Thursday, January 4, 2018

The Lydney Park Temple of Nodens as 'Avalon'

In previous posts I've written about the possibility that the Temple of Nodens at Lydney Park may be *Nemetaballa, the Sacred Grove of the Apple Trees.  The argument, first proposed by Richmond and Crawford, seemed strengthened by the MIRABILIA of the Historia Brittonum, which described a miraculous apple-bearing ash tree at the mouth of the River Wye.  Alas, the Wye estuary is a considerable distance from Lydney.  This fact casts considerable doubt on the identification of the ash tree with the site of the Lydney temple.  Too bad, as the apple-bearing ash may have been conjured from a name like Avalon, where the word was fancifully derived from Welsh afal, 'apple' + onn, 'ash-tree.'

Some map work has, however, allowed me to salvage the identification.

Lydney's etymology (Mills) is "island or river-meadow of the sailor, or of a man called *Lida.  OE lida [sailor] or personal name (genitive -n) + eg."  According to the English Place-Name Society, eg is

"An island. In ancient settlement-names, most frequently refers to dry ground surrounded by marsh. Also used of islands in modern sense. In late OE names: well-watered land."

Anglo-Saxon eg, 'island', brings to mind Geoffrey of Monmouth's insula Avallonis/pomorum and the Welsh Ynys Afallach. 

However, lida/sailor is a replacement of an earlier god name.  So far as I know, this idea was originally proposed by Rhys and is discussed in P.C. Bartram's A CLASSICAL WELSH DICTIONARY:

"Nodens is probably the Nudd of Welsh legend and the Irish Nuada Argatlám, which leads to the
Welsh Lludd Llaw Eraint. This transformation of the name from Nudd to Lludd (see s.n. Lludd Llaw
Eraint) probably occurred fairly early as the latter form evidently survives in the name of the site,
Lydney. (John Rhys, Hib. Lect., p.125; Celtic Folklore, p.448)."

More recent scholars, such as John Koch, agree with this derivation:

"There, the place-name Lydney is first recorded as Old
English Lideneg in a source from c. 853. The meaning of
the place-name is ‘Lida’s island’, where the Anglicized
personal name is to be explained as a borrowing from
archaic Welsh */Lü:d/ < British Nodons. This borrowing
probably occurred in the 7th century, when
English speakers first came into the Lydney area."

As archaeology has revealed, Nodens of Lydney had pronounced marine attributes. In the Life of St. Collen, the king of the false Avalon Glastonbury is Gwyn son of Nudd ( = Nodens).

Now, as it happens, at the mouth of the Wye there is a Lyde Rock.  Place-name experts have no early forms for this name, and I've been unable to find it on maps predating the 1800s.  Other Lyde place-names in England are to be derived either from Old English hlid, 'slope', or hlid, 'gate.'  Either one may apply to Lyde Rock, as the rock may slope into the water, or it may have been metaphorically viewed as a gate to the Beachley Passage of the Severn. It is unlikely to be an Anglicized form of Welsh llwyd, 'grey.'  Certainly, it is not impossible that we have here either OE lid, 'ship', or lida, 'sailor.' I have some queries outstanding to various authorities and if I'm able to come up with anything more certain regarding this place-name, I will add it to the post. (1)

What I think may have happened is this: the author of the story of the apple-bearing ash tree may have wrongly identified the Lyde place-name at the mouth of the Wye with that of the lida of Lydney.  He thus mistakenly transferred *Nemetaballa from Lydney to the Wye estuary.

Lydney and Lyde Rock

Lyde Rock

Lyde Rock

If I'm right about this, then 'Avalon' or Nemetaballa may indeed have been the name of the Nodens shrine at Lydney.


Tradition favors Arthur's deposition at 'Avalon', which may well have been the Nodens temple at Lydney.  However, there are three strong arguments against Lydney.  

1) It was a purely pagan site.  Thus, a burial there would imply that Arthur was himself pagan.
2) Archaeology does not favor the continuation of the Lydney shrine into Arthur's period (see below).
3) The West Hill shrine at Uley did continue - and as a Christian church - and is situated right next to the Uley Bury hillfort, my candidate for Camlan (*Cambolanda).

In my mind, the only way we can justify selecting Lydney over Uley is if Arthur were pagan and his enemies had taken Uley.  It would have been necessary with the loss of the West Hill site to remove the hero's body elsewhere.  A boat taken from the River Cam's mouth at Frampton Pill to Lydney represents no difficult feat. 

Alternately, Arthur was, in fact, buried at the Uley church.  The proximity of *Nemetaballa/'Avalon' to Uley on the Cam would have been sufficient to engender a folk belief that he had been taken to the Place of Apple Trees, a mythological construct.


While working on the Welsh Mabinogion hero Terynon Twrfliant, I came across numerous references to this "Divine Lord of the Roaring Sea/Waters/Flood" being a title for Nodens at Lydney - and this despite the fact that all indications were that he was instead to be identified with Manawydan son of Llyr ("Sea").  Terynon's name is present at modern Llantarnam and I've discussed the history of that place in depth elsewhere.  Suffice it to say here that Llantarnam is not near Lydney.

