The oldest surviving account of the piper and the 130 lost children was written about 150 years after the event. This page reads every source in order, tests seven explanations against them, and states which of them the evidence supports.
Hamelin (in German, Hameln) is a small town on the river Weser in northern Germany, some 40 kilometres south-west of Hanover. The story the town has told since the fifteenth century runs as follows. On 26 June 1284, the feast day of Saints John and Paul, a stranger came over the bridge and in through the gate on the river side: a young man of about thirty, handsome and so well dressed that people stared, carrying a silver pipe. He played it through the streets, and the children of the town, 130 of them, followed him out through the gate on the far side, towards the hill outside the walls where the town carried out its executions, and there they vanished. Their mothers went from town to town looking for them and found nothing. Afterwards, the story says, Hamelin counted its years from that day, the way the Church counted years from a jubilee: the first year after the children went out, the second, the third. The remaining elements of the modern story entered the tradition later. The rats, the town's refusal to pay the piper's fee and the mountain that opened to take the children in first appear in books of the 1550s and 1560s; the lame boy left behind first appears in an English book of 1605, from which Robert Browning took his poem of 1842.
Nothing written at the time survives. The earliest account is a short Latin note added about 1430–50, some 150 years after the event, to a manuscript that is now in the town library of Lüneburg, a city about 150 kilometres to the north-east (I call it the Lüneburg note below); its writer says he copied it from "an old book". The inscriptions carved on Hamelin's houses that give the date and the number begin in 1525, and the first printed accounts date from 1556. Historians therefore argue about two questions: whether anything happened, and, if it did, what it was. Hans Dobbertin, a local historian who collected the sources in 1970, and Jürgen Udolph, a German specialist in place names, hold that the "children" were young people recruited for the German settlement of the lands east of the river Elbe, which was actively under way in the 1280s, and Hamelin's museum today takes the same view. Others read the story as a religious procession of the young that came to grief; as an accident on the hills outside the town (the folklorist Waltraud Woeller in 1957–61, the local historian Gernot Hüsam since 1990); as a levy of young men for a local war (Fein, a garrison preacher, in 1749; Thomas Küntzel in 2010); or as a crime. Some folklorists doubt that any single event lies behind the tale. The version most often repeated in English, that Udolph proved where the children went by feeding Hamelin's surnames into a computer, rests on a press report of 1998; Udolph published no such study.
The evidence is the set of surviving witnesses to the story, read in order of date and, wherever possible, from the originals: the Lüneburg note from the library's online facsimile, the house inscriptions from the standard edition of 1989, the sixteenth-century printed accounts from digitised copies, and the medieval charters of Hamelin and of Pomerania from the printed charter books. The method is to list what the tradition claims and when each claim is first attested, then to score seven explanations against those claims and against the documentary record of the 1280s, and to attach to each explanation a probability expressing my own estimate of its likelihood. The result is that a real loss of young people probably lies behind the tradition (I put it at 80 in 100, with the caution that no surviving witness is older than about 1430); that a recruited departure for the new villages east of the Elbe is the best single explanation, at 42 in 100, which makes it the leading candidate, with the other six explanations together at 58; and that the sources cannot say where the group went. Two documents turned up that I have not seen used for the legend before: a bishop's letter of 1502 that fixes the place the inscriptions name, and a Pomeranian charter of 8 July 1284 that puts a count from the next lordship east of Hamelin at the Pomeranian duke's court twelve days after the date of the exodus. Neither document shows what happened in 1284.
The report has these sections. "What survives, and from when" lays out every witness in order of date. "The earliest text, read from the page" quotes and translates the Lüneburg note. "The town's own formula" gives the house inscriptions and a lost prayer from the town's principal church. "Where the town placed 'Calvarie bi den Koppen'" takes the inscriptions' own words for the place where the children were lost and uses the bishop's letter of 1502 to fix them to a spot on the map. "The explanations, tested" scores seven explanations against the tradition's features and argues the ranking. Then come what is new here and what only agrees with earlier scholars, what evidence would overturn the verdict, and where the report fell short.
The result is a ranking of the seven explanations by estimated likelihood; it does not establish which of them is correct. A real loss probably happened: I put it at 80 in 100 that Hamelin lost a group of its young people in or close to 1284. Of seven explanations for how, a recruited departure for the new villages then being founded east of the Elbe ranks first, at 42 in 100; the other six together carry 58. Emigration is therefore the leading candidate: several explanations remain possible and are put in order, it stands ahead of each rival, and the order can change when one judgement changes. Where they went cannot be recovered from the sources, and the "surname proof" repeated in popular accounts has no published basis.
The earliest written account, set down about 150 years after the event, tells the story as follows. On the feast day of Saints John and Paul (26 June) in 1284, a handsome, well-dressed young man with a silver pipe came into Hamelin by the gate on the river side, the Weser gate. All the pueri who heard him, 130 in number, followed him out through the east gate "as if towards" the place called Calvary (the name of the hill of the Crucifixion; the note itself explains that here it means the town's beheading place), and were never found. Their mothers "ran from town to town and found nothing at all". Every clause of that rests on the Lüneburg note, written about 1430–50 by someone who says he found it "in an old book" and names one eyewitness: the mother of a dean of Hamelin's principal church who left a will in 1378. Later witnesses repeat parts of it. The year, the feast day and the number recur in a rhymed Latin prayer of unknown date, once written in a service book of that church and now lost, and on inscriptions carved on the town's houses from 1525 onward. The Calvary recurs in the same prayer, in the physician Johann Weyer's book on demons of 1566 and on inscriptions from 1572. A piper (or a magician) leading the children out of a gate is in every source from 1556 on; from Weyer's book the well-dressed young man is a piper in a coat of many colours. A search turns up in another form (the parents, "on water and land") in the first printed version, a book of marvels by the Jena physician Jobus Fincelius, 1556. But the entry by the Weser gate, the exit by the east gate, the silver pipe and the mothers running from town to town have no witness except the Lüneburg note. The calendar date "26 June" is a sixteenth-century conversion of the feast day, first on an inscription on a house on the market square (Markt 1), carved in 1525 or thereabouts and known only because Samuel Erich, the rector of the town's Latin school, printed its wording in 1654; "Monday" is my own computation from it. The prayer that some have dated to the fourteenth or fifteenth century is lost and known only from 1654. The explanations ranked below are therefore explanations of a tradition; no documented account of the day exists to reconstruct.
