jehan de Werbecque, boatman, son of the late Diericq de Werbecque, renewed his burgess-right as a burgess’s son born in burgess-right and took the customary oath.  ·  1460, 9 February. Pierart Flan, boatman, son of the late Pierart, native of Condé. Tournai registers, in Gairdner (1898), pp. 334–335.

Et afin que entendez et cognoissiez clerement que sui vostre filz et non aultre, il vous plaira souvenir, quand je parti de vous avec Berlo pour aller en Anvers, vous me deistes adieu en plorant, et mon pere me convoia jusque a la porte de Marvis … Et me recommandez humblement a mon parin, Pierart Flan, a Maistre Jean Stalin, mon oncle … Escrips a Excestre, le xiije jour de Octobre de la main de vostre humble filz, Pierrequin Werbecque.

And so that you may understand and know clearly that I am your son and no other, please remember, when I left you with Berlo to go to Antwerp, you said goodbye to me weeping, and my father accompanied me as far as the Marvis gate … And commend me humbly to my godfather, Pierart Flan, to Master Jean Stalin, my uncle … Written at Exeter, the 13th day of October, by the hand of your humble son, Pierrequin Werbecque. Gairdner (1898), pp. 330–331; Pollard, Reign of Henry VII, I, no. 119.

“ffirst it is to be knowen that I was born in the Towne of Turney, and my ffaders name is called John Osbek; which said John Osbek was controller of the Towne of Turney. And my moders name is Kateryn de ffaro. And one of my Grauntsires vpon my ffaders side was called Deryk Osbek, which died …” Confession, Kingsford p. 219.   The King to the prisoner: “We have heard that you call yourself Richard, son of King Edward. In this place are some who were companions of that lord, look and see if you recognise them. The young man answered that he knew none of them.” Soncino, CSP Milan I, no. 548, reporting the herald who was present.

What it can bear. The claimant was identified as Piers Werbecque of Tournai by three bodies that owed Henry VII nothing: Portuguese witnesses examined at Spanish instance in April 1496, eighteen months before the capture; Charles VIII’s council, which sent Henry a sealed paper with the Portuguese king-at-arms’ declaration that he was “the son of a barber” and offered to send over his father and mother (the trade differs from the boatman of the registers; the town and the living parents do not); and the Tournai registers, which show a boatman Jehan de Werbecque son of Diericq admitted burgess in 1474 and a boatman Pierart Flan in 1460, the “Deryk Osbek” and “Peter Flame” of the confession and the “parin Pierart Flan” of the Exeter letter. The confession was written under Henry’s control and is worth little on its own; the letter to his mother is worth more, because it recalls things (the Marvis gate, a sister dead of plague at the procession, a swineherd-girl) that only a Tournai boy could use to prove himself to a Tournai mother, and it survives in Tournai copies. Once captured he was shown men who had been companions of Richard of York and knew none of them, and none is recorded as knowing him. Sir William Stanley, Henry VII’s chamberlain, executed in 1495 for words in his favour, is reported by Vergil to have said he would not fight the young man si certum haberet Edouardi filium esse, if he knew for certain he was Edward’s son: even his patron had not been shown proof. In 1498 the prisoner swore before Puebla and the Bishop of Cambrai that Margaret “knew as well as himself that he was not the son of King Edward”, a statement made in the King’s presence and weighed accordingly.

Tyrell and Forest

Michael Hicks has called the Arnhem narrative “the earliest record connecting Richard, duke of York, with Sir James Tyrell and Miles Forest”, and Tim Thornton cites it in the same sense (in the journal History, 109, 2024, n. 50); the observation is theirs. The narrative names Sir James Tyrell and “my lord Foriest” among the men who had custody of the boys. More’s History, written about thirty years after the events and some fifteen years after the Arnhem copy, names Tyrell as the commissioner and Miles Forest as one of the two killers, and Tim Thornton has shown (History 106, 2021) that Forest’s sons Edward and Miles were at Henry VIII’s court in More’s own circle in the 1510s, so that More had a plausible route to the family. The two texts, one from the claimant’s camp and one from the Tudor court, agree on who kept the boys and disagree on what became of them. What the overlap most naturally shows is a shared body of Yorkist knowledge, current among the exiles of the 1490s, about the men in charge of the Tower in the late summer of 1483. That Forest was one of the custodians is probably true. The overlap does not show whether Forest then helped to remove Richard, as the Arnhem text says, or helped to kill him, as More says.