However, at the time I neglected to treat of the folklore motif of the foal-snatching claw/arm in the story of Gwri Gwllt Eurin.  Nodens is often related to the phenomenon known as the Severn Bore:

The Bore can literally sweep away livestock.  It thus occurred to me that the claw/arm that steals foals every May Eve may well be a representation of the Bore, perhaps perceived as an arm of the sea, and that Terynon's act of cutting it off is probably symbolic of the receding of the tide.  Furthermore, the story of the Irish Nuadu's amputated arm is well known, and it is tempting to equate this god's limb with that of the Mabinogion story. 

In Irish tradition, the isle of apple-trees is associated with Mannanan mac Lir, the counterpart of the Welsh Manawydan.  Nuadu is equated with Necht/Nectan, a name cognate with the Latin Neptune.  All of this tends to lend more weight to the notion that the Lydney temple may, indeed, have been at least a regional 'Avalon.' 

"In early Irish literary tradition, Nuadu can be equated
with Nechtan, in view of shared aquatic attributes
and also the combination of the name and the epithet
Nuadu Necht, found in the early Old Irish genealogical
poem, from which an extract is quoted below.
3. Swift in ships, he traversed the sea as a warrior
of the west: a red wind, which dyed sword-blades
with a bloody cloud.
4. Fergus Fairrge, Nuadu Necht, strong and brave:
a great champion who did not love punishment from
a rightful lord."


"Nechtan’s name is cognate with Latin Neptunus, the Roman god of the sea (see also Dumézil, Celtica 6.50–61)." - Anne Holley in CELTIC CULTURE: A HISTORICAL ENCYCLOPEDIA. 

Insula Avallonis (the Isle of Avalon) is first mentioned
by Geoffrey of Monmouth in his Historia
Regum Britanniae (c. 1139) as the place where
Arthur’s sword Caliburnus (see Caladbolg) was
forged, and then as the place where Arthur was taken
after the battle of Camlan for his wounds to be
tended. In the Welsh versions of Historia Regum Britanniae
(Brut y Brenhinedd), the place is called Ynys
Afallach. In Geoffrey’s Vita Merlini (Life of Merlin; see
Myrddin), Insula Avallonis is explained as insula pomorum
‘island of apples’ (cf. Welsh afal ‘apple’, afall ‘apple
trees’). Ynys Afallach thus corresponds closely to the
poetic name that occurs in early Irish literature for
the Isle of Man (Ellan Vannin), namely Emain Ablach
‘Emain of the apples’, a name applied to Man specifically
as the blessed and otherworldly domain of the
sea divinity Manannán mac Lir.



The earliest discerned phase is undoubtedly the Iron Age fort with multivallate
defences. There is no need to follow Wheeler in ascribing the outer ditch and rampart to
a post-Roman context. The only justification for this was that '... the outer bank and
outer ditch may safely be ascribed as on tactical grounds to this period.' Since there is
no evidence for dating the inner rampart to a late period the hypothesized outwork may
resume its place in the original plan of the site.

The second phase of activity is the iron mining. The inception of this work is entirely
undated and it may be entirely pre-Roman. A date for the filling of an adit is given by its
being sealed by a third century structure, but this does not mean that industrial activity
had continued to this date.

The next phase of activity falls in the second half of the third century. The coin list
implies a presence that is paralleled on other sites with occupation of this period. The
dominant character of Coin Period 18 (264-86), a coinage which is virtually extirpated
by Diocletian's currency reforms, suggests that purposeful site activity starts no later
than this date.

The problem is to define what building activity can be ascribed to this phase. The
analysis of the coins from the four principle buildings, the Temple, 'Abaton', Guest
House and Bath suggests that all participate in this period. We have already seen that in
two contexts Wheeler's huts prove to be opus signinum floors antecedent to later mosaic
surfaces. Only the rectangular structure under the tail of the rampart qualifies as a freestanding
building separate from those forming the Temple complex. This may have
been accommodation for builders or people making a living from the shrine. One must
assume that part of the Guest House furnished accommodation for higher ranking
servants of the cult, whether priest or administrator. In any event the lower grade
housing did not survive the heightening of the rampart.

The pottery from the rampart and associated contexts (see Pottery below) is
uniformly of later third to earlier fourth century date, and since it is associated with
building material derived from an adjacent structure it seems reasonable to date that
structure to a date earlier than the reconstruction of the Temple itself after the collapse
of the cella.

Elsewhere the best evidence for specific third century activity comes from the Bath
House where a coin of Gallienus is associated with an opus signinum floor now identified
below what Wheeler considered to be the primary floor of Room XXXVIII. In Room
XXXV to judge from the stratigraphical description recorded on the coin envelopes, coins
of Probus and Tetricus were associated with what seems to be a floor predating that
which bears the mosaics. If this interpretation is correct we have a similar situation to
that revealed by excavation in Room XXXVIII.

