# Brunanburh (937): the northern candidates and the full field

Dossier on (1) Burnswark, (2) Lanchester, (3) every other proposed site, (4) the Scottish/Cumbrian route evidence, with a closing balance and master table.

**Verification conventions used below.** `[V]` = read directly in the cited text (OCR of a scan, publisher page, open-access PDF/HTML, or the author's own blog post/comment). `[V-via X]` = the claim about author A is verified only through a quotation or summary in source X (page numbers given are those X supplies). `[U]` = could not be verified; treat as provisional. Paywalled items that I could not open are marked as such. Not every web search could be run, so a few minor points rely on texts already downloaded rather than fresh searches; these are flagged.

---

## 0. The name-evidence everyone argues from (reference table)

Compiled from Cavill's table in "The Place-Name Debate" (Livingston, *Casebook*, 2011, pp. 327-49; the open PDF at https://www.nottingham.ac.uk/-sczsteve/09essay-cavill.pdf is a pre-publication TEAMS/Kalamazoo pagination, pp. 311 ff.) `[V]` and from Neilson's 1909 table `[V]`.

| Source | Form | Note |
|---|---|---|
| ASC A, B, C, D (poem, l. 5 *ymbe Brunanburh*) | *Brunanburh* (A, C: *Brunnanburh*) | accusative after *ymbe* |
| ASC E, F | *Brunanbyrig*, *Brunanbyri* | dative |
| Æthelweard, *Chronicon* | *in loco Brunandune* | dative; the only early *-dun* form with *Brun-* |
| Symeon of Durham, *Libellus de exordio* ii.18 (Arnold, *Symeonis opera* i, 76) | "apud Weondune, quod alio nomine Etbrunnanwerc vel Brunnanbyrig appellatur" `[V: Rolls Series OCR]` | *Et-* = OE *æt* 'at' (so Cavill; Halloran 2010 accepts this) |
| *Historia Regum* s.a. 937 (Arnold ii, 93) | "Ethelstanus rex apud Wendune pugnavit, regemque Onlafum cum DC. et XV. navibus..." `[V: OCR]` | 615 ships |
| William of Malmesbury | *Brunefeld*, *Bruneford* (MS variants) | |
| John of Worcester | *Brunanburh*; fleet "entered the mouth of the Humber" | sole root of the Humber tradition (Cavill; Livingston) |
| Henry of Huntingdon | *Brunebirih*; Arundel 48 MS *Brunesburi(t)h* (Rolls ed. printed *Brunesburh*) | the *-s-* forms are late/idiosyncratic (Cavill 2011 §5.3) |
| Gaimar, *Estoire des Engleis* l. 3522 | *Bruneswerce* (one 14th-c. MS), *Brunewerche*, *Burneweste* (x2) | |
| Chronicle of the Kings of Alba | "bellum Duinbrunde" | Gaelic *dún* = OE *burh* (Campbell 1938, cited Cavill) |
| Annales Cambriae / Brut y Tywysogion | *Brune* / *Brun* | |
| Welsh poetry (Glaswawt Taliessin; Canu y Dewi) | *kat tybrunawc*; *Cad Dybrunawc* | |
| Annals of Clonmacnoise (s.a. 931 = 937) | "the plaines of othlyn(n)" | |
| Egils saga | *Vínheiðr*, *Vínuskógr* | |
| 12th-c. charters / Scottish tradition | *Bruningafeld*, *Brounyngfelde* | |
| The poem, l. 54 | *on Dingesmere* (O: *dinnesmere*) | the sea/mere over which the Norse fled to Dublin |

---

## 1. BURNSWARK (Birrenswark), Dumfriesshire

### 1.1 The site
An Iron Age hillfort on a flat-topped hill (c. 280 m / 920 ft) 3.5 miles north of Ecclefechan in Annandale, with two Roman camps on the north and south slopes and a possible fortlet; the hill was bombarded by Roman artillery c. AD 140 (Wikipedia, "Burnswark Hill", https://en.wikipedia.org/wiki/Burnswark_Hill `[V]`). Excavated by Christison, Barbour and Anderson, *PSAS* 33 (1898-9) (Neilson 1909, p. 51 `[V]`). Wikishire notes that "no archaeological evidence has been uncovered of a tenth century battle here" (https://wikishire.co.uk/wiki/Burnswark `[V]`) and the Battlefields Trust adds "there is no archaeological evidence of reoccupation of the forts during the Dark Ages / Early Middle Ages" (https://www.battlefieldstrust.com/resource-centre/battlepageview.asp?pageid=891 `[V]`).