ElementLetter to Isabella, Aug 1493Setúbal, Apr 1496 and confession, 1497Gelderland text, c. 1500
Elder brother killed; narrator spared by a lord under oath; sent abroad with two guardians; years hiddenYesNoYes (Howard; the two Parcijs)
Paris and northern France“diversas provincias”NoYes
Antwerp, Bergen op Zoom, MiddelburgNoYes (the Antwerp fair with the merchant Berlo; “Barowe marte”; service with the merchant John Strewe at Middelburg)Yes, in that order
To Portugal in the company of Brampton’s wifeNo (“mansi ad tempus in Regno Portugallie”)Yes (Brampton and Rui de Sousa 1496; confession)Yes
Age at Lisbon“vix puericiam egressus”Confession implies about 13–14 in 1487–88“about fourteen”
Ireland; Kildare and DesmondYesYes (Cork, 1491)Yes
Tower custodians named“cuidam domino”Brackenbury, Tyrell, Buckingham, Forest, Mauleverer, Howard, Norris, Tyrwhitt

Finding on the claimant of 1491–97. On the ledger’s weights the chance that Richard of York survived in any form is about 4 in 100 (2 alone, 2 with his brother), nearly all of it the general allowance for evidence not yet found; that he survived as this claimant is a part of that figure. Every document that treats him as York is his own, a backer’s, or a routine grace issued on his own petition. The three independent bodies that examined the question before his capture (the Portuguese witnesses at Setúbal, the Spanish sovereigns through Rui de Sousa, the French council) identified him as a commoner of Tournai. The Gelderland narrative, the fullest of his stories, has absorbed that boy’s itinerary, and the story in his own name misstates the prince’s age. What would change this: a record from 1483–91 of Richard of York’s whereabouts made by someone not later paid to say so, or physical evidence, which the Tournai and English family lines might one day supply.

The murder case, source by source

No writer claims to have seen the boys die, and no body has ever been shown to be theirs. The case for murder therefore rests on how everyone behaved in 1483–84 and on a sequence of reports, each of which has to be asked whether it records a rumour, an inference or knowledge, and whose interest the writer served. The sources are taken in the order they were written, which is not the order of the events. Original words are quoted where an edition was checked; Mancini’s Latin could not be consulted and is given in the English of his editor, C. A. J. Armstrong.

SourceWrittenClassWhat it can bear
Simon Stallworth to Sir William Stonor9 and 21 June 1483Contemporary factBoth boys alive in the Tower on 21 June; York “mery”; “every manne dowtes other”
Memorandum of George Cely, a London wool merchantc. 13–26 June 1483FearThe king’s death among the feared contingencies in London that week; contains false items
Colchester Oath Book (the town’s register), heading for 1483–84c. 29 Sept 1483AmbiguousEdward V “nuper filii” of Edward IV: deposed or dead, the word does not say
Mancini, De occupationeBy 1 Dec 1483 (events to mid-July)Rumour, marked as suchWithdrawal from sight after 13 June; a suspicion of death in London; the writer does not know
Caspar Weinreich’s Danzig chronicleUnder 1483; date of writing uncertainMerchant newsReport in the Hanseatic port of Danzig that Richard “had his brother’s children killed” that summer; paired with a false item
Rochefort at Tours15 Jan 1484Asserted as fact; French crownBoth deaths current as fact at the French court by January 1484; no source stated
Crowland continuationTen days ending 30 Apr 1486Rumour and inference; Yorkist official under Henry VIIBoys under guard in late summer 1483; plotters aimed to free them; a rumour of violent death “was spread”; manner still unknown; rebels turned to Richmond (Henry Tudor)
Attainder of Richard IIINov 1485Tudor government“shedding of Infants blood”, no victim named, no date
Diego de Valera, a Castilian chronicler1 Mar 1486; chronicle 1487Merchant report, “se afirma”The story as it reached Andalusia, already garbled
John Rous, a Warwickshire priest, Historia Regum Angliae1486–91Asserted; Tudor dedicationKilled “within about three months or a little more”; manner “known to very few”; writer far from London
Robert Ricart, town clerk of Bristol, KalendarYear 1483–84; marginal, undatableCommon report“put to scylence in the Towre”
Notes of a London citizen (College of Arms MS 2M6)1483–88 (R. F. Green’s dating)AssertedPut to death “be the vise of the Duke of Buckingham”; earliest English text to name an agent (text not seen here)
Philippe de Commynes, a French royal councillor, Mémoires1489–98Asserted; French courtLouis XI of France held Richard a child-killer before 30 Aug 1483 (if the anecdote holds); Richard in Books I and V, Buckingham in Book VI
Molinet, Chroniques1490sRumour catalogue; Burgundian court“About five weeks” prisoners; chest or quilts; names wrong; Buckingham suspected
Vergil, Anglica Historiac. 1512–13; printed 1534Asserted; Tudor official historyTyrell on Richard’s order during the royal progress (Richard’s tour of the Midlands and north, August–September 1483); manner “non plane liquet”; no confession, no grave
Great Chronicle of London; Fabyanc. 1512; 1516Common fameSeen “shotyng and playyng” in the Tower garden in the mayoral year to Oct 1483; “neuer came abrode after”; three rumoured methods
More, History of King Richard the Thirdc. 1513–18; printed 1557Asserted, from unnamed informants; Morton milieuForest and Dighton, the feather bed, the stair foot, the priest, the 1502 confessions; “some remain yet in doubt”
Divisiekroniek, a chronicle of HollandPrinted 1517Three rumours; pro-Warbeck milieuGloucester smothered them; or Buckingham killed them; or Buckingham killed one and Richard escaped as the White Rose (the Yorkist claimant)
The bones under the stairFound 17 July 1674; examined 1933Physical, unidentifiedTwo children, sifted from rubble, identified from More’s story; never sexed, dated or tested
June 1483: the last certain sightingContemporary fact