If the tradition is broadly true, the explanation that fits most of it is that the young people were recruited by a locator (the contractor who gathered settlers for a lord) and led away to the new villages being founded east of the Elbe. (The words the sources use, Latin pueri and kinder in Middle Low German, the language of medieval northern Germany, mean unmarried young people, servants and crew, not only small children.) One channel is documented: the counts of Spiegelberg, a family of minor nobles who from 1281 held Coppenbrügge, the next lordship east of Hamelin, had ties to Pomerania, the duchy on the Baltic coast about 470 kilometres to the north-east: Count Nicolaus is recorded at the Pomeranian duke's court as the duke's kinsman between November 1282 and 8 July 1284, twelve days after the tradition's date. But no document names a single settler from Hamelin anywhere; the place-name evidence dates nothing; and the story that Jürgen Udolph, a German specialist in place names, fed the town's surnames into a computer and found them clustered in the Prignitz, a district of Brandenburg, comes from a 1998 press report and does not appear in his one scholarly article. So emigration ranks first without being likelier than all the alternatives together, and nearly a fifth of the probability stays with the possibility that the tradition records no single event at all.
Figure 1. Probabilities across seven explanations. Made for this report from the weighting of the features set out in "The explanations, tested". These numbers express a judgement; they were not measured and do not count as evidence. They rank the seven explanations listed and depend on the weights chosen; only the first digit of each figure carries any information. They add to 100, and the reasoning for each is in "The explanations, tested" below. The 42 for emigration places it first; the other six together carry 58. The ranking depends chiefly on one judgement, the 80 in 100 for a real loss. Put that at 50 (keeping the event explanations in the same proportion to each other) and "no single event" rises to about 50 while emigration falls to about 26, still first among the event explanations; put it at 90 and emigration rises to about 46. Four explanations are left off the list because they fail the year given by the Hamelin witnesses, or fail the settlement record: the children's crusade of 1212, the battle of Sedemünder near Hamelin (1259/60), the Black Death (1347 onward) and the German settlement of Transylvania.
No surviving text of the thirteenth or fourteenth century mentions the exodus. The date "about 1300" that most retellings give the stained-glass window was assigned by Hans Dobbertin, the local historian who collected the sources of the legend in 1970 (Quellensammlung, no. 1). Christine Wulf, who edited Hamelin's inscriptions in 1989 for the Deutsche Inschriften, the standard German edition of medieval and early modern inscriptions, rejects it: "Dobbertins Datierung des Textes auf 'um 1300' ist abzulehnen, da sich kein Anhaltspunkt dafür finden läßt, daß diese Inschrift bereits auf dem älteren von Lossius erwähnten Bild gestanden hat" (DI 28, no. 76†), that is, nothing shows the inscription was already on the older picture that a writer named Lossius mentioned in 1571. The only window whose text is known is the one the mayor, Poppendieck, gave to the market church in 1572, destroyed in 1660; the older glass picture is known only from Lossius's Latin couplet of 1571. The church chronicle that Johann von Pohle wrote in Hamelin in 1384 says nothing. The claim that Hamelin's records of 1384 state "it is 100 years since our children left" cannot be traced to any scholarly or early-modern source, and Wulf notes that the few charters carrying a later-added note "post exitum puerorum", "after the children went out", never count the years. The surviving witnesses, in order of date, are these:
| Witness (date of the witness) | Year | Day | 130 | Piper's dress | Calvary / Koppen | Rats | Word for the lost; witness figure |
|---|---|---|---|---|---|---|---|
| Lüneburg, Ratsbücherei, Theol. 2° 25, fol. 268r, note added to Heinrich von Herford's Catena aurea (c. 1430–50). read from the facsimile | 1284 | Joh. et Pauli | ● | "omnino bene vestitus", silver pipe | ● "locum calvarie vel decollacionis" | ○ | pueri; the mother of Dean Johannes de Lüde (dean, will 1378) saw them go; "ista reperi in uno antiquo libro" |
| Rhymed Latin prayer in red ink in the Passionale of the Bonifatius-Stift (lost; 14th/15th c. claimed; known from Erich 1654 and Pastor Herr, 18th c.) | 1284 | Joh. et Pauli | ● | ○ no piper | ● "Calvaria sorpsit" | ○ | pueros; only text that blames "rei", the guilty |
| House inscription Markt 1 (DI 28 no. 40; 1525?, wording via Erich) | 1284 | 26 June | ● | ○ | ○ | ○ | Kinder "auß der Stadt verledet" |
| Neues Tor stone, 1556 (DI 28 no. 45; now Museum Hameln) | 1556−272 | — | ● | magus | ○ | ○ | puellos; the only monument that counts years since the exodus |
| Jobus Fincelius, Wunderzeichen (Jena 1556), first print | "vor 180 jaren" | 22 July | ○ | the Devil "hat gepfiffen" | "an ein berg" | ○ | "kneble vnd meidle"; a girl who followed at a distance; cites "das Stadbuch zu Hammel" |
| Zimmerische Chronik (c. 1559–66) | — | — | ○ | "landtfarer" with a "pfeifle" | "nechst beiligenden berg" | ● first | children "under acht oder neun jaren"; town dates letters "nach verlierung unserer kinder" |
| Johann Weyer, De praestigiis daemonum (3rd ed. 1566; 1568) | 1284 | 26 June | ● | "tibicen Omnicolor" | ● "Calvariae sub monte Koppen" | ● glires | pueri; "unus superstes" |