The fact that the Bath House corridor abuts the Long Building necessarily makes the
'Abaton' stratigraphically earlier than the Bath House. It could of course be argued that
the corridor forms a later addition to the bath, but there is no archaeological evidence for
this argument. A start for this building before the early fourth century is suggested by the
coins of Quintillus, Carausius, Constantius Chlorus and Constantine II as Caesar found on
the 'hut floor of pounded tile' (ie the earlier opus signinum floor of room XLVlll).
The Guest House is very problematic. Its state of preservation was already very bad
when Wheeler dug it in 1928/9. The hut floor that Wheeler claims to have found in
Room XXX could not be found during the limited re-examination in 1980/1. It could be
argued that the Guest House has to be contemporary with the 'Abaton', since it provides
the other post to the gate of the Temple enclosure. However, no evidence survives to
show that the gateposts were integral to the buildings they abutted apart from Plate
Xlllb. The gate may well be a later addition.

Problems relating to dating the Temple itself have already been discussed. It may be
noted that buildings, or additional rooms, not unlike the 'Abaton' are associated with
other Roman-British temples, eg Harlow and Caerwent. These structures are generally
later additions to an earlier sanctuary and this general observation might reinforce the
topographical argument that the Temple predates the 'Abaton'.

The later history of the site saw an embellishing of the structures, including the
laying of mosaics and the extension of the 'Abaton'. The date of these changes is
usually given as post-364 since a number of coins of Gratian are said to have been
found under the mosaics. Without further excavation involving the lifting of surviving
mosaics this chronology cannot be tested. Wheeler's own definition of 'below the
mosaics' is rather ambiguous; there is evidence that excavations in the nineteenth
century did cut through floors, whether it was the re-excavation of these features that
produced coins or whether newly found floors were lifted, which seems inherently
unlikely, is not at all clear.

It appears that the site did not enjoy the attentions of pilgrims for very long. A
comparison of the coins from four similar sanctuaries; Lydney, Harlow, Uley and
Matagne-la-Grande show that Lydney is much closer in losses in Coin Period 27 to
Harlow (which seems to have been abandoned in the post-Valentinianic period) than
to Matagne-la-Grande and Uley, the latter of which certainly continued into the fifth
century. The only material evidence of a late presence at Lydney is a brooch of fifth
century date. This object has no specific findspot and may well be a stray find
dropped by a casual visitor in the post-Roman period. On the other hand the
reduction of Wheeler's chronology to one more in conformity with comparable sites
elsewhere does not of itself prove that Lydney did not have a function in the period
down to, or even beyond, the occupation of the Forest of Dean region by Saxons. An
inability to recognize a distinctive post-Roman material culture is a characteristic
element of early post Roman studies. Nonetheless the reduction of the chronology
tends to undermine the concept of a 'pagan revival', associated with rural religious
sites, in the period when Christianity was ascendant in both the official and private
spheres of life in Britain.


From Laura Johns, Libraries and Information, Gloucestershire Libraries Enquiry Service:

Lyde Rock lies in the River Severn Estuary near the ‘Old’ Severn Bridge, it had a warning beacon installed in 1896 as described here:

 “LYDE ROCK .The light was established in 1896 to help vessels steer clear of Hen & Chickens Rocks. “ (I have attached the PDF Document that details the beacon on Lyde Rock for your information) so we can safely assume that Lyde Rock has had that name for the last 100+ years.

 I haven’t been able to find a precise etymology for the name, so it has been a process of locating resources that bear the name Lyde in reference to the River Severn and then using the dates that the name was in use to provide the answer to your enquiry.

 I  found a reference to Lyde Rock dated 1573 at British History Online:

 “A fishery in the Severn at Lyde Rock to the north of the Beachley passage was bought by John Philpot in 1573”   British History Online

 On the Gloucestershire Archives website I found the following references to the River Severn ‘Lyde’, in these entries you can see the dates that the name was in use:

Messuage in Bettysley (Beachley); 2/9th parts of Beachley Passage or Ferry; fishing in the R. Severn called the Lyde; cottage called Kitcotes; tenement called Crookes Place.



Messuage in Bettysley (Beachley); 2/9th parts of Beachley Passage or Ferry; fishing in the R. Severn called the Lyde; cottage called Kitcotes; tenement called Crookes Place, with addition of messuage called the Green Dragon, later the Ostrich, the George Inn, at Beachley. [Both PHILPOTT family]



754 putchers at a place called the Salmon, 375 putchers at Lyde Rock


14 May 1866

I have had a look into the name “Lyde” and its origin and meaning is quite difficult to work out.  However, the most likely possibility is:

 Lyde: topographic name from Old English hlið, hlid, Old Norse hlíð ‘slope’.

This would make sense as the rock is a slope emerging out of the river at low tide. 

I’ve also had a look through the reference books we have here at Gloucester Library and there is no mention of Lyde Rock at all. There seems to be very little in the way of information regarding this name, and no information on how the rock came to be called Lyde Rock. The earliest mention of it I can find is from 1573 as shown above at British History Online.

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.