Simon Stallworth (servant of the Chancellor, Bishop Russell) to Sir William Stonor, London, 21 June 1483 · The Stonor Letters and Papers, ed. Kingsford, II (Camden 3rd ser. 30, 1919), no. 331, pp. 160–161 · George Cely’s undated memorandum, The Cely Papers, ed. Malden (1900), no. 114, pp. 132–133, between 13 and 26 June 1483.

Worschipfull Sir, I commend me to you, and for tydynges I hold you happy that ye ar oute of the prese, for with huse is myche trobull, and every manne dowtes other. As on Fryday last was the lord Chamberleyn hedded sone apone noon. On Monday last was at Westm. gret plenty of harnest men: ther was the dylyveraunce of the Dewke of Yorke to my lord Cardenale, my lord Chaunceler, and other many lordes Temporale … and so departed with my lord Cardenale to the toure, wher he is, blessid be Jhesus, mery.

Ther ys grett romber in the reme … yff the kyng God saffe his lyfe wher desett, the Dewk of Glosetter wher in any parell, geffe my lorde prynsse wher God defend wher trobellett …

There is great rumour in the realm … if the king, God save his life, were deceased; if the Duke of Gloucester were in any peril; if my lord prince, which God defend, were troubled … Cely, the same week. The note also reports the Bishop of Ely dead, which was false.

What it can bear. Both boys were alive and together in the Tower on 21 June 1483, five days after Richard of York left sanctuary; Hastings had been beheaded on the 13th; London was frightened. Cely’s jotting shows that the king’s death was already among the things people feared, not that it had happened. These are the last dated reports of the boys’ condition by contemporaries who were not later writing history.
Dominic ManciniRumour, marked as such

De occupatione regni Anglie per Riccardum tercium, Lille, Bibliothèque municipale · Latin · completed at Beaugency, 1 December 1483, for Angelo Cato, archbishop of Vienne and councillor of the French court; Mancini left London about mid-July · ed. and tr. C. A. J. Armstrong, The Usurpation of Richard the Third (1936; 2nd ed. 1969), c. pp. 92–93; new translation by Annette Carson, a Ricardian writer (2021). Latin not consulted here.

“But after Hastings was removed, all the attendants who had waited on the king were debarred access to him. He and his brother were withdrawn into the inner apartments of the Tower proper, and day by day began to be seen more rarely behind the bars and windows, till at length they ceased to appear altogether.” … “The physician Argentine, the last of his attendants whose services the king enjoyed, reported that the young king, like a victim prepared for sacrifice, sought remission of his sins by daily confession and penance, because he believed that death was facing him.” … “I have seen many men burst forth into tears and lamentations when mention was made of him after his removal from men’s sight; and already there was a suspicion that he had been done away with. Whether, however, he has been done away with, and by what manner of death, so far I have not at all discovered.” Armstrong’s translation.

What it can bear. It was written within five months of the events by a foreigner who had been in London, for a court hostile to Richard, with no Tudor interest to serve. It establishes the withdrawal of the boys from sight after 13 June, the dismissal of their servants, the boy king’s own expectation of death as reported by his physician, and the existence in London by July of a suspicion (suspicio) that Edward V was dead. Mancini says in terms that he does not know whether he was. Carson argues that the verb he uses (tollere) means “removed” rather than “killed”; on either rendering the report is of a suspicion, not of a death, and it says nothing separately about Richard of York.
The Crowland continuationRumourInference

Anonymous; the writer describes himself as a doctor of canon law and royal councillor under Edward IV · Latin · colophon (the closing note): Acta sunt haec & expleta apud Croylandiam, Anno Domini 1486, per spatium decem dierum quorum postremus fuit ultimus dies mensis Aprilis, so written in ten days ending 30 April 1486 · first printed by William Fulman, Rerum Anglicarum Scriptorum Veterum, I (Oxford, 1684), pp. 567–568, 578 (read on archive.org); H. T. Riley’s translation (1854), pp. 490–491; the modern edition by Pronay and Cox (1986), pp. 162–163.

Interim & dum haec agerentur, remanserunt duo praedicti Regis Edwardi filii sub certa deputata custodia infra Turrim Londoniarum, pro quorum ab hujusmodi captivitate liberatione, coeperunt populi Australes & Occidentales Regni plurima submurmurare, inire coetus & conventicula … ut si quid dictis masculis humanitus in turri contingeret, nihilominus per salvandas personas filiarum Regnum aliquando ad veros rediret haeredes.