| Marktkirche window donated by Poppendieck, 1572 (DI 28 no. 76†; text via Erich; destroyed 1660) | — | Joh. und Pauli | lost | "ALLERLEI GE[KLEDET]" | ● | ○ | "GEBAREN … THO K[AL]VARIE … DEN KOPPEN" |
| Backhaus, Chronicon Hamelense rhythmicum (MS 1589) | 1284 | Joh. und Paul | ● | "bundten Kleidern" | ● "Das Halsgericht allda versteh" | ● first Hamelin text | "Kind"; Bungelosenstraße named |
| Mörsperg travel chronicle, watercolour (1592/93) | — | — | — | parti-coloured | — | ● drowning | oldest picture; not stated to copy the window |
| Rattenfängerhaus, Osterstraße 28, sill beam (DI 28 no. 107 D; 1602/03) | 1284 | 26 June | ● | "MIT ALLERLEI FARVE BEKLEDET" | ● "TO CALVARIE BI DEN KOPPEN" | ○ | "CXXX KINDER VERLEDET … VERLOREN" |
| Richard Verstegan, Restitution of Decayed Intelligence (1605) | 1376 | 22 July | ● | "Pi'd Piper" | "a little Hill" | ● | "Boys"; a lame boy left behind (Browning's source) |
Table 1. The witnesses in order of date, and which elements of the story each carries. The table was compiled for this report from the sources listed under "Primary" below; "DI 28" is Wulf's 1989 volume of Hamelin inscriptions. Dots: ● present, ○ absent; red ● marks the rat motif, which enters between 1559 and 1568 and is never on a Hamelin inscription. Every quotation and link is given in full in the sources dossier kept with this report.
The year, the feast day and the number are constant in the Hamelin witnesses from the earliest layer on. Since that layer is already 150 years after the event, their constancy shows only that the tradition was fixed by about 1430; it does not show that the three details are true (the "no single event" reading has to explain them away; the others take them as true). Every other element was added later. The silver pipe of the Lüneburg note becomes the coat of many colours on the inscriptions of 1525–1602, then the Devil (Fincelius, 1556), then a rat-catcher cheated of his fee (first in the Zimmern chronicle, a Swabian family history of about 1560, and in Weyer), then the lame boy left behind (Richard Verstegan, an English antiquary writing in 1605, whose version was Robert Browning's source). The rats are a south German wandering motif that fused with the Hamelin story in the 1560s, as the folklorist Willy Krogmann showed in 1934. An explanation therefore has to account for the exodus alone.
The note sits at the foot of leaf 268 (folio 268r) of a paper manuscript in the Lüneburg town library, a copy of the Catena aurea, a Latin compendium by the Dominican friar Heinrich von Herford of Minden (died 1370). It is written in a hand of the second quarter of the fifteenth century, among other local marvels (a tale of a basilisk in a Hamelin cesspit in 1347 comes next). I checked the transcription against the facsimile that the Herzog August Library in Wolfenbüttel has put online, and the text that circulates on Wikipedia is wrong in several places. There is no heading "Fabula exodi Hamelnsis" on the page; that is the cataloguer's title. There is no "fere" ("about") before the number; the syllable is the first half of the broken word fe-stulam. The scribe crossed out portam and wrote valvam. The note contains no Latin verses.
Notandum miraculum valde rarum quod accidit in opido Hamelen Mindensis diocesis anno Domini M°CC°LXXXIIII ipso die Johannis et Pauli. Quidam adolescens de XXX annis, pulcher et omnino bene vestitus, ita quod omnes videntes eum in persona eciam vestitum eius admirabantur, intravit per pontem et Wesere portam, habens argenteam festulam mire forme. Incepit festulare per totum opidum, et omnes pueri audientes illam festulam, in numero CXXX, sequebantur eum extra valvam orientalem, quasi ad locum Calvarie vel decollacionis, et evanuerunt et recesserunt, quod nullus potuit investigare ubi unus eorum remansit. Matres vero puerorum cucurrerunt de civitate ad civitatem et penitus nihil invenerunt. Unde "vox in Rama audita est" et quelibet mater deplangit filium suum. Et velut computatur per annos Domini vel per annum primum, secundum, tercium post iubileum, sic computaverunt in Hamelen per primum, secundum, tercium annum post exitum et recessum puerorum. Ista reperi in uno antiquo libro. Et mater domini Johannis de Lude decani vidit pueros recedentes.
A very rare marvel is to be noted, which happened in the town of Hamelin in the diocese of Minden in the year of the Lord 1284, on the very day of John and Paul. A young man of about thirty, handsome and altogether well dressed, so that all who saw him admired his person and even his clothing, came in over the bridge and by the Weser gate, carrying a silver pipe of wondrous form. He began to pipe through the whole town, and all the children who heard that pipe, 130 in number, followed him out through the east gate, as if to the place of Calvary, that is of beheading; and they vanished and went away, so that no one could find out where a single one of them remained. The children's mothers ran from town to town and found nothing whatever. Hence "a voice was heard in Ramah", and every mother bewails her son. And just as one reckons by the years of the Lord, or by the first, second, third year after a jubilee, so they reckoned in Hamelin by the first, second, third year after the going out and departure of the children. These things I found in an old book. And the mother of Lord Johannes de Lüde, the dean, saw the children going away.
Lüneburg, Ratsbücherei, Ms. Theol. 2° 25, fol. 268ra–rb. Facsimile: diglib.hab.de/mss/ed000070, image 00539. Catalogue: I. Fischer, Handschriften der Ratsbücherei Lüneburg 2 (1972), pp. 47–48. First publicised by Heinrich Spanuth (1936; edition in his dissertation, 1951, p. 16).