### 1.2 George Neilson (1899; 1909; 1910)
- **Neilson, *Annals of the Solway until A.D. 1307* (Glasgow, 1899), p. 34**: first statement of the case (self-cited in 1909, n. 5) `[V]`.
- **Neilson, "Brunanburh and Burnswork", *Scottish Historical Review* 7, no. 25 (Oct. 1909), 37-55** (plan and sections of the hill, pp. 48-50). Full OCR: https://archive.org/download/ScottishHistoricalReview7/Scottish_Historical_Review_7_djvu.txt `[V]`. His argument:
  - Three elements of evidence: "the name of the place ... the mention or absence of mention of the Humber, and ... the account of Anlaf's flight" (p. 38). A table of the name-forms (p. 38) reproduced above.
  - "The mention of the Humber as the place of entry into England is a pure error of Florence of Worcester, repeated by some later writers after him. It is negatived ... by the natural inference of a West Coast junction spot for Irish, Cumbrians, and Scots ... and finally, the fact that the routed Norsemen fled back by sea to Dublin is conclusive proof that the site of battle was in the West" (pp. 38-9).
  - The one "distinct and identifiable statement of locality" is the Beverley *Alia Miracula S. Johannis* (c. 1175; a revision of William Ketell's c. 1150 narrative) which places Athelstan's crossing "in finibus ... flumen quod dicitur Scotorum Vadum" — the Solway ("Scottiswath") — so the battle was north of the Solway (p. 39).
  - Symeon's *Brunanwerch* and Gaimar's *Bruneswerc* "emphatically" point to Burnswark; supported by Hodgkin, *Political History of England* i (1906), p. 335 (p. 39 and nn. 3-7).
  - Early forms he prints (p. 39 n. 6): **Burnyswarke (1542)**, Burniswork (1608), Burneswark (1623), Burnswark (1661).
  - Eight topographical criteria drawn from Egils saga (rendezvous on the Roman road; unmistakable hill; flat summit; a *borg* on each slope, both Roman; steeper south camp; a burial mound on the plateau) which he says Burnswark satisfies (pp. 49-53); "Brunan burh and Brunan werch, plainly doublet forms, alike attest the borg" (p. 51 n.).
  - Minsca (Minscaw 1680) three miles away ≈ *Vínuskógr*; Hazelberry and Shawhill ≈ the saga's hazel-poles and wood (p. 52).
  - "The 'plains of Othlyn' have not been identified, but Burnswork stands in and looks down upon the flat riverside parish of Hoddom, anciently called Hodelme or Holdelm ... Othlyn is quite near enough to Hodelme for an Irish annalist" (p. 54 n. 1).
  - He dismisses the English sites as standing "on that solitary leg, the mere place-name": "Brunedown in Devonshire, Brumby near Doncaster, Lincolnshire, Burnham in North Lincolnshire, Bourne in South Lincolnshire, and Aldborough ... are all on the wrong side of the country. Bromborough on the Mersey is far away from the Solway. Bromfield (formerly Brunefeld) near Wigton in Cumberland, alone lies in the Solway region. But it is on the south" (p. 49).
- **The SHR rejoinders (vol. 7, 1910)** — not by W. S. Angus; they are:
  - **James Wilson, "The Site of the Battle of Brunanburh", SHR 7: 212-14** `[V]`: grants that "the Irish participation ... requires a site on the western coast" and that "the stage suits the play", but treats the saga/Beverley material as folk-tradition ("I take a desponding view of their historical value"); as a *reductio* he offers Burwen Castle, Elslack (Ribble valley, Yorkshire border) as a site that would satisfy Neilson's criteria equally, adding "I do not believe in Burwen Castle any more than in Burnswork or Aldborough"; "the mere jingle of place-names is as treacherous as a friar's lantern".
  - **Alice Law, "Burnswork and Brunanburh", SHR 7: 431-5, with Neilson's note, pp. 435-6** `[V]`: (i) the Beverley story belongs to the 934 Scottish expedition, not 937 (citing Earle: Athelstan's gifts to Beverley and St Cuthbert "belong to the campaign, not of Brunanburh, but of 934"); (ii) Egils saga's two *borgs* were "some hours of riding apart", whereas Burnswark's two camps are "just about 600 yards" apart — "Dr. Neilson ... must either discard his own theory of Burnswork, or discard the 'trustworthy' evidence of Egil"; (iii) strategically "absurd to suppose that the battle ... was fought almost at the very spot of supposed junction": the coalition "were invading England ... they would naturally march southwards", Malmesbury says Anlaf "multum in Angliam processerat", and the ASC has Constantine flee "to his country North", "which he would not have needed to do had he been already on his own side of the Solway". Neilson's reply concedes nothing, notes that Hodgkin's support "goes back to 1885", W. H. Stevenson's to 1891, and that "the latest advocate ... appears to be Professor Oman (see Athenaeum, 18th June [1910])".
- Support in the early 20th c.: Hodgkin (1885/1906), Stevenson (1891), Oman (1910) `[V-via Neilson 1910]`; Wikipedia adds that Burnswark "was the leading theory in the early 1900s" (https://en.wikipedia.org/wiki/Battle_of_Brunanburh `[V]`). Tudsbery (1907) already reports "Solwayside folk" holding the view (Tudsbery, *Brunanburgh A.D. 937*, OCR https://archive.org/download/brunanburghad93700tudsuoft/brunanburghad93700tudsuoft_djvu.txt `[V]`). Alistair Campbell's 1938 edition (site discussion pp. 57-80) rejected all localisations: "an honest *nescio* is greatly to be preferred to ambitious localisations built upon sand" `[V-via Wikishire; page unverified]`.

### 1.3 Kevin Halloran, "The Brunanburh Campaign: A Reappraisal", SHR 84:2 (Oct. 2005), 133-48 (DOI 10.3366/shr.2005.84.2.133) — paywalled; abstract and mediated content
- **Abstract (publisher/Crossref)** `[V]`: "Modern accounts of the battle of Brunanburh have generally suggested a location in the Northumbrian-Mercian borderlands east or west of the Pennines, a conclusion based in part on analysis of the aims and strategy of Anlaf Guthfrithson, Viking king of Dublin. This article re-examines the political dynamics of the coalition against Athelstan, taking account of the territorial and political ambitions of the kings of Alba and Strathclyde, and proposes a radically different interpretation of the campaign of 937. It also questions the reliability of the variant form Brunanburh as a guide to the battle's location and concludes that the most likely site was Burnswark in Annandale."
- **What the article argues, so far as recoverable:**
  - *Strategic aim.* The war of 937 is read as a consequence of Athelstan's annexation of Northumbria (927) and invasion of Scotland (934), and the coalition's objectives are those of Alba and Strathclyde rather than Anlaf's supposed bid for York; hence a battlefield in the Solway borderland, the zone of Strathclyde/Cumbrian expansion, rather than deep in England. `[V-via abstract; via Clarkson, Senchus 2009 post and Jan. 2010 comment; via Halloran's own comments on Senchus 2012]` (https://senchus.wordpress.com/2009/04/30/brunanburh-and-burnswark/ ; https://senchus.wordpress.com/2012/08/29/more-brunanburh-links/). In 2005 he told *The Herald* (24 Oct. 2005) that the 30+ English sites "are based on mistaken assumptions about both the course of the campaign and the place-name evidence" (via HNN, https://www.hnn.us/article/david-ross-famous-english-battle-site-brunanburh-p `[V]`).
  - *Name-forms.* "There is no doubt that early forms of the place-name Burnswark would fit with both Etbrunnanwerc and Bruneswerce" (Halloran 2005, pp. 145-6, quoted in Cavill 2008, p. 310, and Cavill 2011, p. 324 n. 43 `[V-via Cavill]`). The *burh* form is treated as "a variant of the real place-name that was poetically more satisfying", "a poetic paraphrase of the *werc(e)* original" (quoted Cavill 2011, §3.2 `[V-via Cavill]`).
  - *Æthelweard's Brunandun / Symeon's Wendun = the hill.* "Another argument put forward by Halloran ... is that the *dun* referred to in Æthelweard's *Brunandun* and Symeon's *Wendun* associates the battle site with 'a prominent hill'" (Halloran 2005, pp. 144, 147; Cavill 2011, p. 324 `[V-via Cavill]`).
  - *Symeon's phrase.* In 2005 (p. 145) he translated "apud Weondune quod alio nomine Etbrunnanwerc uel Brunnanbyrig appellatur" as "at Weondune which is otherwise named Etbrunnanwerc or called Brunnanbyrig", implying three distinct features; he withdrew this in 2010 (Cavill 2014, p. 100 `[V-via Cavill]`).
  - *Roman camps.* The two Roman camps and the summit fort are the *werc* and *burh* (developed in 2010, below).
  - *Dingesmere.* In 2005 read as a descriptive phrase, "sea of storms" (Halloran's own comment, 18 Mar. 2010, on Senchus `[V]`).
  - *Military/tactical reconstruction* at pp. 138-40 (Livingston's n. 59 targets these pages) `[V-via Cavill 2014]`.
  - *Humber tradition.* Rejected (Neilson's line; Clarkson's summaries). **The specific claim that Halloran attributes John of Worcester's Humber entry to confusion with 1066 could not be verified** `[U]`. The 1066/1069 comparison is documented elsewhere: Woolf 2007 ("This is fundamentally unlikely, since he came from Dublin ... [unlike the fleets] of Harald of Norway (1066) and Sven of Denmark (1069) which did indeed make landfall in the Humber", p. c.169-70, see §4.2 `[V]`) and Cavill, "The Yorkshire Hypothesis" (John's 937 entry "is very similar in expression to the Humber entry of Tostig in 1066", n. 30 `[V]`).
  - *Othlynn.* Halloran's treatment not recovered `[U]`; Neilson's Hoddom identification (above) is the Burnswark school's standard answer.
- Popular restatement: Halloran, "Brunanburh Reconsidered", *History Today* 56:6 (June 2006) (paywalled; teaser: "the claims of the Wirral ... may be misplaced", https://www.historytoday.com/archive/brunanburh-reconsidered `[V]`).