Meanwhile, and while these things were going on, the two sons of King Edward aforesaid remained under a certain appointed guard in the Tower of London, and for their deliverance from this captivity the people of the southern and western parts of the kingdom began to murmur greatly, to hold meetings and confederacies … so that if anything should happen to the said boys in the Tower, as happens to mortals, the kingdom might nevertheless, through the safety of the daughters, some day return to the true heirs.

… factis proclamationibus publicis quod Dux Bukynghamiae Henricus … facti paenitens, Capitaneus & principalis existeret, vulgatum est, dictos Regis Edwardi pueros, quo genere violenti interitus ignoratur, decessisse in fata. Quamobrem hi omnes qui hanc commotionem incoeperant, videntes quod si novum suae conquestionis Capitaneum invenire non possent, brevi de omnibus actum esset, meminerunt Henrici Comitis Richmundiae …

… public proclamation having been made that Henry, Duke of Buckingham, repenting of what he had done, would be captain and chief, it was put about that King Edward’s said boys had gone to their fate, by what kind of violent end is not known. Wherefore all those who had begun this commotion, seeing that unless they could find a new captain for their enterprise it would shortly be all over with them, bethought themselves of Henry, Earl of Richmond …

What it can bear. The best-placed English writer, a former councillor writing eight months after Bosworth when blaming Richard was safe, still reports the deaths as something “put about” (vulgatum est) and says, in the present tense, that the manner “is not known” (ignoratur). What he does record as fact is the sequence: the boys under guard in the late summer; a movement whose first object was to free them; a rumour of their violent death that coincided with the proclamation of Buckingham; and the rebels’ consequent turn to Richmond, that is Henry Tudor, and Elizabeth. The plotters’ provision for “anything happening” to the boys shows that in August they thought them alive; the switch shows that by October they thought them dead. That is an inference from conduct; in this case, conduct is the strongest evidence available.
Guillaume de Rochefort at ToursAsserted as factFrench crown

Opening address of the Chancellor of France to the Estates General, 15 January 1484 · Latin, in Jehan Masselin’s journal · ed. A. Bernier, Journal des États généraux de France tenus à Tours en 1484 (Paris, 1835), p. 38, with Bernier’s French at p. 39 (read on archive.org).

Aspicite, quaeso, quidnam post mortem regis Eduardi in ea terra contigerit, ejus scilicet jam adultos, et egregios liberos impune trucidari, et regni diadema in horum extinctorem, populis faventibus, delatum.

Look, I pray you, at what has happened in that land since the death of King Edward: his children, already grown and distinguished, slaughtered with impunity, and the crown of the kingdom carried, with the people’s approval, to their destroyer. The text reads impune trucidari, not the trucidatos often quoted.

What it can bear. It shows that by mid-January 1484 the deaths of both boys were being asserted as fact at the French court, in a set speech whose purpose was to warn the Estates against the perils of a minority, with England as the cautionary tale. Rochefort gives no source; Mancini’s patron Cato sat at the same court, and Rochefort may have hardened Mancini’s suspicion into a certainty. It is the earliest surviving public assertion of the murders, made by a government hostile to Richard.
The rebels of October and the oath at RennesInference from conduct

Crowland, as above · Vergil, Anglica Historia, XXV (1555 text; Ellis’s sixteenth-century English, Camden Soc. 29, pp. 203–204) · John Howard, Duke of Norfolk, to John Paston, a Norfolk gentleman, 10 October 1483 (the Paston family letters, ed. Gairdner, VI, no. 994). No document of the oath survives.

… iam advenit Christi dies natalis … Henricus primum dato iureiurando promittit se simul atque regno potitus fuerit ducturum in matrimoniam Elizabeth Edouardi regis filiam …

“… the day of Chrystes natyvytee … erle Henry uppon his othe promysyd, that so soone as he showld be king he wold mary Elyzabeth, king Edwards dowghter; than aftir they swore unto him homage as thowghe he had bene already created king …” Vergil, Ellis tr.

It is soo that the Kentysshmen be up in the weld, and sey that they wol come and robbe the cite, which I shall lett yf I may. Norfolk to Paston, 10 October 1483. Neither the Paston nor the Plumpton family letters, the correspondence of two gentry families, mention the boys that autumn.

What it can bear. A rising raised to free Edward V ended, within weeks, as a rising for Henry Tudor and Elizabeth of York, and on Christmas Day 1483 the Yorkist exiles at Rennes, who included men of Edward IV’s household, swore homage to Richmond as king on condition that he marry Edward IV’s daughter. Both narrators wrote after 1485. But the conduct they describe is not in dispute, and it is intelligible only if the men closest to the boys believed by the autumn of 1483 that they were dead or beyond recovery. No one at Rennes was in a position to know the boys’ fate; the oath attests their belief, not the fact.
The short notices, English and foreignCommon fame

Ricart, The Maire of Bristowe is Kalendar, ed. Toulmin Smith (1872), p. 46 · Weinreich, in Visser-Fuchs, The Ricardian 7 (1986), p. 316 · Fabyan, New Chronicles, ed. Ellis (1811), pp. 669–670 · Great Chronicle of London, ed. Thomas and Thornley (1938), pp. 234, 236–237 (quoted from a secondary printing) · Rous, Historia Regum Angliae, ed. Hearne (1745), pp. 214–215 · Commynes, ed. Mandrot, I, pp. 59, 455; II, pp. 64–65 · Molinet, ed. Buchon, II, pp. 400–403 · Divisiekroniek (Leiden, 1517), fol. 392r–v · Valera, Crónica, ed. Carriazo, p. 212.