Three features of the text bear on what follows. The writer explains Calvaria himself: it is the locus decollacionis, the town's beheading place, and the children went "as if" towards it. The note records a search that failed, "from town to town"; it does not record a departure to a known destination. The chain of transmission is named: an "old book", and the mother of a dean. Johannes de Lüde was a real person, dean of St Boniface's, the collegiate church (one run by a body of canons) that was Hamelin's principal church, who left a will in 1378 and appears in Pohle's chronicle. If his mother watched the children leave as a girl of ten to fourteen, as Heinrich Spanuth, who first published the note in 1936, guessed, she was born around 1270–74 and her son held office in the middle of the fourteenth century, so the "old book" the Lüneburg writer copied was probably a record of the collegiate church from the dean's own lifetime. That chain is plausible, but it rests on inference; no fourteenth-century document survives to confirm it.
From 1525 Hamelin carved the story onto its buildings in a fixed wording; the so-called Rattenfängerhaus ("Rat-Catcher's House"), a councillor's house of 1602/03, repeats it on the beam under its upper storey. Wulf's edition gives that text without the "AM" that popular versions insert:
ANNO · 1284 · DAGE · IOHANNIS · ET · PAVLI · WAR DER · 26 IVNII · DORCH · EINEN · PIPER · MIT · ALLERLEI · FARVE · BEKLEDET · GEWESEN · CXXX · KINDER · VERLEDET BINNEN · HAMELEN · GEBON. TO CALVARIE · BI · DEN · KOPPEN · VERLOREN ·
"In the year 1284, [on the] day of John and Paul, which was the 26th of June, through a piper clad in all sorts of colours 130 children born within Hamelin were led astray; lost at Calvary by the Koppen." (GEBON is the carver's error for GEBOREN.)
DI 28, Hameln, no. 107 D (Christine Wulf), inschriften.net. Compare Markt 1, 1525 (no. 40 A): "Anno Christi 1284 am Tage Johannis Pauli ist gewesen der 26. Junij […] C und XXX. Kinder auß der Stadt verledet"; Neues Tor, 1556 (no. 45): "Centum ter denos cum magus ab urbe puellos / duxerat ante annos CCLXXII condita porta fuit"; Hochzeitshaus, 1610–17 (no. 111): "Nach Christi Geburt 1284 Jahr / gingen bei den Koppen unter Verwahr / hundert und dreissig Kinder in Hameln geboren / von einem Pfeiffer verführt und verloren".
The Passionale of St Boniface's, a service book of saints' lives now lost, carried the same date and number in a rhymed Latin prayer, which Samuel Erich printed in 1654 as "alte Knüttel- oder Münchsverse" (old doggerel, or monks' verses) and a Hamelin clergyman, Pastor Herr, copied again in the eighteenth century. It mentions no piper. The only agents are "the guilty" and the hill itself:
Post duo CC. mille, post octuaginta quaterque, / annus hic est ille, quo languet sexus uterque, / orbantis pueros centumque triginta Iohannis / et Pauli caros Hamelenses non sine damnis. / Fatur ut omnes eos vivos Calvaria sorpsit. / Christe tuere tuos [Herr: reos], ne tam mala res quibus obsit.
"After two hundred and a thousand, after eighty and four: this is the year in which either sex languishes, [the year] that on John and Paul's day robbed [Hamelin] of a hundred and thirty dear Hamelin children, not without losses. It is said that Calvary swallowed them all alive. Christ, protect thine own [Herr: the guilty], lest so evil a thing harm anyone."
S. Erich, Exodus Hamelensis (Hannover 1654), pp. 39–40 (MDZ bsb10019001); the reos variant after Meinardus 1882 via Gutch, Folk-Lore 3 (1892). Herr's prose gloss: "Anno 1284 in die Johannis et Pauli perdiderunt Hamelenses centum et triginta pueros, qui intraverunt montem Calvariam."
Two further documents tie the story to the town's own streets and records. The street the children are said to have left by was already "uppe der bungehelosenstrate" in a document of 16 September 1496 (published by Meinardus in the journal of the Lower Saxony historical society, Zeitschrift des Historischen Vereins für Niedersachsen, 1884): the "drumless street", where by custom no music is played. So the tradition was fixed to the town's streets before any surviving inscription. And Johann Weyer, who visited in 1567, wrote that he had seen the annals, the painted windows and the dating formula with his own eyes, "cuius rei sum testis oculatus" (of which I am an eyewitness). In writing, none of this is older than the 1440s.
Three incompatible theories depend on this phrase. Dobbertin read Koppen as Kopahn, a manor of the Knights Hospitaller on the Baltic coast of Pomerania near Rügenwalde (today Darłowo in Poland), and had the children drown there in a shipwreck. Waltraud Woeller, a folklorist writing in 1957–61, and Gernot Hüsam, a local historian who has written on the legend since 1990, read it as the Koppenberg or the Ith ridge at Coppenbrügge, thirteen kilometres east, and had the group die in a sinkhole or at a pagan Midsummer rite. Udolph, rightly, said that Koppe is just Kuppe, a hilltop, one of dozens of such names. A document printed in the Hamelin Urkundenbuch, the collection of the town's medieval charters, which as far as I can find none of these authors cites, settles where the town of about 1500 put the phrase. It is a letter of 6 July 1502 from the bishop of Minden, whose diocese Hamelin belonged to, consenting to a chapel. What it can and cannot say about 1284 comes after it.
