England, Ireland and Burgundy · 1483 to 1499
Did the boy king Edward V and his younger brother Richard, Duke of York, die in the Tower of London in 1483, or did one or both survive to become the two claimants to the throne of 1487 and the 1490s?
What happened. King Edward IV of England, the first king of the house of York, died in April 1483, leaving two sons: Edward V, aged twelve, and Richard, Duke of York, aged nine. Their uncle Richard, Duke of Gloucester, the dead king's younger brother, took charge of the young king, lodged both boys in the Tower of London, then a royal residence as well as a fortress and a prison, had them declared illegitimate, accepted the crown on 26 June and was crowned Richard III on 6 July. Over that summer the boys' servants were dismissed and, in the words of Dominic Mancini, an Italian visitor who wrote up what he had seen for a French patron that December, they were seen "more rarely behind the bars and windows, till at length they ceased to appear altogether"; no one recorded a dated sighting of either boy again. By October, men who had plotted to free them had heard that they were dead and rose instead for Henry Tudor, an exile in Brittany with a distant claim to the throne, who killed Richard at Bosworth in August 1485 and became Henry VII.
Why it is a mystery. No one who wrote at the time claims to know whether the boys died, or when, or how: Richard III never said; Henry VII never produced bodies or a published confession; the first full account of a murder was written about thirty years later by Thomas More, the lawyer who later became Henry VIII's Lord Chancellor; and two children's bones found under a Tower staircase in 1674 have never been dated, sexed or tested. Most historians of the period conclude that the 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, 2024) and Sean Cunningham (2003), and of the recent biographies by David Horspool (2015) and Chris Skidmore (2017), though the Duke of Buckingham, Richard's chief ally who turned against him that autumn, and Henry VII himself have also been proposed as the killer. Richard III's defenders, known as Ricardians and organised since 1924 in the Richard III Society, answer that no death is recorded and that two later claimants to the throne may have been the princes: the boy crowned "King Edward" at Dublin in 1487, called Lambert Simnel, an Oxford organ-maker's son, by Henry VII's government, and the "Richard, Duke of York" of the 1490s, who after his capture in 1497 confessed to being Piers Werbecque, called Perkin Warbeck, born at Tournai, a town in what is now Belgium. In 2023 Philippa Langley, the writer who led the search that found Richard III's grave in 2012, and her Missing Princes Project put four documents forward as evidence that they were: the Lille receipt, a receipt for pikes in the accounts of the artillery of Maximilian, the Habsburg King of the Romans (the elected German king) who governed the Low Countries, kept at Lille in northern France and dated December 1487, which calls the Dublin king "son of King Edward"; the Gelderland narrative, a first-person account of Richard of York's escape in a manuscript of about 1500 in the provincial archive of Gelderland at Arnhem in the Netherlands; the Dresden pledge, 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 Maximilian's "three marks" letter, a letter from the same winter naming three bodily marks by which the claimant could be known.
How it was judged. Every document on both sides was read in print or online, in its own language with a translation where an edition could be reached (Mancini's Latin, the full Dutch of the Gelderland narrative and the Portuguese witness statements of 1496 could not be, and were read in translation or in summary; the Dresden pledge and the "three marks" letter are known only from the Missing Princes Project's descriptions), and tagged by what it can bear: for the survival documents, whether it shows a prince alive or only that a claimant or his backer said so; for the murder sources, whether the writer knew, inferred or reported a rumour, and whose side he was on. The evidence was then sorted into five clusters (the events of 1483; Henry VII's conduct from 1485 to 1502; the 1487 claimant; the 1490s claimant; and the traces of death, More's narrative and the bones) and weighed in a ledger, computed by a script published with the full report, which is linked from the page: each cluster is given a number for how likely that evidence would be under each of five outcomes (both boys died in 1483; Edward lived; Richard lived; both lived; something else), the numbers, which are judgments rather than measurements, are combined by Bayes's rule, the standard way of weighing evidence, from a stated starting weight, and the ledger is re-run with clusters dropped or softened to show how far the verdict moves when any number changes. Three things tipped the balance: the Lille receipt is contradicted by every other contemporary record that says whom the Dublin king claimed to be, and by the dates on his own documents, which count his reign as beginning in 1487, as Edward V's could not have done, 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, an argument the survival side has not answered; the 1490s claimant was placed in a Tournai family by Portuguese witnesses whose depositions, sworn statements, were taken at Setúbal, near Lisbon, in April 1496, a year before Henry VII had him; and Richard, who could have ended the rumour of the boys' deaths by producing two living nephews, did not do so in the two years of his reign. The result is that both boys most probably died in the Tower in 1483, in the late summer or autumn, in Richard III's custody, about 85 chances in 100, against about 6 in 100 that one or both lived on, nearly all of it an allowance for evidence not yet found, and about 9 in 100 for a death at some other time or place; a record of either boy alive after the autumn of 1483 made by someone not later paid to say so, or radiocarbon, sex and DNA tests on the 1674 bones, would change it.