And this yere the two sonnes of King E. were put to scylence in the Towre of London. Ricart. The entry is a marginal addition of no fixable date.

Item vort diesen somer lisz sich Ritzert des konigs bruder mechtig machen und kronen in Engelandt und liesz seines bruders kinder todten und die konigin auch hemlich wegbringen.

Item, later this summer Richard, the King’s brother, had himself put in power and crowned in England, and had his brother’s children killed, and the queen also secretly put away. Weinreich, Danzig, under 1483. The queen was not put away.

In which passe tyme the prynce, or of ryght kynge Edwarde the. v., with his broder the duke of Yorke, were put vnder suer kepynge within the Tower, in suche wyse that they neuer came abrode after. … the common fame went that kynge Richarde hadde within the Tower put vnto secrete deth the. ii. sonnes of his broder Edwarde the. iiii. Fabyan.

… infra circiter tres menses vel parum ultra cum fratre suo interfecit … ita quod ex post paucissimis notum fuit qua morte martirizati sunt.

… within about three months or a little more he killed him together with his brother … so that afterwards it was known to very few by what death they were martyred. Rous. On the same page he has Richard poison his queen and spend two years in the womb.

Ils furent environ cinq sepmaines prisonniers; et par le capitaine de la tour, le duc Richard les fit occultement mourir et estaindre. Aulcuns disent qu’il les fit bouter en une grande huge … Aultres disent qu’ils furent estains entre deux quieutes …

They were prisoners about five weeks; and by the captain of the Tower Duke Richard had them secretly put to death. Some say he had them shut in a great chest … Others say they were smothered between two quilts … Molinet. He calls the boys Pierre and Georges and has Buckingham suspected.

Nu wouden enige seggen dat se die hertoge van Cloucester, hoer oem voerseit, dede versmachten om selver coninc te sine. Enige andere willen seggen dat die grave van Bockingen dese kinderen ter doot brochte … Enige seggen oec dat dese grave Henric van Bockinghen alleen een kint ter doot brochte, ende dat ander … lietet heimelic uuten lande voeren. Ende dit kint was geheten Rychaert … die Witte Rose.

Now some would say that the Duke of Gloucester, their uncle aforesaid, had them smothered in order to be king himself. Some others will say that the Earl of Buckingham brought these children to death … Some say also that this Earl Henry of Buckingham put only one child to death, and had the other secretly carried out of the country. And this child was called Richard … the White Rose. Divisiekroniek, Holland, 1517. This is the earliest printed survival story, offered as one of three rumours.

What it can bear. They show that the belief in the boys’ deaths was general and geographically wide, and in Danzig and France early: Bristol, Danzig, the French and Burgundian courts, Andalusia. That is worth something, because it shows the belief was not manufactured in 1485. Beyond that these texts cancel each other: Richard or Buckingham, a chest or a quilt or malmsey wine or poison, five weeks or three months, bodies recovered or never found. Each writer who ventures a manner admits, or shows, that he does not know it. Weinreich’s is the one entry that may be strictly contemporary, a merchant’s note that Richard “had his brother’s children killed” that summer, and it stands beside a false report about the queen.
Vergil and More, with Thornton’s findingsAsserted, from unnamed informantsTudor court

Vergil, Anglica Historia, XXV, drafted c. 1512–13, printed Basel 1534 (Latin of the 1555 edition; Ellis’s English, pp. 188–189) · More, The History of King Richard the Third, written c. 1513–18, printed from his manuscript by Rastell in 1557, ed. Lumby (1883), pp. 80–85 · Tim Thornton, History 106 (2021), pp. 4–25; Historical Research 97 (2024), pp. 218–225; History 109 (2024), pp. 445–460, with TNA PROB 11/19.

Ille imperata facere compulsus dolenter Londinum proficiscitur ac regios pueros exemplo in omni memoria prope inaudito necavit. Ita Edouardus princeps cum Ricardo fratre interiit, sed quo genere mortis miselli pueri affecti fuerint non plane liquet.

He [Tyrell], compelled to do as he was ordered, went sorrowfully to London and killed the royal boys by an example almost unheard of in all memory. Thus Prince Edward perished with his brother Richard, but by what kind of death the poor little boys were despatched is not at all clear. Vergil. He places the order during the royal progress of August–September 1483 and knows no confession and no grave.