### 1.4 Halloran, "The Identity of Etbrunnanwerc", SHR 89:2 (Oct. 2010), 248-53 (DOI 10.3366/shr.2010.0207) — paywalled; content via Cavill 2014 (which cites page numbers) and Halloran's own 2012 blog comments
Reference list from Crossref `[V]`: Arnold's Rolls Series Symeon; Dodgson (EPNS Cheshire); Simon Taylor, "Generic-element variation, with special reference to eastern Scotland", *Nomina* 20 (1997), 5-22; Christison, *PSAS* 33 (1899); Cavill, *JEPNS* (2003/4). The argument:
- **The "Et".** He now accepts the translation "at Weondune which is called by another name AetBrunnanwerc or Brunnabyrig" (2010, p. 248 n. 3) — i.e. *Et-* is the Old English preposition *æt* fossilised onto the name, not part of it. **No argument about a Gaelic or Latin preposition was found**; the "Celtic" component of the article is elsewhere (next point). `[V-via Cavill 2014, p. 100]`
- **A Cumbric/Brittonic first element.** Following Alan James's *Brittonic Language in the Old North* (s.v. *brïnn* 'hill'), he proposes that Burnswark's first element is a "Brit/Pict" simplex *brun[n]* 'hill', "taken up by Northumbrian English speakers to form Brunes-weorc" (2010, p. 251 and n. 16) `[V-via Cavill 2014, p. 101]`. In his own words (Senchus, 27 Sept. 2012): "I suggest the 'wen' form is the OE Northumbrian name and the 'brun' forms are a composite derived from an original Cumbrian simplex form 'brun' with generic OE elements added by later English settlers into the Solway area" `[V]`; and (17 Sept. 2012) "my argument in 'Etbrunnanwerc' regarding the 'wen' (OE) and 'brun' (Cumbric) forms as synonymous and meaning 'a swelling'" `[V]`. Hence *Wendune* ("Wendune rather than Weondune is the correct form and refers to the hill of Burnswark", 2010, p. 250 `[V-via Cavill]`) = 'wen (swelling)-hill' = Burnswark, whose "wen-like appearance was striking" from the road (Senchus 2012 `[V]`).
- **burh vs werc = two features.** The variants "represent the alterations of the name from Brunnanbyrig to Etbrunnanwerc" (p. 249); "the *burh* element refers to the fort on the summit at Burnswark and the *werc* element to the earthworks of the Roman camps on the slopes" `[V-via Cavill 2014, p. 100]`; the Roman camps "were not recognised as such until the 19th century and hence always simply referred to as 'works'. OE 'werce'" (Senchus 2012 `[V]`).
- **Generic-element variation (Taylor 1997).** Used to argue that *dune/burh/werce/feld* can all attach to one site, and that "'paraphrasing' or name substitution invariably involves the substitution of a rare element by a common one ... Therefore ... if the 'werce' (rare) and 'burh' (common) forms were paraphrases then the 'werce' form would almost certainly be the correct original. This destroys the theoretical foundation of the Dodgson/Cavill argument that Brunnanwerc was merely a paraphrase of Brunanburh" (Senchus, 28 Sept. 2012 `[V]`).
- **Dunbrunde.** The Chronicle of the Kings of Alba's form is taken as evidence that "the place had a Gaelic name and an English name" — implying a borderland location (Senchus 2012 `[V]`; cf. the article, p. 252 `[U]`).
- **Dingesmere.** Reinterpreted as a proper name, "Sea of Noise" (Solway), positing OE *\*dinne*/*\*dynge* 'storm/noise' (2010, p. 253 n. 24) and reading *nægledcnearrum* as "in" rather than "to" the nailed ships `[V-via Cavill 2014, pp. 100, 105-6; Halloran comment 18 Mar. 2010 [V]]`.
- **Method.** The onomastic approach "relies too much on an analysis of forms that derive variously from copied, altered, difficult to read, and conflicting sources of uncertain provenance" (pp. 252-3 `[V-via Cavill 2022, n. 6]`); Cavill's burn+wark etymology is "unfeasibly late" (p. 252); he cites Sir John Clerk's *Brunswork* (1730) against the *Burn-* spellings (p. 252) `[V-via Cavill 2014]`.

### 1.5 Halloran's later work
- "Welsh kings at the English court, 928-956", *Welsh History Review* 25 (2011), 297-313 `[V: cited on Senchus 2011]`.
- Online papers on Academia.edu (2012): on Athelstan's 934 invasion; "Brunanburh: the defeat of empire" (Senchus, 29 Aug. & 19 Oct. 2012 `[V]`); "Brunanburh: Causes, War Aims and Strategy" (https://www.academia.edu/63944009/ — page blocked; date `[U]`).
- "Anlaf Guthfrithson at York: A Non-existent Kingship?", *Northern History* 50:2 (2013), 180-5 (DOI 10.1179/0078172X13Z.00000000042): argues that the Anlaf who took York in 939 was Anlaf Cuaran, and that Anlaf Guthfrithson's York kingship is a modern conflation `[V: publisher abstract via search]`.
- **No peer-reviewed publication by Halloran on Brunanburh after 2013 was found** (Cavill 2022 and 2023 cite nothing later than SHR 2010) `[V-negative]`.

