Examine the evidence
Was King Arthur real?
What the early texts and associated places can establish and where the evidence stops.

Reviewed 18 October 2026.
The careful answer
A historical Arthur has not been securely identified by the evidence reviewed here. There may have been a person, or memories of several people, behind parts of the tradition; that possibility is not the same as a demonstrated biography. We can be much more confident about the existence and development of the stories.
The useful question is not simply whether a source is old. Ask how close it is to the events it describes, what kind of work it is, whether it names Arthur, and whether it depends on another account.
| Source | What it supplies | What it cannot establish alone |
|---|---|---|
| Gildas, sixth century | A moral account of Britain’s troubles, naming Ambrosius Aurelianus and describing the siege of Badon. | Gildas does not name Arthur. His silence neither identifies an Arthur nor proves that no such person existed. |
| Historia Brittonum, ninth century | A battle list attached to Arthur and separate marvels associated with him. | A later compilation with wondrous material is not a contemporary campaign record. |
| Annales Cambriae | Brief notices connecting Arthur with Badon and with Medraut at Camlann. | The familiar dates 516 and 537 are labels in a later annalistic tradition, not proof that those entries were written during the events. |
| Geoffrey of Monmouth, twelfth century | An expansive account of Arthur’s family, wars, court and departure. | A developed literary history, written many centuries later, is not independent confirmation of each event. |
| Medieval romances, including Malory | Detailed plots involving the court, swords, lovers and quests. | Narrative detail is not evidence that the characters’ actions happened as described. |
Badon and Camlann need particularly careful reading
Gildas’s Badon passage does not identify its commander as Arthur. The later sources do connect Arthur with battle, but that is an additional claim to evaluate. The Camlann annal says that Arthur and Medraut fell; it does not describe their relationship or say they killed each other. Malory’s father-and-son conflict cannot fill those gaps as though it were an explanatory footnote written at the time. Gildas §§25–26; Historia Brittonum §56; Annales Cambriae.
What archaeology changes and what it does not
At Tintagel, early medieval settlement and imported goods illuminate a real society. They do not identify Arthur. At Glastonbury, the 1191 grave claim belongs to the medieval history of the abbey and the legend. Neither place supplies a securely identified personal record of the king. The individual place guides set out the evidence and its limits.
A monument can be real while its association with a named hero is later. Conversely, a story’s later development does not make the physical history of its setting unimportant.
What would strengthen an argument?
A proposed identification needs evidence with a secure date, provenance and context, a convincing explanation of the name, and an account of why competing explanations are weaker. Repeating a claim across websites is not multiple independent confirmation. Nor can a resemblance between names do all the work.
This is an evidence guide for the sources actually examined, not a verdict on every proposed Arthurian candidate. Welsh poetic and triadic material, manuscript transmission and the arguments of modern specialists need further dedicated treatment. The book list marks further reading that has been located but not yet examined.
Explore the annotated books · Follow the source register · Return to Arthur’s figure guide
Sources for this guide
- Gildas, De excidio Britanniae (On the Ruin of Britain). Sixth-century work; nineteenth-century English translation at Fordham. Sections 25–26. Ambrosius and Badon; absence of the name Arthur. Read source · About this book
- Anonymous compiler, traditionally attributed to Nennius, Historia Brittonum (History of the Britons). Ninth-century compilation; Fordham transcription of Giles (1848), with §§56 and 73 supplied from the Camelot Project. Sections 56 and 73. Later battle list and marvels. Read source · About this book
- Anonymous annalists, Annales Cambriae (Annals of Wales). Medieval annalistic compilation; English transcription at Fordham. Entries conventionally dated 516 and 537. Limits of the brief notices. Read source · About this book
- Geoffrey of Monmouth, History of the Kings of Britain, Book XI. Composed c.1136–38; Giles collection (1848). Book XI, chapters 1–2. Later developed account of Arthur’s final conflict. Read source · About this book
Research coverage
This guide compares the named translated passages and links to the separately sourced place guides. It does not claim a comprehensive survey of Arthurian historical scholarship.
The Myths · Arthurian collection · Editorial approach