… some remain yet in doubt, whither they wer in his dayes destroyde or no. … I shall rehearse you the dolorous end of those babes, not after euery way that I haue heard, but after that way that I haue so hard by suche men and by such meanes, as me thinketh it wer hard but it should be true. … [Forest and Dighton] keping down by force the fetherbed and pillowes hard vnto their mouthes, that within a while smored and stifled … caused those murtherers to burye them at the stayre foote, metely depe in the grounde vnder a great heape of stones. … a prieste of syr Robert Brakenbury toke vp the bodyes again, and secretelye entered them in such place, as by the occasion of his deathe, whiche onely knew it, could neuer synce come to light. … bothe Dighton and he were examined, and confessed the murther in maner aboue writen, but whither the bodies were remoued thei could nothing tel.

I bequeath to my sonne Sir Giles his fadres Cheyne which was Yonge kynge Edward the Vth. Will of Margaret Capell, 1516, proved 1522, TNA PROB 11/19; in print since 1826; Thornton 2024.

What it can bear. Vergil, the official historian, names Tyrell and Richard, dates the deed to the progress, and confesses that the manner is unknown. More, thirty years on and five years old at the time, supplies the manner, the men, the grave and a confession, and warns that some still doubted whether the boys died in Richard’s reign at all. Thornton has strengthened More on two points and weakened him on one. Miles Forest’s sons were at Henry VIII’s court in More’s circle in the 1510s, one of them a messenger on More’s own 1515 embassy, so that More had a route to the family of a named killer; and a chain “which was Yonge kynge Edward the Vth” was in 1516 in the hands of Margaret Capell, whose half-sister was Tyrell’s wife and whose family were clients of Thomas More’s father, the judge John More; this connects an object of the boy’s to Tyrell’s household. Against this, Thornton’s own search of the 1502 records finds Henry VII at the Tower in the days around Tyrell’s trial and no direct evidence of any confession, and the will has been in print since 1826 (N. H. Nicolas’s Testamenta Vetusta, 1826) and was used by several scholars, none of whom saw a link to Tyrell in it, since Tyrell’s kin could have come by a dead boy’s chain in more ways than one. More’s narrative is the Tudor court’s version, from informants he does not name, and it describes a burial that he says was undone.
The bones under the stairPhysical, unidentified

Francis Sandford, a herald, A Genealogical History of the Kings of England (1677), pp. 402–404, from “a Gentleman, an eye-witness” · urn of 1678 in Henry VII’s Chapel, Westminster Abbey, inscription in Keepe, Monumenta Westmonasteriensia (1683) · L. E. Tanner, the Abbey’s archivist (Keeper of the Muniments), and W. Wright, professor of anatomy, Archaeologia 84 (1935), pp. 1–26 · critiques: Hammond and White, in Richard III: Loyalty, Lordship and Law (1986), pp. 104–147; Molleson, London Archaeologist 5 (1987).

… digging down the Stairs which led from the Kings Lodgings, to the Chappel in the said Tower, about ten foot in the ground, were found the Bones of two Striplings in (as it seemed) a wooden Chest, which upon the survey were found proportionable to the ages of those two Brothers, viz. about thirteen and eleven years. … the Labourers, who not being sensible of what they had in hand, cast the rubbish and them away together, wherefore they were caused to sift the rubbish, and by that means preserved all the Bones. The Circumstances from Story being considered … upon the presumptions that these were the Bones of the said Princes, His Majesty King Charles the Second, was graciously pleased to command that the said Bones should be put into a Marble Urne …

Hos fratres germanos, Turre Londinensi conclusos, injectisque culcitris suffocatos, abdite & inhoneste tumulari jussit patruus Richardus perfidus regni praedo. The urn inscription of 1678. It repeats More’s account.

What it can bear. Two children’s skeletons were dug from about ten feet of rubble at the foot of a Tower stair on 17 July 1674, thrown on the spoil heap, sifted out again, kept loose for four years, and identified “upon the presumptions” and “from Story”, that is, from More. The single examination of 1933 was conducted with More’s account as its working hypothesis; it estimated ages of about 12–13 and 9–11, found no way to sex the bones, and could not date them. The Abbey declined the requests made in 1993, 1995, 2013 and 2018, and as of March 2026, the latest report found, no radiocarbon, sex or DNA test had been made. Until one is, the bones cannot bear a date, an identity or a cause of death. They are also the one piece of evidence that could settle the whole question: children of the right ages in the wrong century would end the murder story as told, and a match to the Yorkist maternal line with a date in the 1480s would end the survival story.

Finding on the murder sources. Nothing here is knowledge of a death. What the sources establish is a sequence: two boys alive on 21 June 1483; withdrawn from sight after Hastings’ execution; a suspicion of death in London by July and a report of killing that summer in the Hanseatic port of Danzig; a rebellion raised to free them that turned, by October, into a rebellion for another king; the boys’ own household men swearing to that king at Christmas; both deaths asserted as fact in France by January; and a king who, through two years in which the rumour was costing him his southern support, never produced the boys, never explained their absence, and in March 1484 gave their mother an oath that covered her daughters only. The later narratives add names and methods that contradict one another and prove nothing beyond the strength of the belief. The case for death in 1483 is a case from conduct, Richard’s above all. It is strong because the conduct is not in dispute and has no good explanation on any hypothesis in which the boys were alive and could have been shown. It is not proof; the numbers in the next section quantify a judgment and do not prove it.