… quidam in Christo fideles devoti certam capellam in honore illius dolorissimi transitus domini nostri Jesu Christi … gravissima via in eadem passi et facti usque ad montem Calvarie, ubi cruci affixus est … ante valvam orientalem opidi Hamelensis nostre diocesis cum signis et distanciis sicuti in pluribus aliis locis habetur ad limites ville parve Afferde, vulgariter beneden den Koppen, ex certa devotione in memoriam passionis Christi … dudum edificare et instaurare presumpsissent … Sed quia per … decanum et capitulum ecclesie Hamelensis … impediti … eandem capellam … a dicto loco ammovissent et iam de novo extra phalangas dicti opidi Hamelen vulgariter buten vor der lanthwer to Afferde pro parte in via publica versus Hildensem … erexerunt …
"… certain devout faithful had some time ago presumed to build a chapel in honour of the most sorrowful passage of our Lord Jesus Christ … along that most grievous way to Mount Calvary, where he was nailed to the cross … before the east gate of the town of Hamelin in our diocese, with markers and distances as is the custom in many other places, up to the boundary of the village of Klein-Afferde, commonly [called] 'beneath the Koppen' … But because the dean and chapter of the Hamelin church hindered them, they moved the chapel from that place and have now set it up anew outside the Landwehr of the town, 'buten vor der lanthwer to Afferde', partly on the public road towards Hildesheim …"
Bishop Heinrich of Minden consents to the chapel and grants a forty-day indulgence, 6 July 1502. Urkundenbuch des Stiftes und der Stadt Hameln II (1408–1576), ed. E. Fink, Hannover 1903, no. 663, p. 479 (archive.org bub_gb_gBs3AQAAMAAJ). Original in the Hauptstaatsarchiv Hannover, Dep. Hameln 405. I have not found this document used for the legend before
So by 1502 a Way of the Cross with measured stations ran from Hamelin's east gate to the field boundary of Klein-Afferde, a village just outside the town, at a place "commonly called beneath the Koppen", on the road to Hildesheim, the next city to the east. Samuel Erich put the "Calvarien- oder Kopffelberg" at the same place in 1654: "vor dem Oster Thor gegen Morgen … etwa ein Büchsen Schuß weit von der Stadt an der Heerstrassen … da das Halsgerichte stehet", before the east gate, a gunshot from the town on the highway, where the gallows stood, with two cross-stones in front of it. Backhaus, who wrote a rhymed chronicle of the town in 1589, explained "Vnter dem Berg Calvariae" as "das Halsgericht allda versteh" (understand the place of execution there), and the Lüneburg writer had already written locus Calvarie vel decollacionis. For every witness between the 1440s and 1654, then, the phrase meant the execution hill by the hilltops just outside the east gate. The order of that usage runs: a place called Calvary, glossed as the beheading place, in the Lüneburg note (c. 1430–50); a devotional Calvary with stations laid out there before 1502; the "Calvarie bi den Koppen" formula on houses from 1525.
What the charter can establish is where the phrase pointed when the inscriptions were carved. That is enough to rule out Kopahn and Coppenbrügge as the meaning of "Koppen" in the town's formula: the theories that derive a Pomeranian manor or the Ith ridge from the inscriptions lose their textual basis, because the same town, within a generation, used those words for a site a gunshot from its walls. It establishes nothing about 1284. It does not show that a group went that way, that the Calvary existed in 1284 (its earliest witness is the Lüneburg note, 150 years later, and the stations are a fifteenth-century devotion), or where anyone went afterwards. The "direction east" used in this report comes from the tradition's extra valvam orientalem; the charter does not supply it.
I test the explanations against the tradition as first attested; there are no established facts about 1284 to test them against. Each feature below carries its earliest witnesses, and an explanation has to account for it or explain it away. (1) A feast day in the year 1284: the Lüneburg note (c. 1430–50), the lost Passionale verses (undated) and house inscriptions from 1525. (2) A large, specific number of young people, 130, from a town of perhaps two thousand: the same three witnesses. (3) An orderly, voluntary departure behind a musician: the Lüneburg note, then every witness from 1556 on; the east gate is the Lüneburg note alone; that the day was a Monday two days after Midsummer is a computation from the feast, true only if the year and the feast are. (4) No return, and a search that found nothing: the Lüneburg note (the mothers, "from town to town") and, in another form, Fincelius in 1556 (the parents, "on water and land in all places"). (5) A memory framed as loss and mourning (perdiderunt, "they lost"; verloren, "lost"; "vox in Rama", the voice heard in Ramah from the Gospel's massacre of the innocents), with the guilty left unnamed: Lüneburg note, Passionale, inscriptions. (6) Nothing in the town's own documents for 150 years; this is a fact about what survives rather than a feature of the tradition. Only the "no single event" reading treats (1) to (5) as products of the tradition rather than as things that happened. The others take them as roughly true, which is the 80-in-100 judgement above in another form.
| Explanation | Proposed by | 1284 date | 130 young people | Musician, orderly exit | No return, failed search | What breaks it |
|---|---|---|---|---|---|---|
| Locator-led emigration to the colonial east | Wann 1949 (Moravia); Dobbertin 1955–96 (Pomerania, Prussia); Udolph 1997 (Prignitz, Uckermark) | ● | ● | ● | only if the group came to grief or the memory is the mothers' view | No document names a Hamelin settler; place names undated; a legal emigration should have been traceable (Hucker 1985) |
| Procession, pilgrimage, dancing "Kinderfahrt" | Britannica 1911; G. Spanuth 1958; Rumpf 1985; comparanda Erfurt 1237, Utrecht 1278, Aachen 1374 | ● Midsummer fits | plausible | ● | needs a distant disaster | Early texts describe leading away, not dancing; the well-attested epidemics begin 1374 |
| Accident near the town (Koppen, Ith) | Woeller 1957/61; Hüsam 1990–2024 | ● | ○ no mechanism, no bodies | partly | ○ the mothers searched other towns | The inscriptions' "Koppen" is the Afferde site (1502 charter), so the phrase gives Coppenbrügge no support; nothing found |
| Military levy or ruse | Fein 1749 (Sedemünder, 1259/60); Küntzel 2010 (Everstein siege, 1284) | Everstein yes; Sedemünder no | ● | ● fife and drum | ○ a battle 30 km away is known | Sedemünder is 1259/60; a public defeat is not remembered as a disappearance |
| Abduction, sect, murder | Manchester 1992; Hüsam 1990; Fincelius's Devil | ● | ○ 130 is too many | ● | ● | No evidence; Manchester misdates to 1484 |
| Children's crusade 1212; plague 1347+; Transylvania | popular since Leibniz (attribution unverified); Wolfers 1965; Grimm 1816 | ○ | — | — | — | Wrong century; no Lower Saxon origin traces in Transylvania (Kroner 1984) |
| No single event; aetiological legend | Krogmann 1934 (partly); Rostek-Lühmann 1995 | must explain the date | must explain 130 | — | ● | Does not explain a fixed feast day, a fixed number and a named eyewitness chain |
Table 2. Seven explanations scored against the tradition's features as attested between c. 1430 and 1602, not against established facts of 1284. The table was compiled for this report from the works listed under "Scholarship cited" below; the "Proposed by" column gives author and year of works listed there.
