NGRef: !NT 0255 8078
OSMap: LR65
Type: Fort, Minor Settlement, Camp
The Carriden Site

The Site of Carriden/Veluniate
looking west towards Carriden House
Probable Road: E (11) to Cramond (Edinburgh, Lothian)
Antonine Wall: W (3.75) to Kinneil (Central)

The Veluniate Stone Found in 1956


"To Jupiter Best and Greatest,
the resident villagers of the fortified settlement of Veluniate,
administered through the agency of Aelius Mansueto,
willingly, gladly and deservedly fulfill their vow."

(J.R.S. xlvii, 1957, pp.229-30)

This fort and settlement lies on the southern shore of the Firth of Forth overlooking Torry Bay in Fife, and is situated very close to the eastern terminus of the Antonine Wall at Bo'ness, which lay only a couple of miles to the west. It has been suggested that Carriden and other forts at Cramond and Inveresk formed a chain of forts along the Lothian (southern) shore of the Firth of Forth, perhaps augmented by others as yet undiscovered. This idea is sound, as it would mirror the situation at the western end of Hadrian's Wall to the south, where a chain of forts and watchtowers have been identified all along the coast of Cumbria, the so-called 'Western Sea Defences'.

"Observation from the air in 1945 disclosed at Carriden the defences of the long-lost Roman fort. Photographs record three ditches, forming the defences of the east and part of the south side. The east side was some 440 ft. long, and contained a gate 150 ft. from the south-east angle ; the length of the south rampart appears to have been at least 400 ft. Trial trenches dug in 1946 located the defences on the east side and yielded pottery of Antonine date. The western half of the fort lies within the grounds of Carriden House, while even such part of the site as is available for digging seems to have been heavily denuded." (St. Joseph, 1951)

Photographs from 1945 showed a triple ditch system on the east and south sides of this fort, which measures roughly 440 ft. from north-south by 400 ft. transversely (c.134 x 122 m), enclosing an area of around 4 acres (c.1.6 ha).

Like many Roman forts in Scotland, Carriden has a couple of marching camps nearby; at Kinglass Park (NT0080) and Muirhouses (NT0180), both near Bo'ness, Central.

"Cohort Stone" from Veluniate


"The Eighth Cohort, century of Stateles."

(RIB 2138)

The only inscribed Roman stone reported in the RIB from the Carriden site is undoubtedly a legionary building stone (vide RIB 2138 supra), which proves the presence of the military, however, an altar to Jupiter was found at the site "ploughed up about 150 yards to the East of the fort" in 1956, dedicated by the vicani consistentes proving that there was a civil settlement here also. A photograph of the altarstone appears in the handbook 'The Antonine Wall' by Anne Robertson, published by the Glasgow Archaeological Society in 1960. The stone now resides in the National Museum of Antiquities in Edinburgh. The text of the stone, as it appears in Robertson's photograph, is shown above with a tentative translation.

The name of this minor settlement was not listed among the four towns attributed to the Selgovae tribe by Ptolemy, but there is classical confirmation of the Roman name for Carriden within the Ravenna Cosmology (R&C#191). This seventh-century geographical work records the entry Velunia - between the unidentified stations Rumabo and Volitanio - which may be rendered in the nominative as Veluniate.

In addition to the Veluniate stone, a Roman dedicatory building inscription has been recovered from the Wall nearby Bo'ness (vide RIB 2139 infra).

Building Inscription from Bridgeness, west of Carriden


"For Imperator Caesar Titus Aelius Hadrianus Antoninus Augustus Pius, Father of his Country, The Second Augustan Legion, were responsible for building four thousand six-hundred and fifty-two paces [of the rampart wall]."

(RIB 2139; dated: 139-161AD)

This distance slab is decorated with a relief of a Roman religious festival known as the suovetaurilia. Also depicted on the stone is a vexillum or legionary divisionary standard bearing a secondary inscription LEG II AVG "Legio Secundae Augusta". The stone was found in 1868 at the butt-end of the Antonine Wall on Windmill Hill, overlooking the Firth of Forth.

The Dateable Pottery Evidence

The only pottery of note is a piece of samian Form 37 bearing the stamp of Cinnamus, which is dated to the Antonine period.

The Numismatic Evidence

Only 2 coins have been recovered from the Carriden area, an aureus? (possibly only brass) of Vespasian, now lost, and a sestertius of Hadrian found on the ground surface.

See: The Roman Wall in Scotland by Sir George MacDonald (Oxford, 2nd Ed. 1934) pp.190/191;
Air Reconnaissance of North Britain by J.K. St. Joseph in J.R.S. xli (1951) p.61;
The Roman Inscriptions of Britain by R.G. Collingwood and R.P. Wright (Oxford 1965);
The Roman Occupations of Scotland by B.R. Hartley in Britannia III (1972) pp.1-55;
A Survey of the Coin Finds from the Antonine Wall by Richard Abdy in Britannia xxxiii (2002) pp.189-217.
Many thanks to Derrick Matthew for the invaluable information and an excellent picture of the Veluniate altarstone.