Weighing it

The evidence falls into five clusters. Items inside a cluster are not independent: Mancini, Crowland and Rochefort largely report one London rumour, so the cluster gets one judgement, not three. For each cluster the table gives the probability of seeing that body of evidence if each outcome were true. Only the ratios matter. These numbers (likelihoods, in the language of probability) are judgments about how strongly each body of evidence favours each outcome; the ledger makes them explicit, keeps correlated sources together, and shows what follows from them. It does not turn them into evidence, and a reader who assigns different numbers will get a different verdict; the sensitivity lines below say how different. The survival outcome is split three ways because the 1487 evidence bears on Edward V and the 1490s evidence on Richard, and the two must not be allowed to cancel each other out.

Evidence clusterDied 1483Edward livedRichard livedBoth livedOther
C1 1483: withdrawn from view; London believes them dead by December (Mancini); the rebels turn to Tudor and Elizabeth Woodville backs the Tudor marriage; Rochefort’s speech; the 1484 oath covers daughters only; Richard never produces them in two years0.850.150.150.120.20
C2 Henry VII, 1485–1502: “shedding of Infants blood” but no bodies, inquest or published confession; repeals Titulus Regius and marries Elizabeth of York0.600.450.500.400.40
C3 The 1487 claimant: Symonds, Warwick shown, Henry to the Pope, “Edward the vjt”, “1 Edward VI”, the Butler patent’s first year, Voorne, Mechelen, Molinet, de But; against them the Lille receipt and André0.300.030.300.030.30
C4 “Perkin Warbeck”: the Setúbal depositions, the Tournai records, the 1497 confession, no recognition by anyone who had known Richard of York; against them the Gelderland narrative, the recognitions by Margaret, Maximilian, James IV and the Penitentiary, the Dresden pledge, the marks letter0.600.600.100.100.60
C5 Traces of death: More’s narrative with Thornton’s findings on Forest’s sons, Dighton and the Capell chain; the 1674 bones0.350.120.120.080.25

Starting from a prior (the weight given to each outcome before this evidence is considered) of 40 per cent for death in 1483, 30 per cent for survival in some form and 30 per cent for something else, the ledger gives 92 per cent, under 1 per cent and 8 per cent. That last figure for survival is too confident for a case built on documents: a misread or forged text on either side, or a lost one, would move it. Mixing the result with a flat distribution at one part in ten, as an allowance for what has not been seen, gives the figures reported above, 85, 6 and 9. The prior does not drive the result: a flat prior gives 82, 7 and 11. Dropping the largest cluster, C1, altogether still leaves 67, 7 and 25. Halving the strength of the three clusters that cut hardest against survival (C1, C3, C4) gives 77, 8 and 15. The script, ledger.py, and all the numbers are published alongside this page.

The ledger makes two things visible. First, the murder verdict is carried by 1483, not by More or the bones: remove C5 and the answer barely moves. Second, the survival documents are not worthless; C4 rates the Gelderland narrative and the recognitions as somewhat more likely under survival than under imposture. They are outweighed because the Setúbal depositions and the Irish regnal year are evidence that neither the claimant nor his backers produced.

The strongest case for survival

Put at its best, the case runs like this. Henry VII never named the boys as victims, never produced bodies, never published Tyrell’s confession, and never addressed the question publicly. Elizabeth Woodville came to terms with Richard in 1484 and was shut away by Henry in 1487, weeks before a Yorkist rising. The Earl of Lincoln, who had a better claim than Warwick, went to war behind a boy. A Habsburg receipt calls that boy a son of Edward IV and so, in his way, does Henry’s own court poet; the official Lambert Simnel is ten years old while the crowned youth leads an army. The 1490s claimant was polished, was recognised by the aunt who had seen the real boy in 1480, was vouched for by the King of the Romans down to marks on his body, was married by James IV to a kinswoman of his own, and told a coherent story of his escape that survives in a manuscript of about 1500 at Arnhem. Meanwhile the murder story rests on Mancini, who says he could learn nothing; on Crowland, who reports a rumour; on a French chancellor making a political speech; and on More, who wrote thirty years later and describes a burial that he says was undone.

None of this is false, but none of it survives two tests. The 1487 boy was dated in his first regnal year by the men who governed in his name, which Edward V could not have been unless the count restarted, for which no precedent is known; the 1490s claimant was placed in a Tournai family by Portuguese witnesses deposed in April 1496, a year before Henry VII had him, and by Tournai’s own records. Patrons at war with Henry VII had reason to recognise any claimant: Charles VIII received the same man and dropped him at the treaty of Étaples, his peace with Henry VII in November 1492. Richard’s silence weighs more than Henry’s. A king in Richard’s position, whose southern gentry had risen on a report that his nephews were dead, would have shown two living nephews if he had them. He did not do so in 1483, in 1484 after the death of his own son made the succession urgent, or in 1485 when Henry Tudor’s invasion was expected.