### 1.6 Counter-arguments to Burnswark
1. **Philology (Cavill).** "The case for Burnswark was first made seriously by Neilson" — Cavill, "The Site of the Battle of Brunanburh: Manuscripts and Maps, Grammar and Geography", in *A Commodity of Good Names: Essays in Honour of Margaret Gelling*, ed. Padel & Parsons (Donington, 2008), pp. 303-19, Burnswark at pp. 310-11 `[V: OL search-inside snippets, ia commodityofgoodn0000unse]`: "the earliest known spelling is from 1542, Burnyswarke; Neilson also lists Burniswork (1608), Burneswark (1623), Burnswark (1661)"; "the similarity of the one manuscript spelling of Bruneswerce in Gaimar" is all the case rests on. In the *Casebook* (2011, §5.1) `[V]`: "there are no known spellings of the place-name Burnswark within 600 years of the battle, there is no philological 'fit' in the forms ... the best explanation for the first element of the Dumfries name is *burn* [a spring, a stream], in possessive and/or plural form ... None of the recorded spellings for Burnswark suggest that it originally had a weak Old English noun in the first element"; *dun* is "not used 'to describe great prehistoric hill-forts'" and *dun* names are "significantly lacking in ... south-west Scotland"; and the poem's *ealgian* implies "Brunanburh was English, was part of the English king's patrimony in a way that the Solway region never was at this time". In "Battlefield Despatches" (Harding, Griffiths & Royles eds, *In Search of Vikings*, CRC 2014, pp. 95-108; PDF https://www.nottingham.ac.uk/-sczsteve/Cavill_2014.pdf `[V]`): the 2010 article "simply reasserts views that fail to engage with the evidence"; the equation "Etbrunnanwerc = Bruneswerce = Burnswark depends simply on the notion that the names look a bit similar"; the *\*brunn* etymology "is a red herring"; an alternative derivation from *Birrens* (< OE *burgæsn* 'cairn', the name of the Roman fort Blatobulgium) + *wark* is offered; conclusion: "More thorough examination of the onomastic evidence relating to Burnswark makes any connection with Brunanburh vanishingly remote". Cavill 2022 ("Spellings of Brunanburh Revisited", *Northern History* 59:2, 162-76, n. 13) lists Burnswark among places "where the name is never actually recorded" `[V]`.
2. **Military/strategic (Livingston; Law; Battlefields Trust; Foot).** Livingston, *Casebook* p. 19 n. 59: Halloran's military-tactical interpretation "is nowhere as secure as he suggests" `[V-via Cavill 2014, p. 104]`. Alice Law's 1910 points (§1.2) remain the classic strategic objection: the invaders were aggressors on English soil and Constantine fled *north*. Battlefields Trust: "unlikely the battle was fought this far north, especially as Anlaf's objective had been to regain York" `[V]`. (Foot, *Æthelstan* 2011, pp. 172-9 discusses the candidates; not consulted directly `[U]`.)
3. **Would Athelstan have campaigned that far north after 934?** This objection is implicit in the "English soil" argument (Cavill) and in Law; Halloran's reply is that the site was the coalition's rendezvous and the war was about the north, and that the Scots left by land (poem offers no ships for Constantine) — Clarkson, Jan. 2010 comment `[V]`.
4. **Distance from Dublin.** Not in fact an objection (the Solway is a short sail); the real "distance" objection is Halloran's own, turned against Bromborough: that Scots and Cumbrians could not operate so far south "as late in the year as Brunanburh (probably October)" (Halloran, Senchus, 18 Mar. 2010 `[V]`). Livingston's counter is that a Dublin fleet is more plausible on a straight Irish Sea crossing to the Wirral than anywhere east (https://www.michaellivingston.com/the-search-for-brunanburh-a-summary/ `[V]`).
5. **Woolf 2007, p. 171** `[V: OL search-inside]`: "One Scottish site has been suggested, Burnswark, an Iron Age hill-fort between Lockerbie and Ecclefechan, and this hypothesis has recently been revived [n. 9: Halloran 2005]. It must be said, however, that the evidence is ..." (snippet ends; the sense is dismissive).
6. **Clarkson's trajectory** (author of *Strathclyde and the Anglo-Saxons*): 2009 — Cavill's analysis has "effectively swept aside" Burnswark; Jan. 2010 — after discussion with Halloran, "Kevin's suggestion of Burnswark requires no complicated logistical theories and therefore has much to commend it"; Oct. 2010 — "more likely that the battle took place within the bounds of Strathclyde ... anywhere from Penrith northwards"; 2014-15 — Amounderness (§4.1). All `[V]` (Senchus posts/comments).
7. **Halloran's own reservation**: "if my argument in 'Etbrunnanwerc' ... was correct (big if)" (Senchus, 17 Sept. 2012 `[V]`).

---

## 2. LANCHESTER (Longovicium), County Durham

### 2.1 The site and its antecedents
The Roman fort of Longovicium stands above the point where Dere Street crosses the River Browney (Wikipedia; Battlefields Trust `[V]`). A Browney location is much older than Breeze: Bosworth (1848) glossed the battle as fought "on the plain between the river Tyne and the Browney" (Momentous Britain list, s.v. Brancepeth; also visible in J. L. Hall's 1902 *Judith, Phoenix* notes `[V: OL snippet]`), and Momentous Britain records Edwin Guest (1838) as proposing "Browney" (reference known only second-hand `[U]`). Ekwall's *Concise Oxford Dictionary of English Place-Names* gives the Browney as "originally OE Brūne 'brown river'" `[V: OL snippet]` — the premise on which Breeze builds.

### 2.2 Breeze's publications on Brunanburh
- "The Battle of Brunanburh and Welsh Tradition", *Neophilologus* 83 (1999), 479-82: "Brun(e) was a place-name known to the Welsh"; "the English name Bruna or Brune was presumably understood by the Welsh as meaning 'stream'" `[V-via Cavill 2014, p. 102]`.
- Lecture, "Brunanburh in 937: Bromborough or Lanchester?", Society of Antiquaries of London, 4 Dec. 2014 (on YouTube). **This is a lecture title, not a journal article**; the SAL abstract (Wayback copy of https://www.sal.org.uk/events/2014/12/brunanburh-in-937-bromborough-or-lanchester/ `[V]`): "Recent years have seen a shaky consensus in favour of Bromborough, in the Wirral. Yet recent research must rule this out. The battle was surely fought near Lanchester in County Durham, where a Roman fort or 'burh' stands on a major route north near the River 'Brune' or Browney. Location of Brunanburh near Durham thus solves permanently an old problem".
- "The Battle of Brunanburh and Cambridge, CCC, MS 183", *Northern History* 53:1 (2016), 138-45 (DOI 10.1080/0078172X.2016.1127631) `[V: Crossref]`: argues that the Life of St Cuthbert manuscript Athelstan gave to the Chester-le-Street community was presented after the victory, "when Athelstan will have been in the Chester-le-Street area", the book having been prepared as 'propaganda' in the summer of 937 `[V-via search summary; not read]`.
- "Brunanburh Located: The Battle and the Poem", in M. Ogura & H. Sauer (eds), *Aspects of Medieval English Language and Literature* (Berlin: Peter Lang, 2018), pp. 61-80 `[V: Wikipedia/Cavill refs]`.
- *British Battles 493-937: Mount Badon to Brunanburh* (Anthem Press, 2020; pb 2021), ch. 13, "937: 'Brunanburh' and English Triumph at Lanchester, County Durham" `[V: Senchus review]`; Pickles notes chs 1, 3-4, 6-8, 10-13 are reworked earlier papers `[V]`.

### 2.3 Breeze's argument (reconstructed)
1. **Brunan- = the Browney.** The river's Old English name was *Brūne*; treated as a weak feminine noun its genitive is *Brūnan*, so *Brunanburh* = "fort on the Brune", *Brunandun* = "hill of the Browney" (Cavill 2023 abstract and summary `[V-via Cavill]`; SAL abstract `[V]`).
2. **The burh = Longovicium**, "a Roman fort or 'burh' ... on a major route north" (SAL `[V]`); Clarkson's report: Breeze "believes that the Roman fort of Lanchester, slightly north of the Browney, may be the lost 'fort of Bruna'" (https://senchus.wordpress.com/2014/12/09/brunanburh-in-937-bromborough-or-lanchester/ `[V]`).
3. **We(o)ndun = 'wen-shaped hill'** (Cavill 2023 abstract `[V]`), applied to the hill at Lanchester.
4. **Dingesmere → *dingles mere*** ('pool of the dingle/dell'): Breeze "proposes that dinges- in the Old English poem ... should be emended to dingles-" (Cavill 2023 abstract `[V]`). **Not a Welsh/Brittonic derivation.**
5. **Route.** Breeze accepts John of Worcester's Humber entry: the Norse "sailed in via the Humber estuary ... mooring their ships and marching to the battlefield" (Clarkson's summary `[V]`); the Scots/Cumbrians presumably came down Dere Street (inference; not verified in Breeze's text `[U]`).
6. **Durham interest.** Symeon's Durham tradition preserves the *Weondune/Etbrunnanwerc* names; the CCC 183 gift to St Cuthbert's community fits a victory nearby (Breeze 2016 `[V-via summary]`). Morillo notes Breeze places the battle "on the Browney near where the Roman road from England into Scotland runs", where Athelstan would intercept a Scots-Norse force (TMR 21.04.01 `[V]`).
7. **Othlynn.** No Breeze interpretation found `[U]`.