Veluniate Related Lynx

Bo-ness.ORG.UK - Barrowstounness local history site Mastered by Ken Wright
An Excellent Site! Recently Updated - Includes the Ace sections
Bo-ness.ORG.UK - Time Line
Bo-ness.ORG.UK - History Trails Flash Graphics!

Roman Name

The attribution of this name to this place is ranked: secure

Ravenna Cosmography: Velunia; Inscription

The name is securely located to Carriden by an inscription found nearby

The etymology of this name has been much debated because of its important link to Pictish. Nennius tells us that the wall started from a placed called "Cennil" by the (Gaelic) Scots and "Peneltun" in Old English. Bede gives the Old English as "Penneltun" and also mentions that it was called "Peanfahel" by the Picts and "Penguaul" in Welsh. This provides us with one of the very few known Pictish words together with either its apparent translation, or at least its reproduction in another languages.

These forms are similar as "Gu" is used in Welsh to represent "W", likewise, "V" in Latin to represent "W", and "F" in linguistically very close to "V" (being the voiced and voiceless forms of the labiodental fricative as in knife and knives). As such "gual", "vel", "fahel" and "wall" are all phonetically close enough that they can be variants of the same word written in different languages. So they could be either meaningless phonetically similar reproductions of the same word in different languages or meaningful translations in languages where the same word was phonetically similar.

The easiest to translate is the Welsh "Penguaul". "Pen" in Welsh is a common landscape prefix meaning "head" & Gwal in modern Welsh means wall. So it appears to mean "head of ..." or more likely: "end of the wall". Old English had the word "weall" which meant "wall" but interestingly the Old English is "Penneltun". So that the Welsh "Pen appears to be written "Penel" which has no meaning in Old English. Likewise Cennil has no easy translation in Gaelic. The Gaelic name has been explained by a tendency of Welsh "P" to be written in Gaelic as "K" or "C" so "Penel" -> "Cenel". Thus both the Old English "Penel" and Gaelic "Cennil" appear to be related and neither is obviously a translations in these languages. So they appear instead to be meaningless phonetic reproductions. This strongly suggest they derive from the Welsh or Pictish forms.

As "h" is often silent, we can see that Pictish "fahel" is phonetically close to both Latin "Vel" and words like "Wall" & "Gwal", but we do not know enough Pictish to speculate about its meaning. However as Pictish prefix "pean" seems to be a form of Welsh prefix "Pen" it seems reasonable that Pictish "fahel" is also a form of Welsh "gwal" meaning wall. But linguistics does not help resolves whether the Pictish form came first or derives from the Welsh.

Note also the similarity to Latin VALLUM meaning rampart or ditch which also occurs in other European languages. Note also that VELUNION on Hadrian's wall and VOLITANIO on the Antonine Wall have very similar "V-L" forms. This strengthens the likelihood that whatever the language it originally came from, "vel" in Velunia is connected with the wall. Thus it is safe to conclude that the "Vel" in Velunia (or Veluniate) likely had a meaning derived from "Wall". However, as there is no evidence of a preceding wall, this implies the name came after the Roman occupation. So we cannot rule out the possibility that the word was made up in Latin or a language spoken by soldiers building the wall.

There's a similar name in England at Wallingford recorded early as "Welinga ford" which in Latin would be VELINGA. This has been variously translated with "place of the foreigner" being a common suggesting. But another possibility is from the suffix -ing or -ung which forms nouns out of verbs. As such "walling-ford" would be the noun form of "to wall" written in Latin as "Vellung". There are a couple of modern places of note. A kilometre south is "Walton" (NT026793) and a bit further "woolston". wealand v. wealh-land (a foreign land) wealian (To be impudent, bold, wanton) weallan (of water, &c. issuing from a source, to well, bubble forth, spring out, flow) wealian (To be impudent, bold, wanton. v. wealh) As an aside, it is also worth noting that this same "Kair Eden" has been used to derive the supposed Gaelic etymology of Edinburgh which was never recorded with any such name. In reality, the earliest recorded name is Edwinesburgh which is easily translated as meaning Edwin's Burg from a Northumbrian King of that name who ruled that place in the 7th century.

Some have derived the etymology from Irish folladh meaning ruling, however as there seems to be no similar word in Scots Gaelic this seems unlikely. The nearest words in Scots Gaelic are phonetically more distant being related to flath (chief). So, not only is the Irish an apparent rare isolate and a poor fit, but the fit gets worse where we expect it to be closer in Scots Gaelic. Moreover, if the original were Gaelic why is the later Gaelic Cennil so different if it is supposedly the same language? Gaelic seems unlikely. However, it must be noted that the assumed modern place that carries the same name as Nennius' "Cennil" is not Carriden but Kinneil (earlier forms: Karreden/Kair Eden) several miles inland and close to another Roman fortlet. Either the Kinneil name moved inland, perhaps because the estate moved lock stock and barrell from the coast, or two different names, at closely located places were being referred to.

Welsh Gaelic Old English Other
gwal (wall) folladh (ruling) weall, (wall)
will, (well, spring)
willan (wells, springs)
[be]feolan (to command)
fahel (wall)

Note also that Carriden (earlier forms Karreden/Kair Eden) does not appear to derive from the Roman name.

Page Citation: Mike Haseler, Kevan White (2018) "Roman Britain: VELVNIATE"