What would overturn this verdict

The identity of the Dublin king would be reopened by a dated Irish document of 1487 reckoning “King Edward’s” regnal year from April 1483, or by proof that the memoranda-roll abstracts were mis-copied.

The survival question would be reopened by a record independent of claimants and their patrons showing a prince alive after the autumn of 1483: a household or estate entry, a safe-conduct, a letter, securely naming Edward V or Richard of York rather than “the lord bastard” (which fits John of Gloucester, Richard III’s illegitimate son, as well) or “the Duke of York” as a claimant styled himself. So would proof that the Setúbal depositions or Symonds’s confession were fabricated, or that Tournai’s Werbecque family did not exist as described. So would a demonstration that the Gelderland narrative contains checkable facts that only the real Richard of York could have known, as opposed to facts that Yorkist exiles at Margaret’s court would have known; nothing of that kind has been identified.

The bones in the urn could also decide it. Radiocarbon dating and sexing of the 1674 bones, with a comparison of mitochondrial DNA (inherited through the mother) against a Woodville maternal line reported as traced in 2018, would not say who was responsible. But bones of two boys of the right ages, of the right century and of that line, found where More said the bodies were first laid, would end the survival hypothesis; bones of the wrong century or sex would remove C5 and lower the verdict to about 82 per cent (the ledger’s drop-C5 line) without touching 1483. As of March 2026, the latest report found, neither the Abbey’s governing body, the Dean and Chapter, nor the Crown had changed the position taken in 1993, 1995 and 2018.

What is new here

Nothing on this page is a new document, a new reading of a manuscript or a new fact. What follows says where each argument comes from, so that the checking and arrangement done here can be told from the scholarship it rests on.

The verdict is not new. That both boys most probably died in 1483 in Richard III’s custody is the position of Ross (1981), Horrox (1989), Pollard (1991), Hicks (2003, 2024), Cunningham (2003), Horspool (2015) and Skidmore (2017). Re-reading the sources behind it found it to stand.

The regnal-year argument is Randolph Jones’s and Chris Phillips’s. Jones (2014, 2021, 2023) worked it out from the Irish exchequer abstracts and the Butler patent; Phillips (2024, pp. 9–10) restated it. It was checked here against the printed records and against Smith’s 1996 dating of the patent. The Lille receipt’s printed history and impossible chronology are Phillips’s too (2024): Finot abstracted it in 1895, Bletzer quoted it in 2013, and its “June 1487” purchase cannot fit a force that sailed by early May. Finot’s abstract was read for this page and its French is quoted above.

The identification of the 1490s claimant as Piers Werbecque of Tournai is Gairdner’s (1898), with Wroe (2003) on Setúbal. The age discrepancy in the 1493 letter (“nonum fere agens annum”) is Gairdner’s observation (p. 268 n.). The reading of the Gelderland text as propaganda that either hypothesis requires is Hicks’s (2024), anticipated by Enklaar’s question in 1951; that it tracks the 1493 letter and the 1497 confession in outline has been said by Amin and, in the opposite sense, by the Richard III Society’s research pages. The Tyrell and Forest overlap with More is Hicks’s and Thornton’s. Forest’s sons at court, the 1502 itinerary and the Capell chain are Thornton’s (2021, 2024). The argument from Henry VII’s silence is Potter’s (1983) and older.

What is new is arrangement and checking. The three-jurisdiction table for 1487, built on Phillips’s 2024 inventory with Smith (1996) and Jones (2014) added, is arranged to test the survival thesis’s claim that the English records are Tudor propaganda. The Gelderland point is sharpened: the stations in the Arnhem text that are absent from the 1493 letter are exactly the Setúbal and confession stations, in the same order. The murder sources are set out side by side in their original words and classed as rumour, inference or claimed knowledge, with one small textual note: Masselin’s printed text of Rochefort’s speech reads impune trucidari, where trucidatos is sometimes quoted. And the five-cluster ledger with its sensitivity tests shows that the verdict is carried by the events of 1483 and survives the removal of More, the bones and Thornton’s findings together. None of this changes the answer; each part makes an existing argument easier to check.

What was not done. The Gelderland manuscript was not read in full in Dutch, only in Langley’s translation and the quoted extracts; the Setúbal depositions and the Tournai records were read in Gairdner’s and Wroe’s printings, not the originals; Mancini’s Latin was not available; Hicks’s 2024 note was available only in abstract; the Ricardian Bulletin items of 2023–25 were read through intermediaries; Langley’s book itself was read in extracts and reviews, not in full. None of these is likely to change the reading, but each is a place where it could be checked further.

Sources

Primary texts are listed with the edition read; where only a summary or a secondary quotation was available, this is said. Archive references are given as published. Web addresses were live on 14 September 2026.

Manuscripts and archival records

Printed primary sources

Scholarship and debate

Method and files