### 2.4 Reviews and specific objections
- **Cavill, "The Battle of Brunanburh: The Lanchester Hypothesis", *Neophilologus* 107:3 (Sept. 2023), 431-49 (DOI 10.1007/s11061-022-09758-1)** — the principal rebuttal. Abstract `[V: Crossref]`: "interrogates the claims of Andrew Breeze, in several works, to have securely located the battle at Lanchester ... Breeze's main proposals are discussed: that Brunanburh refers to the River Browney in County Durham; that the name We(o)ndun ... refers to a wen-shaped hill; and that dinges- ... should be emended to dingles-. Alternative interpretations of the material are given ... It is argued that the Lanchester hypothesis does not stand linguistic and critical analysis." Further points as quoted in online summaries of the article (PDF is behind a Springer bot-wall; `[V-via summary]`): "Breeze's whole case rests on the supposition that the name of the Browney was a weak feminine Old English Brune, with an oblique inflection Brunan, and on the evidence presented, the claim cannot be substantiated"; "Breeze identified the hill at Lanchester with Brunandun 'hill of the Browney', but the Welsh evidence does not support the view that references to Brune, Brun and -brun- must be to the river name Browney"; the *dingles* emendation "imports a Middle English word while necessitating change in four letters".
- **Cavill 2022 (NH 59:2), n. 13**: Lanchester is a place where the name Brunanburh "is never actually recorded" `[V]`. Cavill, "The Yorkshire Hypothesis" (*English Studies* 104:1, 2023, 19-38) adds a general point against any north-eastern site: "The lack of interest of northern sources in the battle suggests that it was not fought in the north-east" `[V]`.
- **Thomas Pickles, *Northern History* 60:1 (2023), 117-18** (open copy: https://chesterrep.openrepository.com/handle/10034/627517 `[V]`): the book "is about the etymology of names ... not ... about war"; he "felt ill-equipped to make judgements on the technical linguistic arguments ... This was thrown into relief by comparing Paul Cavill's careful elucidation of the etymological evidence for the roots of Brunanburh and its likely equation with Bromborough on Wirral ... revisited recently in Northern History, 59:2 (2022), with the arguments set out here."
- **Stephen Morillo, *The Medieval Review* 21.04.01 (2021)** `[V]`: the Browney/Lanchester equation "makes sense linguistically and culturally" and is strategically attractive, but is not "proven"; more generally, without "explanations of known syntactic and sound change rules ... what is in fact a reasonable ... argument can simply look like 'this could be this'"; the book "could have used more maps".
- **Nicole Meier, *Archiv für das Studium der neueren Sprachen und Literaturen* 258 (2021)** — exists; content not seen `[U]`.
- **Tim Clarkson** (Senchus, 9 Dec. 2014 and 28 May 2020 `[V]`): "I'm not convinced by the Lanchester theory"; the earliest source "says that Anlaf fled across the sea to Dublin" (against a Humber/east-coast landing); commenter Chris Pickles: Lanchester fort "overlooks a steep sided and narrow valley", hard to square with the Clonmacnoise "plains", and an army landing in the Humber "would achieve little by marching north to Lanchester".
- Summary of the objections: (a) no source ever calls Lanchester or the Browney *Brunanburh*/*Brunan-* anything; (b) the weak-feminine inflection *Brūnan* is unattested for the river; (c) the Welsh *Brun* forms do not require a river; (d) *dingles* is a four-letter, Middle-English-importing emendation; (e) the wen-shaped hill is unprovable; (f) the reconstruction depends on John of Worcester's Humber entry, which most now reject (Neilson; Livingston; Cavill; Woolf: "fundamentally unlikely"); (g) topography (narrow valley) vs the annals' "plains"; (h) no archaeology.

---

## 3. THE FULL FIELD OF CANDIDATES

Principal lists used: Momentous Britain, "Full List of Brunanburh Battlefield Candidates" (46 sites + 8 unverifiable + 5 unseen analyses; https://momentousbritain.co.uk/go/Brunanburh_Battlefield_Candidates `[V]`); Cavill, *Casebook* §5 `[V]`; Wikipedia "Battle of Brunanburh", Location section and footnotes (raw wikitext read `[V]`); Wikishire `[V]`; Neilson 1909 p. 49 `[V]`; Paul Hill, *The Age of Athelstan* (2004), pp. 141-2 (30+ sites; snippets only `[V-partial]`); Cockburn, *The Battle of Brunanburh and its Period Elucidated by Place-Names* (1931) (not seen; cited by Hill and Cavill).

Notes on specific items:
- **Axminster (Devon).** The tradition, as given by Baring-Gould, *A Book of the West*, i (1899), ch. "Axminster" (Gutenberg #48439 `[V]`): "Axminster was the scene of a great battle in the reign of Athelstan, in which five kings and seven earls fell ... Athelstan now endowed a college there for six priests to pray for the souls of those who fell in the battle"; the Newenham Abbey register (temp. Edward III) says the battle was fought "at Munt S. Calyxt en Devansyr" (Coaxdon) and "ended at Colecroft under Axminster"; Baring-Gould links it with Brunanburh (the bishop of Sherborne's death; the two æthelings buried at Malmesbury "because it was the nearest great monastery") and even finds "Brinnacombe" nearby. Momentous Britain traces the identification to Leland's *Itinerary* (vol. 1) and Lysons, *Magna Britannia* (1822); Wikishire attributes a modern restatement to Lawrence Snell (1966) `[V]`; Hill 2004 and the *Casebook* (Parker, "The Victorian Imagination", pp. 400-1) discuss it as antiquarian legend `[V-partial]`. Standard objection: far too far south for a Scots-Dublin coalition; the "five kings and seven earls" is simply the ASC poem's tally attached to a local minster legend.
- **Whalley, Brunstock, Chapel Brampton, Brinklow, Brinsop, "Netherlands".** None appears in any candidate list I could check (Momentous, Cavill, Wikipedia, Wikishire, Hill snippets). Whalley enters the literature only as part of Hardwick's Ribble-valley battle traditions (Hardwick, *On some Ancient Battle-fields in Lancashire*, 1882, ch. IV on Brunanburh `[V: OL snippet]`); Brunstock (Cumberland) appears only in place-name dictionaries (Ekwall: *Bruneskayth* 1253, ON *skeið*). Treat all six as **not real candidates** unless a source is produced. (The Terschelling "Waendune" mentioned by Halloran in 2012 as Mick Deakin's find is a name-parallel, not a Dutch battle site `[V]`.)
- **Dunbar, Bamburgh, Brunton, Broomridge, Brinkburn.** Hill 2004 lists Dunbar as candidate no. 4 ("at, or near, Dunbar, Athelstan is said to have ..." — from Malmesbury's 934 narrative) and Brinkburn as "another candidate" `[V: OL snippet]`; Momentous traces Dunbar, Bamburgh and Brumford to Cockburn 1931 as cited by Hill `[V]`. Broomridge/"Brumridg near Brumeford in Northumbria" is Camden's (1600/1607) and Tyrrell's (1700); Brunton is Sharon Turner's (1820) refinement of Camden; Brinkburn is Francis Grose (1813) `[V: Momentous]`.