Emigration is the one mechanism that needs no special assumption, because recruiting young people and leading them away was a locator's ordinary business. A locator was a contractor who recruited settlers for a territorial lord: he advertised in market towns with music and fine clothes, gathered young unmarried people, equipped them and led them east in early summer, in time to build and sow winter rye. A party of 130 would fill one or two new villages of forty to sixty hides (a hide being one family's farm). If the tradition's date and gate are right, the date fits the recruiting season and the exit east fits the road to the settlement lands of the Altmark, Brandenburg and Pomerania, and pueri/kinder is the right word for the recruits. A substantial line of opinion since the 1950s (Dobbertin; Udolph; Norbert Humburg, editor of the 1985 volume of the Hamelin conference on the legend; Wolfgang Mieder, author of the English handbook on the legend of 2007; and the town's museum today) prefers this reading. The objection raised in 1985 by the medieval historian Bernd Ulrich Hucker, below, is the standing counter-case, and the accident, procession and levy readings each have their own advocates. It remains a reconstruction, because no source says that anyone recruited in Hamelin in 1284.
The documented channel is the family that held the land just east of the Koppen. In 1281 Count Moritz I of Spiegelberg and his sons Nicolaus, Moritz and Hermann acquired the lordship of Coppenbrügge on the road out of Hamelin (Calenberger Urkundenbuch, the charter book of the Calenberg region, Wunstorf nos. 29–31; Hermann was still a minor). The Pomeranian charter book then shows the same sons at the court of Bogislaw IV, duke of Pomerania, their kinsman through their mother:
Testes huius rei sunt nobilis vir Nicolaus comes de Speghelberg et fratres ipsius Mauricius et Hermanus et dominus Jo. de Scolentin et dominus Marquardus de Lübecin et dominus Hobe, Heinricus de Reno coquinarius, Jacobus de Guntersberg et Gothfridus Scriptor. Actum et datum Stettyn anno domini M.CC.LXXXIIII. VII. idus Julii.
"Witnesses of this are the noble man Nicolaus, count of Spiegelberg, and his brothers Moritz and Hermann, and Lord Johann von Schöningen [Scolentin] and Lord Marquard von Lübecin and Lord Hobe, Heinrich von Reno the master of the kitchen, Jakob von Güntersberg and Gottfried the scribe. Done and given at Stettin in the year of the Lord 1284, on the seventh day before the Ides of July [the editor heads it 8 July]."
Pommersches Urkundenbuch II.2 (1278–1286), ed. R. Prümers, Stettin 1885, no. 1308, p. 529 (Bogislaw IV grants six hides at Böck to St Mary's, Stettin); archive.org kpbc.umk.pl.Pomorze_062_04_23540. The register (PUB III.2, s.v. Spiegelberg) lists Nicolaus at Bogislaw's court from 11 November 1282 (Pyritz) through eight charters of 1283 ("Nicolao comite de Spegelberge, consanguineo nostro", Stettin, 26 May 1283) to this one, and Hermann as "domicellus" at Greifenberg on 25 March 1284. After July 1284 none of the three appears in the Pomeranian record again. I verified this in the printed edition
Dobbertin built his case on this charter, and it supports the existence of his channel better than his critics allow. Wikipedia's genealogy gives "Nicolaus I († 1283)", a misreading of "last attested 1283"; in fact the count was alive and in Stettin (the duke's seat, today Szczecin in Poland) twelve days after the tradition's date for the exodus, with both his brothers. The charter does not support the use Dobbertin made of it. A mounted count could ride the 470 kilometres from Hamelin to Stettin in twelve days. A column of 130 on foot, at 25 kilometres a day, could not have got there before mid-July. If Nicolaus recruited in the Weser valley in 1284, he did it through agents, and the "handsome young man of thirty" is at most his locator. Dobbertin's shipwreck at Kopahn on 22 July is a construction (the day is Verstegan's date for the event), and the medievalist Hans Eberhard Mayer's review of 1958 already called the whole scenario a temptation "die Klärung der Vergangenheit bis zur letzten Konsequenz und zum letzten Detail treiben zu wollen", to push the clearing-up of the past to its last consequence and its last detail. Contact between the Weser country and Pomerania in these years is documented. Hamelin's former lords, the counts of Everstein, had a Pomeranian branch at Naugard (Nowogard) from 1274; Duke Barnim I of Pomerania had given the church of Pyritz (Pyrzyce) to the convent of Wülfinghausen in the Weser country; the war between Brandenburg and Pomerania of the early 1280s ended in the peace of 13 August 1284; and 1284 was the first year after the great Prussian revolt against the Teutonic Order, the crusading order that ruled Prussia, when the Order's village-founding began in earnest. A recruiter working the Weser towns in June 1284 would have found lords wanting settlers. None of this places a Hamelin group on that road; it shows that the route was in use and that settlers were wanted.
