Is there a barrow in Barrow?
Old English beorg meant a mound or a rounded hill, and it gives us the word barrow. I tested 129 English places named from it against 6,100 protected burial mounds. Do the names lead to the barrows?
A hill word, not a grave word
Place-name dictionaries often gloss beorg as "burial-mound/hill" because no one can tell which sense a namer meant. One way to find out is to count. If beorg places were named after barrows, more scheduled barrows should survive near them than near places named after plain hills.
That is not what the data show. Within 2 km, beorg places have about 1.2 times the barrows their local area would lead you to expect. Places named dūn or hōh, both ordinary hill words, score 1.1 to 1.2. The gap is well inside the noise. In most cases a beorg was a hill.
Each dot shows the barrows found within 2 km of places with that element, divided by the number expected from barrow density within 30 km of each place. 1 means "no more than the surroundings". The bars are 95% bootstrap intervals.
Most Barrows were groves
The village names that look most like "barrow" are the weakest evidence of all. Of the places in the Key to English Place-Names called Barrow-something, eight come from bearu, "grove" or "wood". They include Barrow upon Soar, Barrow upon Trent, Barrow upon Humber, Barrow Gurney and Barrowford. Only four come from beorg. Barrow-in-Furness has a different origin again.
Even where the dictionaries do say "burial mound", the mound has usually gone. KEPN glosses 12 beorg names as burial mounds. Nine of them have no scheduled barrow within 2 km. That includes Barrow in Rutland, where the editors point to a tumulus near The Green, and Barrowden, "burial-mound hill", with "a number of tumuli known in this area".
| Place glossed "burial-mound" | Barrows within 2 km | Expected |
|---|
Scheduled monuments are only the barrows that survive well enough to be protected. Centuries of ploughing have levelled thousands more, especially in the arable east where many beorg names sit. A zero here means no surviving, scheduled barrow, not that there never was one.
The one that might
Hlāw, the word behind Taplow, Lowe and many northern "laws", scores highest: 1.5 times the expected barrows within 2 km, falling to 1.1 by 3 km. Taplow is named after a real Anglo-Saxon burial mound in its churchyard. Still, with 88 places the interval runs from 0.9 to 2.3, so this is a lead, not a finding.
The controls mostly behave as you would hope. Burh, "fortification", scores 1.3, probably because hill-forts and barrows share the same high ground. Bearu (grove) sits below 1, as a word with nothing to do with mounds should. So, oddly, does Old Norse haugr, which can mean a burial mound. Its 26 places have only 3 barrows within 2 km.
Method
- Places come from the Key to English Place-Names (KEPN, University of Nottingham). I used every place it lists under each element. Each place is a single point, usually the settlement.
- Barrows are scheduled monuments from Historic England's National Heritage List whose names include barrow, tumulus, burial mound, round cairn or bowl cairn: 6,100 records.
- For each place I counted barrows within 1, 2 and 3 km. The expected count is the number within 30 km scaled by area. This controls for the fact that barrows cluster on the chalk and the moors.
- Results are pooled by element as total observed divided by total expected, with 2,000 bootstrap resamples of places for the intervals.
Scheduled monuments cover England only, so this test is England only. Some places appear under two elements (Barrowden is beorg and dūn). Grid references mark a monument's centre, so a long barrow cemetery counts once per listing, not once per mound.