### Master table — every candidate found

| Site | Proposer, year | Key argument | Main objection |
|---|---|---|---|
| **Bromborough, Wirral** (incl. Brackenwood/Bebington golf course, Brimstage, Wallasey Pool) | Gibson 1692; Baines 1867; Tudsbery 1907; Dodgson 1957 (Saga-Book 14); Cavill/Harding/Jesch 2004; Livingston 2011, 2021; Downham 2021; Wirral Archaeology 2019 | Only surviving name that actually descends from OE *Brunanburh* (13th-c. *Bruneburgh*); Norse settlement; Thingwall → *Dingesmere* 'mere of the Thing'; straight Irish-Sea crossing | Too far south/west for Scots-Cumbrians (Halloran; Clarkson 2010); *Brunan-*+*burh* common elements (Campbell); Dingesmere = wetland "over which ships sailed" (critics); Higham: Mersey not a shipping lane; Deakin 2022: first element 'brown'; 2019 finds unconfirmed (Williams 2021) |
| **Burnswark, Dumfriesshire** | Hodgkin 1885; Stevenson 1891; Neilson 1899/1909; Oman 1910; Halloran 2005/2010 | Solway rendezvous; Beverley *Scotorum vadum*; hillfort + Roman camps = *burh*/*werc*; Gaimar *Bruneswerce*; Wendune = 'wen-hill'; *Dunbrunde* | No spelling before 1541/2 and those are *Burnys-/Burnis-* (burn+wark or Birrens+wark); *dun* ≠ hillfort; battle on English soil; Egil's distances (Law 1910); no archaeology |
| **Lanchester (Longovicium), Co. Durham** | Breeze 2014-2020 | Browney = OE *Brūne*, gen. *Brūnan*; Roman fort on Dere Street; wen-hill; *dingles mere*; Durham/Chester-le-Street links | Weak-fem. inflection unattested; name never recorded there; *dingles* emendation; depends on Humber entry; narrow valley (Cavill 2023; Clarkson) |
| **Browney / Brancepeth (Durham)** | Guest 1838 (second-hand); Bosworth 1848 | "plain between the Tyne and Browney" | As for Lanchester; no name evidence |
| **Hunwick, Co. Durham** | Björnsson & Vernharðsson, *Brunanburh: Located through Egil's Saga* (2020) | Saga topography | Saga is "unsupported in practically all its details" (Campbell) |
| **Binchester (Durham)** | mentioned by Battlefields Trust | Roman fort on Dere Street | none beyond generic |
| **Aldborough (Ouse), Yorks** | Skene, *Celtic Scotland* i (1876/1886); accepted by Arnold 1885 | Ouse/Humber tradition; Malmesbury's "far into England" | No name link; Humber entry doubted |
| **Boroughbridge/Ripon** | Baring-Gould 1908 | as above | as above |
| **Brinsworth / Tinsley (Rotherham)** | Goodall 1913; Cockburn 1931; Burne 1950; Wood 1980 (Saga-Book 20), 1999/2001 | Strategic Roman-road nexus; White Hill temple = *Weondun* 'holy hill'; *Brynesford* ≈ *Bruneford* | *Bryni* ≠ *Bruna*; persistent -s- genitive; White Hill = *wiht* 'bend' (Page 1982; Cavill 2011) |
| **Went valley: Burghwallis / Barnsdale Bar / Robin Hood's Well** | Wood 2013 (YAJ 85), 2017 | Domesday *Burg* + spring = *Brunnanburh* 'spring-fort'; Went ≈ *Wendun*; Great North Road; Humber entry | Burghwallis never called "fort at the spring"; Went Hill not a *dūn*; Went < *Wenet*; Humber entry unreliable (Cavill 2022/2023) |
| **Bramham Moor** | Pearson 1995 | Roman road junction | none specific |
| **Harrogate** | J. H. Turner 1871 | local names | none specific |
| **Elslack / Burwen Castle (Broughton)** | James Wilson 1910 (SHR 7) — offered half-seriously | Roman fort, Danish siege tradition, plain of Otley ≈ Othlyn | proposer himself disbelieved it |
| **Brough-on-Humber / South Cave** | Adrian C. Grant 2019 (v4 Aug. 2019) | Roman road Petuaria–South Cave; Beverley; "All Saints" dedications | No name evidence; Humber entry |
| **Barton-upon-Humber** | Hesleden 1846; Robert Brown 1906; Deakin 2020 (retracted) | earthworks; artesian spring | none of the names |
| **Kirkburn (Driffield)** | "Mr Holderness" 1892; Deakin 2012 | early forms *Burn/Burne* | -burn ≠ *Brunan-* |
| **Londesborough / Nunburnholme / Burnby / Little Weighton (Wolds)** | Napper 1888 (Burnby); Wake 1885 & Bulmer 1892 (Little Weighton); Sally England 2013/2020 (Nunburnholme cross) | Wolds between Beverley and York; Pseudo-Ingulf's pilgrims | late/derivative sources |
| **Brompton (Northallerton)** | Kirsten Møller 2022 | name | none specific |
| **Brumby, Lincs** | Ingram 1823; Giles 1912/14 | name similarity | name is ON *Brúni*-bý; wrong coast (Neilson) |
| **Burnham, N. Lincs** | Rev. Alfred Hunt 1905 | name | as above |
| **Bourne, Lincs / Bourn, Cambs** | Ramsay 1898; Cyril Hart 1992 (*Danelaw*); Deakin 2025 (*Battalia* III) | ON *brunnr* 'spring' dat. pl.; Bourn Brook valley | *brunnr* not inflected *brunan*; flat, no *dūn*; Gaimar knew Bruneswald and never conflated it (Cavill 2011 §5.4) |
| **Bromswold / Bruneswald (Hunts–Northants)** | Smyth 1975/79 | *Bruneswald* ≈ *Brunesburh* (HH Rolls ed.) | *Brunesburh* is an editorial reconstruction; forest not a battlefield; no *feld/dun/burh* names (Cavill) |
| **Burnley / River Brun (incl. Saxifield, Worsthorne, Castercliff)** | T. T. Wilkinson 1856/57 (THSLC 9); Marquis 1909; Partington 1909; Bullen 2021 | River Brun; Cuerdale route Dublin–Ribble–York; Winewall ≈ *Wendune*; Otley ≈ Othlyn | Brun is a back-formation from Burnley (Cavill, *Casebook* p. 331); Cuerdale hoard c. 905-10, not 937 |
| **Broadclough (Rossendale)** | Newbigging 1893 | Broadclough Dykes; Danish chieftain grave | folklore |
| **Bamber Bridge / Brownedge; Bamford (Rochdale)** | Charles Hardwick 1857 | Cuerdale hoard; Ribble crossing | hoard predates 937 |
| **Blackburn / Livesey** | Porteus 1927; Livesey Hist. Soc. | *Leofsige* 'beloved victory' | folk-etymology |
| **Bruna Hill (Garstang), Amounderness** | Clarkson 2014 | Æthelstan's NW frontier; Scots' land route | Clarkson's own doubts after Deakin |
| **Bourne Hill (Thornton, Fylde) / Burn Naze** | Deakin 2015; John R. Kirby 2019 | *Brunnan-* 'spring'; "sea of noise" | unpublished/name-only |
| **Heysham** | Heysham Heritage 2012 | *Brun* names | name-only |
| **Wigan (Bottling Wood)** | Jonathan Starkey / Momentous Britain 2022 | route north; local names | name-only |
| **Bromfield, Cumberland** | Goodall (indirect); noted by Neilson 1909 | *Brunefeld* | south of Solway; *feld* only |
| **Brown Hill, Dunbartonshire** | O. G. S. Crawford 1934 | name | none |
| **Dunbar** | Cockburn 1931 (via Hill 2004) | Malmesbury's 934 "Dunbar" | wrong campaign |
| **Bamburgh / Brumford** | Cockburn 1931 (via Hill) | name | none |
| **Broomridge / Brumridg near Brumeford (Northumberland)** | Camden 1600/1607; Tyrrell 1700 | name | none |
| **Brunton** | Sharon Turner 1820 | Camden refined | none |
| **Brinkburn** | Grose 1813; listed by Hill 2004 | name | none |
| **Breandown (Somerset); Broomfield/Bromfield (Taunton)** | Bulleid 1888; Birch 1885 | *Brune-dun*; *Brunefeld* | far south |
| **Axminster / Kilmington (Devon)** | Leland c. 1540; Lysons 1822; Snell 1966 | minster founded for the slain; Newenham register | far south; legend |
| **Winchester (Guy of Warwick); Dunlow/"Donelew in Wilthschire" (Eulogium)** | medieval romance/chronicle | — | never taken seriously (Cavill §5) |