Both Edward V and Richard, Duke of York, most probably died in the Tower in 1483, in Richard III's custody: about 85 in 100. One or both survived to adulthood: about 6 in 100; something else: about 9 in 100.
A judgment on the balance of published evidence rather than a solution: the 85 per cent depends on the five outcomes and the cluster weights set in the ledger, and the case for murder rests on how people behaved rather than on proof. Death in 1483 is also the position of the named studies of the reign, known here through secondary summaries.
The 85 covers a death on Richard's order, an unauthorised killing (Buckingham has been proposed) and a natural death of the elder boy; the ledger does not separate them, and the sources point to his order but do not establish it. The 6 for survival, as "Simnel", "Warbeck" or otherwise, is nearly all the 10% allowance for evidence not yet found: the ledger without that allowance gives 0.003. The 9 for something else covers a death in 1484 or 1485 still in Richard's keeping, a death under Henry VII, or a death abroad as children.
Every figure depends on the five outcomes and the cluster weights set in the ledger, and the likelihoods, the numbers given to each cluster under each outcome, are judgments rather than measurements. The share for death in 1483 depends most on the cluster of 1483 events: across the ledger's runs it falls no lower than 67, when that whole cluster is dropped, and rises no higher than 92, with no allowance for unknowns; removing More, the bones and the findings of Tim Thornton, a historian whose articles of 2021 and 2024 trace More's informants and the later history of Edward V's belongings, all together barely moves it.
Put at its strongest, the survival case runs like this. Henry VII never named the boys as victims, never produced bodies 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; Elizabeth Woodville, the boys' mother, came to terms with Richard in 1484 and was shut away by Henry in 1487, weeks before a Yorkist rising; the Earl of Lincoln, Richard III's nephew, who led that rising and was killed when its army was destroyed at Stoke, near Newark, on 16 June 1487, went to war behind a boy whom a Habsburg receipt calls a son of Edward IV; and the 1490s claimant was recognised by the boys' aunt Margaret, the widowed Duchess of Burgundy, who had seen the real boy in 1480, vouched for by Maximilian down to marks on his body, and married by James IV of Scotland to a kinswoman of his own. Each of those facts is accurate, and two tests decide against the case built on them: 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; and the 1490s claimant was placed in a Tournai family by Portuguese witnesses deposed in April 1496, a year before Henry VII had him. Patrons at war with Henry VII had reason to recognise any claimant: Charles VIII, King of France, received the same man and dropped him at the treaty of Étaples, his peace with Henry VII. 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, and 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.
The shares are the five-cluster ledger's output, with its three survival outcomes (Edward lived, Richard lived, both lived) added into one, after a 10% allowance for evidence not yet found: 0.848 / 0.063 / 0.090 for died in 1483 / survived / something else, against 0.920 / 0.003 / 0.077 without it. The weights move them: a flat prior, an equal starting weight for every outcome, gives 0.820 / 0.071 / 0.110; dropping the 1483 cluster entirely, 0.674 / 0.072 / 0.254; dropping More, Thornton and the bones, 0.818 / 0.068 / 0.114; halving the strength of the three clusters that cut hardest against survival, 0.773 / 0.077 / 0.150.
Every document on both sides was read in print or online, in its own language with a translation where an edition could be reached. On the survival side: the four documents put forward in 2023 by Philippa Langley, the writer who led the search that found Richard III's grave in 2012, and the records around them, the claimant's own letters of 1493 and 1497, the depositions taken at Setúbal in 1496 and the town records of Tournai. On the murder side: every source from a letter of 21 June 1483 to Thomas More's history, written about thirty years later, and the bones found under a Tower staircase in 1674.
Some could not be reached: the Latin of Dominic Mancini, the Italian visitor who wrote in December 1483, the Dutch of the Arnhem manuscript, which has no edition and was read in Langley's translation as quoted, the Setúbal depositions themselves, read in summary, and the Great Chronicle of London, a city chronicle written up about 1512. The Dresden pledge and Maximilian's "three marks" letter are known only from Langley's descriptions of them, the historian Michael Hicks's 2024 article from its abstract, and the standard studies of the reign through secondary summaries. Nothing was seen in manuscript.
Each document was tagged by what it can bear: for the survival documents, whether it shows a prince alive or only that a claimant or his backer said so; for the murder sources, whether the writer knew, inferred or reported a rumour, and whose side he was on. The evidence was then sorted into five clusters and weighed in a ledger: each cluster was given a number for how likely that body of evidence would be under each of five outcomes, and the numbers were combined by Bayes's rule, the standard way of weighing evidence, from a stated starting weight, then re-run with clusters dropped or softened.
Figure 1 shows the three outcomes as one bar, its widths the ledger's shares: 85 per cent that both boys died in 1483 in Richard III's custody, 6 that one or both survived to adulthood, 9 for something else. Table 1 is the ledger: the five clusters and, for each, the probability of seeing that body of evidence if each of the five outcomes were true; only the ratios matter, and each number is a judgment about the evidence. The shares depend on those weights: the 85 falls to 67 when the whole 1483 cluster is dropped and rises to 92 with no allowance for what has not been seen.