The popular account misrepresents the place-name evidence. Udolph's one scholarly article (Niedersächsisches Jahrbuch 69, 1997, pp. 125–183) is an audit of place names, and most of it rejects earlier identifications. Nearly every equation with a village in Moravia (in today's Czech Republic) proposed by Wann in 1949 and by Spanuth fails (Hamlíkov is a Czech formation, not "Hamlingen"; Bruno, the Westphalian bishop of Olmütz under whom Westphalians had settled in Moravia, died in 1281 and the inflow stopped), and Dobbertin's list contains, in Udolph's words, "schlimme Fehler", bad mistakes. What survives is a dozen place names carried from the Weser and Leine rivers to the Prignitz and the Uckermark, two districts of Brandenburg, and to Pomerania. The best is Hamelspringe → Hammelspring near Templin (1375 Havelspryng, 1438 Hamelspringe); Beverungen → Beveringen near Pritzwalk (1368) rests on the sound point that -ingen names were no longer being coined. Udolph's own conclusion is careful: "Zur genauen Datierung des Auszugs der Hamelner Jungleute vermögen die Ortsnamen natürlich nichts beizutragen" (the place names can of course contribute nothing to the exact dating of the departure), and "Wenn es einen Rattenfänger gegeben haben sollte …, dann hätte er die Hamelner … nur nach Brandenburg, in die Prignitz, die Uckermark oder nach Pommern geführt" (if there was a rat-catcher, he could only have led the Hamelners to Brandenburg, the Prignitz, the Uckermark or Pomerania). The article contains no study of surname distributions. Its section on personal names lists medieval bynames only (a Lodekinus de Hamel in 1317; families called Quernhamel, from the older form of the town's name, in Riga in 1303, Danzig in 1357 and Frankfurt an der Oder in 1441–86), and Udolph himself thinks Hamel-type names are more likely nicknames from Hammel, a wether, than names of origin. The sentence that circulates in English, that he "entered all the known family names in the village at that time" into a computer and found them clustered in the Prignitz, comes from a 1998 article in TIME; his own website calls the episode an agency's exaggeration. Several of his "secure" transfers are attested before 1284 (Hindenburg near Prenzlau in 1269, Depenau/Tiefenau in 1250), the village networks of the Prignitz and Uckermark were essentially complete by 1300, and eleven of the twelve names come from elsewhere in the Weser hills, not from Hamelin. The names show the route in use, mostly by knightly families, for two generations either side of 1284; none places a party from Hamelin on it in 1284.
The fourth feature, the failed search, keeps the share at 42 rather than 70. Hucker put it plainly in 1985: given the density of trade and family contact between the Weser and the Baltic through the Hanseatic towns, "von einem spurlosen Verschwinden der Hamelner hätte keine Rede sein können" (there could have been no question of the Hamelners vanishing without trace), and a fruitless search by the mothers would be "ziemlich rätselhaft", rather puzzling. Dobbertin and Udolph reply that one disaster on the road or at sea would be enough to turn a departure into a catastrophe in the town's memory. That is possible but unproven, and the whole explanation depends on it. An emigration sponsored by a count of Coppenbrügge, kinsman of the duke of Pomerania, would have been known to the council that later carved "verledet" and "verloren" on its buildings. So either the group was lost, or the memory preserved is the mothers' and the church's rather than the council's. No document supports either.
A procession or "Kinderfahrt" (13%). The red-ink entry Erich saw in an old town book put the event on St John the Baptist's day, "wann sonsten das junge Volck, altem Gebrauch nach, eine sonderliche Johannis Freude zuhalten pfleget" (when the young people, by old custom, hold a special Midsummer celebration), and 26 June is two days after Midsummer. Young people leaving towns behind a musician are on record: the Erfurt children who danced to Arnstadt in 1237 (the medieval entry is uncited in Hecker's 1832 history of the dancing manias and I could not verify it), the two hundred said to have danced on a bridge in 1278, and the mass youth pilgrimages to Mont-Saint-Michel in 1456–59, in which Rhineland towns lost boys by the hundred. A departure like that, coming to grief far away, would explain the search, the mourning and the silence about who was to blame. Against this reading, every early Hamelin text has the children led rather than dancing, and the great dancing epidemics begin in 1374. This is the second-ranked event explanation, and it overlaps with the first, since a locator recruiting on a feast day would have looked much the same.
An accident close to home (9%). Woeller's sinkhole and Hüsam's massacre on the Ith put the end at Coppenbrügge, on the strength of the inscriptions' "Koppen". The 1502 charter shows that the town's phrase meant the site at Afferde, a gunshot from the walls, so the inscriptions no longer point to the Ith. An accident there is still possible on its own terms, but nothing supports it, a catastrophe at Afferde would have left bodies, and either way the mothers would not have searched other towns. Erich in 1654 did see "ein eingefallene Sincke, als ob es ein kleiner Erdfall were", a collapsed hollow like a small sinkhole, at the Koppelberg, which is probably how the motif of the mountain swallowing the children attached itself to the place.
A military levy (8%). The theory of Fein, the garrison preacher of 1749, that the story disguises the town's defeat at the battle of Sedemünder fails on the date, but 1284 was the year Duke Heinrich of Brunswick-Grubenhagen took both Everstein castles by siege, thirty kilometres away, and Thomas Küntzel has asked (in the Hamelin museum society's yearbook for 2010, pp. 112–127, which I could not obtain) whether the exodus was a ruse of war in that struggle. Fife and drum are how levies were raised, and pueri is a normal word for armed young men. Against it, a town that lost 130 men at a siege to which it was a party would have known where they fell.
Crime or sect (4%) is the sixteenth century's own reading (Fincelius's Devil, the Zimmern chronicle's landtfarer, a vagrant) and the twentieth's (the American historian William Manchester, 1992); 130 is too many for either. No single event (18%) deserves more weight than the specialists give it. The first text is a "miraculum", a marvel, copied 150 years later into a manuscript of local marvels, next to the basilisk tale; the town's charters, well preserved and edited up to 1407, show no trace at the time; Pohle's chronicle of 1384 is silent. It does not come first because legends do not usually carry a feast day, a number and the name of an eyewitness's son who can be found in the archives with a will.