Momentous Britain's "unseen analyses" (Norwood 1922; Spilman 1875; Bryant n.d.; Black 1936; Harrison 1983/4 in *Durham Arch. Journal* 1) may contain further sites `[U]`.

---

## 4. THE SCOTTISH/CUMBRIAN ROUTE EVIDENCE

### 4.1 Tim Clarkson, *Strathclyde and the Anglo-Saxons in the Viking Age* (Edinburgh: John Donald/Birlinn, 2014), ch. 5
- Position: "Amounderness, the district between the rivers Lune and Ribble, as a possible conflict zone containing the battlefield", with two candidate names, **Bruna Hill near Garstang and Bourne Hill near Thornton on the Fylde** (Clarkson's own blog, https://vikingstrathclyde.wordpress.com/2015/07/08/brunanburh-on-the-fylde/ `[V]`; book chapter pages not verified `[U]`).
- Reasoning (his words): "I take the view that Amounderness was the most northwestern part of Athelstan's 'England' at the time of the battle"; "Athelstan had previously purchased the territory ... from landowners who were most likely of Viking stock" and "granted it to the Archbishop of York in 934"; beyond it Lonsdale and Westmorland "were under the authority of Anglo-Scandinavian lords who answered to York rather than to Athelstan" `[V]`. The underlying logistic argument (from his 2010 comments): the Old English poem gives the Norse a seaborne flight but has Constantine simply return "to his own land", so the Scots and Cumbrians came and went by land; Bromborough "seems too far south to be a viable objective for the northern Celtic powers" `[V]`. He later conceded Deakin's objections to Bruna Hill and found Bourne Hill "increasingly plausible" `[V]`. Reviewer (Hazel Tree, 2015): "long and careful consideration ... a lost site" `[V]`.

### 4.2 Alex Woolf, *From Pictland to Alba, 789-1070* (Edinburgh, 2007), pp. 169-72 (index: "Brunanburh, battle of (Dun Brunde), 126, 128-9, 169-70, 172, 174-5, 177, 181, 184"; Map 4.3 "Suggested locations for the battle of Brunanburh, 937", p. 170; Burnswark, Bromborough, Bromswold all p. 171) — snippets via Open Library search-inside of ia:frompictlandtoal0000wool `[V]`:
- "The site of the battle is unknown. CKA calls it Dun Brunde, the poem in ASC calls it Brunanburh (which has been followed by most modern scholars) ..."
- On the Humber: "[John of] Worcester, who states that Anlaf landed in the Humber. This is fundamentally unlikely, since he came from Dublin, and probably ... [contrast the fleets] of Harald of Norway (1066) and Sven of Denmark (1069) which did indeed make landfall in the Humber. As noted earlier, the obvious route between Dublin and York lay across central Lancashire ..."
- The plausible landing zone "would seem to cover the whole coastline from the Wirral, in Cheshire, northwards into Dumfries and Galloway".
- "A favourite location, which has much to commend it, is Bromborough in the Wirral, in precisely the kind of location where Amlaib [would be expected to land] ..."
- "One Scottish site has been suggested, Burnswark ... and this hypothesis has recently been revived [n. 9 = Halloran 2005]. It must be said, however, that the evidence is ..." (sceptical).
- **Othlynn:** a full-text search of the book for *Othlyn/Othlynn/Othlinn* returns nothing; the index lists the Annals of Clonmacnoise only at p. 163 (the 934 campaign). **The suggestion that Woolf discusses the significance of "Othlynn" is therefore unsupported** `[V-negative]`.
- **Clyde/Forth route:** no passage suggesting the Dublin fleet came via the Clyde or Forth was found; hits for "Clyde"/"Forth" concern other episodes. **Unverified/probably a misattribution** `[V-negative, snippet search only]`.
- Woolf considers the Constantine-in-the-raid material in the CKA/Egil tradition a late addition "derived from a now-lost saga or poem" (pp. 178-81, as summarised by Wikipedia's "Constantine II of Scotland" `[V-via Wikipedia]`).

### 4.3 The Chronicle of the Kings of Alba: "bellum Duinbrunde"
- Text (Poppleton MS; Skene, *Chronicles of the Picts and Scots*, 1867, p. 9 `[V: OCR]`): "Bellum Tinemore factum est in xviii. anno inter Constantinum et Regnall, et Scotti habuerunt victoriam. Et bellum Duinbrunde in xxxiiij. ejus anno ubi cecidit filius Constantini. Et post unum annum mortuus est Dubucan filius Indrechtaig, mormair Oengusa. Adalstan filius Advar rig Saxan, et Eochaid filius Alpini, mortui sunt. Et in senectute decrepitus baculum cepit, et Domino servivit ..."
- Translation (A. O. Anderson, *Early Sources* i (1922), adapted `[V]`): "And the battle of Duinbrunde [Brunanburh] in his 34th year [an error for 37th], and in it fell Constantine's son and after one year he died [i.e. Dubucan]." Editions: Skene 1867; M. O. Anderson, *Kings and Kingship* (2nd ed. 1980), pp. 249-53; B. T. Hudson, "The Scottish Chronicle", *SHR* 77 (1998), 129-61 `[V: citations]`. Composition: early 11th c., shortly after Kenneth II (Wikipedia; Cowan 1981; Broun, "The birth of Scottish history", *SHR* 76 (1997) `[V: cited on duffus.com; not read]`). **No specific statement by Dauvit Broun about the "Duinbrunde" entry was located** `[U]`; Broun's ODNB "Constantine II" (2004) presumably touches it `[U]`.
- Interpretation: Campbell (1938) — "dun in Gaelic place-names is equivalent to burh"; Cavill: the form "apparently reflects a Gaelic tradition ... an accurate translation" of *Brunanburh* and so "discredits" the idea that *-burh* was a poetic invention (*Casebook* §3.2 `[V]`). Halloran: it may show the site "had a Gaelic name as well as an English one", implying a bilingual borderland (Senchus 2012 `[V]`). Neilson listed it as "Dunbrunde" (1909, p. 38 `[V]`).