Three things tipped the balance: the Lille receipt is contradicted by every other contemporary record of whom the Dublin king claimed to be, tabulated by jurisdiction, and by the dates on his own documents, which count his reign from 1487, as Edward V's could not have done; the 1490s claimant was placed in a Tournai family by Portuguese witnesses deposed at Setúbal in April 1496, a year before Henry VII had him; and Richard III, who could have ended the rumour by producing two living nephews, never did so in his two years as king. The result is about 85 per cent that both boys died in 1483, in Richard III's custody.
| Evidence cluster | Died 1483 | Edward lived | Richard lived | Both lived | Other |
|---|---|---|---|---|---|
| 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 years | 0.85 | 0.15 | 0.15 | 0.12 | 0.20 |
| C2 Henry VII, 1485–1502: “shedding of Infants blood” but no bodies, inquest or published confession; repeals Titulus Regius and marries Elizabeth of York | 0.60 | 0.45 | 0.50 | 0.40 | 0.40 |
| 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 André | 0.30 | 0.03 | 0.30 | 0.03 | 0.30 |
| 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 letter | 0.60 | 0.60 | 0.10 | 0.10 | 0.60 |
| C5 Traces of death: More’s narrative with Thornton’s findings on Forest’s sons, Dighton and the Capell chain; the 1674 bones | 0.35 | 0.12 | 0.12 | 0.08 | 0.25 |
Already established by others. The verdict agrees with Charles Ross (1981), Rosemary Horrox (1989), A. J. Pollard (1991), Michael Hicks (2003, 2024) and Sean Cunningham (2003), and with the recent biographies by David Horspool (2015) and Chris Skidmore (2017); their books are known here through summaries in the secondary literature, and Hicks's 2024 note was read in abstract. The regnal-year argument is Randolph Jones's (2014, 2021, 2023) and Chris Phillips's (2024). The account of the Lille receipt's printing in Finot's 1895 inventory and of its impossible June 1487 chronology is Phillips's (2024), with Michael Bletzer (2013), whose study of the Habsburg captains at Bosworth and Stoke had quoted the receipt, for the earlier quotation. The identification of the 1490s claimant as a native of Tournai was made by James Gairdner (1898), the Victorian historian of the reign, with Ann Wroe (2003), whose Perkin is the fullest modern study of the claimant, on Setúbal; the age error in the claimant's 1493 letter, which makes the prince a year younger than he was, is Gairdner's (1898, p. 268 n.). The reading of the Gelderland narrative as propaganda that either hypothesis needs, whether the claimant was the real Richard of York or a pretender, since a pretender required exactly such a text, is Hicks's (2024), anticipated by a question asked in 1951 by D. Th. Enklaar, the Utrecht professor to whom the manuscript was first sent; that the narrative follows the 1493 letter and the 1497 confession in outline had been noted by Amin and by the Richard III Society's research pages. The observation that the Gelderland narrative and More name the same two men, Tyrell and Miles Forest, one of the two killers More names, among those who had charge of the boys is Hicks's (2024) and Thornton's (2024). Three further findings are Thornton's (2021, 2024): that Forest's sons were at Henry VIII's court in More's own circle in the 1510s, so that More had a route to the family of a named killer; that Henry VII was at the Tower in the days around Tyrell's trial in 1502, with no direct evidence of any confession; and the Capell chain, a chain that a 1516 will describes as the young King Edward V's, which connects an object of the boy's to Tyrell's household. The argument from Henry VII's silence was made by Jeremy Potter (1983), a Ricardian writer, and by older writers.
What historians have said. Charles Ross (1981), Rosemary Horrox (1989), A. J. Pollard (1991), Michael Hicks (2003, 2024) and Sean Cunningham (2003), and the recent biographies by David Horspool (2015) and Chris Skidmore (2017), hold that both boys most probably died in 1483 in Richard III's custody. Against that position Philippa Langley's Missing Princes Project (2023; updated 2025), with the Richard III Society's research pages, argues from four named documents, the Lille receipt of December 1487, the Gelderland narrative at Arnhem, the Dresden pledge of October 1493 and Maximilian's "three marks" letter, that both boys survived. Earlier Ricardian writers proposed other answers: Paul Murray Kendall (1955) blamed Buckingham, Clements Markham (1906) and Bertram Fields (1998) blamed Henry VII, and David Baldwin (2007), Matthew Lewis (2018) and John Ashdown-Hill (2018) argued for survival.
The survival documents were re-read one by one, in the original language where a printed text could be reached and otherwise in Langley's translation or the Missing Princes Project's descriptions, beside the other records of their own archives, and the position of the named studies was found to stand. Two tests decide it: the regnal year in the Dublin king's own records, against the Lille receipt, and the Portuguese depositions of April 1496 that place the 1490s claimant in a Tournai family, against the other three documents. Chris Phillips reached the same conclusion on 1487 in 2024, in a note written with four members of Langley's own Dutch Research Group, the Dutch researchers of her team, and so did Michael Hicks in the journal Historical Research the same year.