The source corrections below can be checked. None is new evidence in the sense of an unpublished document; each is a new use or a re-reading of something printed or digitised. The reconstruction of what happened is inference from a fifteenth-century tradition; the explanations are ranked, and none is established.
a printed source not used beforeThe bishop of Minden's letter of 1502 (Urkundenbuch des Stiftes und der Stadt Hameln II, no. 663) documents a Calvary with stations "before the east gate … to the boundary of Klein-Afferde, commonly 'beneden den Koppen'". It fixes the town's phrase to a real, named site on the road east, agrees with Erich's location of 1654, and rules out Kopahn and Coppenbrügge as the meaning of the inscriptions. The finding concerns the geography of the later tradition; the charter cannot show that the group went that way or what became of it. I have not found it cited by Udolph (1997), by Alfred Cammann (writing in 1984), in Wulf's commentary in DI 28, or in any of the theory literature I could read. Hüsam's 1993 paper on "Calvaria" I could not obtain, so the claim of novelty is qualified.
checked against the originalCount Nicolaus of Spiegelberg was alive on 8 July 1284 and in Stettin with both brothers (Pommersches Urkundenbuch II, no. 1308). Dobbertin knew this; Wikipedia's "† 1283", repeated in recent summaries, is wrong. The same charter shows he could not have walked out of Hamelin with the column, and that the family disappears from the Pomeranian record after that month. This corrects a date in the literature and shows that a channel existed; no evidence shows that a Hamelin group used it. The Calenberg charters of 1274 and 1281 add that Hermann was a minor in 1281, which fits his title domicellus, "young lord", in 1284.
checked against the originalThe Lüneburg note, re-read from the facsimile: no heading, no "fere", valvam written over a cancelled portam, and no verses (the verses belong to the lost Passionale). The eyewitness chain runs through Dean Johannes de Lüde (will 1378); it is the only link the sources offer to a fourteenth-century record.
agrees with Wulf 1989The window "of about 1300" is Dobbertin's undocumented date; the oldest inscription is 1525; the "1384: 100 years since our children left" record cannot be traced to any scholarly or early-modern source, and no surviving charter counts years since the exodus.
a correctionThe modern-surname argument has no published basis. Udolph's article is a place-name audit with a short section on medieval bynames; the "computer" and "telephone book" claims come from press reports of 1998. Modern surname distributions could not date a 1284 cohort in any case: hereditary surnames were only forming, and the resettlements of 1945 reshaped Brandenburg's stock of names, as Udolph himself notes.
agrees with the specialistsEmigration stays the top-ranked explanation, which agrees with Dobbertin, Udolph, Humburg, Mieder and the Museum Hameln; I rank it first at 42 in 100 rather than treating it as settled, and I do not pick a destination. The arithmetic used is this: 26 June 1284 (Julian) was a Monday; Easter fell on 9 April; Hamelin to Stettin is about 470 km by road, to Pritzwalk in the Prignitz about 280, to Olomouc in Moravia about 780.
A charter, a locator's contract or a town book of about 1284–1295 naming settlers "de Hamelen" or "de Quernhamelen" in a new village in Pomerania, the Neumark (Brandenburg's lands east of the Oder), Prussia or Brandenburg would, on these weights, push emigration above 70 and fix the destination. The Pomeranian volumes for 1284–1300 contain no such entry; the Prussian, Brandenburg and Mecklenburg charter books and the earliest burgher books of Riga, Danzig, Elbing, Stettin and Prenzlau have not been searched systematically. A fourteenth-century Hamelin document mentioning the exodus, found among the collegiate church's records or in the "old book" the Lüneburg writer used, would settle whether it happened; its absence after a full search of the Hamelin records deposited in the Lower Saxony state archive in Hanover would push "no single event" up. A dated foundation in the north-east of about 1285–90 carrying a name specific to Hamelin (not Hamelspringe, a village that belongs to Bad Münder, the next town north-east of Hamelin) would connect Udolph's route to a dated event.
Archaeology at the Koppen or on the Ith turning up a mass grave of the 1280s would revive the accident theories; nothing of the kind has appeared in seven centuries of quarrying, railway and industrial building on the Afferde hills. And Hartmann's 1941 genealogy of the Spiegelberg counts, which I could not obtain, may record Nicolaus after July 1284; if he reappears in Saxony in 1285, the reading in which he himself led the group to Pomerania fails, though the channel remains possible.
I could not get hold of: Dobbertin's Quellensammlung itself (1970/1996); Hartmann, "Die Grafen von Poppenburg-Spiegelberg" (1941); Hüsam's papers of 1993 and 2024; Küntzel (2010); Meinardus (1882); Wauer's name-book of the Uckermark (2000); Bichlmeier's review of 2024; volume I of the Hamelin Urkundenbuch (to 1407), so I could not attempt a list of the members of the 1284 council; and the medieval entry for Erfurt in 1237. The Brandenburg land register of 1375 (the Landbuch) was available only as a scan too poor to search for names. I found no new document from the thirteenth or fourteenth century. The probabilities are my judgement of the tradition's features; anyone who can read the Hamelin records in the Hanover archive should revise them. The narrative of the event here is the tradition's narrative, and the reconstruction built on it is speculative in the plain sense that no source describes a recruitment, a route or a destination.
README.md (the verdict in brief) · sources_dossier.md (every witness, quoted and translated, with links) · onomastics_review.md (Udolph's evidence audited) · theories_context.md (every explanation, Hamelin in 1284, the eastern settlement in 1284) · PUB_Spiegelberg_extracts.md (the Pomeranian charters) · UB_Hameln_II_663_Calvary_Afferde_1502.md (the 1502 charter, with what it can and cannot establish) · computations.md and compute.py (weekday, Easter, distances)