### 4.4 Other route evidence bearing on a northern site
- Annals of Ulster 937: "a great and horrible battle ... between the Saxons and Norsemen ... Amlaibh escaped" (Neilson's quotation `[V]`); AU 938: Amlaíb "in Áth Cliath again" (Cavill 2014 `[V]`).
- The poem: Norse "sought Dublin over deep water"; Constantine went "north to his own land" — used by Law (1910), Clarkson (2010), Halloran for a west-coast/land-route geography `[V]`.
- Egils saga's geography is used by Neilson but dismissed by Campbell; Wilson (1910) and Law (1910) show it can be fitted to almost any site `[V]`.
- Livingston's "Requirement 4" (no Humber) and his logistics: an eastern landing would add "at least a thousand miles" of sailing for a Dublin fleet `[V]`.

---

## 5. BALANCE

**Burnswark.** *Single strongest point:* it is the one candidate that makes the coalition's geography easy — a Solway rendezvous reachable by the Dublin fleet in a day, by Strathclyde and Alba overland, on the Roman trunk road, at a conspicuous flat-topped hill that really does carry a summit fort and two great earthwork "works", with the Historia Regum/Gaimar *-werc* forms and the Gaelic *Dún Brunde* lending superficial support (Neilson; Halloran; Clarkson's 2010 concession). *Single fatal weakness:* the name. No spelling of Burnswark exists before 1541/2, and every early spelling (*Burniswerk*, *Burnyswarke*, *Burniswork*, *Burneswark*) shows a strong *-s-* genitive of *burn(s)* (or of *Birrens*) plus Scots *wark* — nothing that can descend from OE *Brūnan-*; the identification therefore rests on one 14th-century Gaimar MS reading (*Bruneswerce*) and on look-alike logic (Cavill 2008, 2011, 2014). Add that the poem frames the fight as a defence of English homeland and has Constantine flee *north*, and that Halloran himself calls his Cumbric *brun*/OE *wen* synonymy a "big if". Verdict: strategically the most coherent of the northern sites, philologically the weakest.

**Lanchester.** *Single strongest point:* a genuine Roman *burh* on Dere Street, on a river whose Old English name (*Brūne*, per Ekwall) is at least of the right shape, in the sphere of the Durham community that preserved the *Weondune/Etbrunnanwerc* names and that Athelstan patronised (CCC 183). *Single fatal weakness:* every link in the chain is an unattested hypothesis — the weak-feminine genitive *Brūnan* for the river (Cavill 2023: "cannot be substantiated"), the "wen-shaped" reading of *Weondun*, the four-letter emendation *dinges* → *dingles*, and a campaign narrative that needs John of Worcester's Humber landing (which Woolf calls "fundamentally unlikely") followed by a pointless march north into a narrow valley, contradicting the poem's flight to Dublin over *Dingesmere*. Nothing in any source names the place. Verdict: a name-shaped guess dressed in Roman masonry; the reviews (Pickles, Morillo) politely decline to endorse it and the one full philological examination (Cavill 2023) rejects it.

Against both, the standing objection of the "nescio" school (Campbell 1938; Wilson 1910) still applies: the sources are "elastic enough for diverse geographical adaptation", and only Bromborough has a name that demonstrably descends from *Brunanburh* — which is also why Bromborough's critics attack the name evidence as insufficient rather than offering a better-attested name of their own.

The master table is in §3.

---

## Sources actually opened (URLs)
- Neilson 1909; Wilson 1910; Law 1910 + Neilson's reply: SHR vol. 7 OCR, https://archive.org/download/ScottishHistoricalReview7/Scottish_Historical_Review_7_djvu.txt
- Halloran 2005 abstract: https://www.euppublishing.com/doi/abs/10.3366/shr.2005.84.2.133 (Crossref record); Halloran 2010: https://www.jstor.org/stable/23073508 (Crossref reference list only); Halloran 2013: https://www.tandfonline.com/doi/abs/10.1179/0078172X13Z.00000000042
- Cavill, *Casebook* essay PDF: https://www.nottingham.ac.uk/-sczsteve/09essay-cavill.pdf ; Cavill 2014: https://www.nottingham.ac.uk/-sczsteve/Cavill_2014.pdf ; Cavill 2022 "Spellings": https://www.tandfonline.com/doi/full/10.1080/0078172X.2022.2088437 ; Cavill 2023 "Yorkshire Hypothesis": https://www.tandfonline.com/doi/full/10.1080/0013838X.2022.2154045 ; Cavill 2023 "Lanchester Hypothesis": https://link.springer.com/article/10.1007/s11061-022-09758-1 (abstract only); Cavill 2008 via Open Library search-inside (ia commodityofgoodn0000unse)
- Senchus (Clarkson): 2009 https://senchus.wordpress.com/2009/04/30/brunanburh-and-burnswark/ ; 2010 https://senchus.wordpress.com/2010/11/01/the-torbeckhill-sword/ ; 2012 https://senchus.wordpress.com/2012/08/29/more-brunanburh-links/ ; 2014 https://senchus.wordpress.com/2014/12/09/brunanburh-in-937-bromborough-or-lanchester/ ; 2020 https://senchus.wordpress.com/2020/05/28/british-battles-493-937/ ; SASVA 2015 https://vikingstrathclyde.wordpress.com/2015/07/08/brunanburh-on-the-fylde/
- Breeze SAL 2014 abstract (Wayback): https://web.archive.org/web/2016/https://www.sal.org.uk/events/2014/12/brunanburh-in-937-bromborough-or-lanchester/ ; Breeze 2016 Crossref: https://www.tandfonline.com/doi/full/10.1080/0078172X.2016.1127631
- Pickles review: https://chesterrep.openrepository.com/handle/10034/627517 ; Morillo: https://scholarworks.iu.edu/journals/index.php/tmr/article/view/32325/36155
- Woolf 2007 via Open Library search-inside (ia frompictlandtoal0000wool); Skene 1867 OCR: https://archive.org/download/chroniclesofpict00sken/chroniclesofpict00sken_djvu.txt ; CKA translation: www.duffus.com (PDF mirrored on archive.org)
- Symeon (Arnold, Rolls Series) OCR vols i-ii (copies in texts/editions_ocr/: symeonismonachiopera1/2.txt)
- Wikipedia (raw wikitext) "Battle of Brunanburh", "Constantine II of Scotland", "Owain ap Dyfnwal (fl. 934)", "Burnswark Hill", "Chronicle of the Kings of Alba"; Wikishire "Brunanburh", "Burnswark"; Battlefields Trust pp. 891-2; Momentous Britain candidates page; Livingston, "The Search for Brunanburh: A Summary"; HNN/Herald 2005; History Today 2006 teaser; Baring-Gould (Gutenberg 48439); Tudsbery 1907 OCR; Hill 2004 via Open Library snippets.