MMWR3/1

Westwood/1879, 17; `It stands, like the preceding, in the garden of Mr. Nicholl's residence'.

Rhys/1899, 156: `The third day we went to Merthyr Mawr, on the banks of the Ogmore, and examined the two stones in the grounds. The first is the Great Cross, standing by the ruined chapel in the grounds'.

Rhys/1905, 38: `The great cross at Merthyr Mawr comes from a spot near Witney farm, between Merthyr Mawr and Laleston: it stood about ten yards from the fence not in it. This is its history, as kindly corrected by Mr. J. Illtyd D. Nicholl, the squire of Merthyr Mawr, who was good enough last year to have this stone and another near the house buried in the ground so as to clean them of lichen. He invited me last August to come and re-examine both; and the result as regards the present stone was that I was able to read more than before, and to correct at some points what I had guessed on my previous visits'.

Macalister/1949, 161--162: `To the E. of a small chapel in the grounds of Merthyr Mawr house...The stone is locally called the `Goblin' stone: according to Evanson (1909) it stood `in the second field through which the path from Whitney Farm passes to Laleston, near a stile in the south fence, about 20 yards from the S.E. corner' but was removed about the thirties of the last century to its present site. Westwood (Lap. Wall.) says that this stone was No. III [MMWR2/1]; but Evanson, having been rector of Merthyr Mawr and therefore presumably acquainted with the local traditions, is more likely to be correct. The field, he says, is called by the farmer Cae y garreg lwyd (Field of the Grey Stone) and on the Tithe Map Cae'r groes (Field of the Cross)'.

Nash-Williams/1950, 154: `Near Whitney Farm'.

RCAHMW/1976, 57: `originally standing to the N. of Merthyr Mawr,[1] and subsequently moved (with No. 927) to the grounds of Merthyr Mawr House where it has been re-set inside St. Roque's Chapel.

[1] Trans. Cardiff Nat. Soc., XLI (1908), p. 25; Stowe MS. 1024, fols. 20-1, `in a field called Kaer Groes'; drawing shows cross-head complete'.

Macalister/1949, 161: `a slab surmounted by a cross pattée, having triquetras on the arms'.

Nash-Williams/1950, 154--155: `Massive free-standing slab-cross with debased wheel-head of Anglian type (fractured away on the left) and broad shaft (partly fractured away on the right) with moulded angles tapering slightly to the foot. 88+" h. (overall) x 32--3 1/2" w. and 11" t. (shaft); head, 33" h. x? 36" (full) diam. x 8 1/2" t. The cross is decorated on all faces with, lightly carved or incised patterns, and is also inscribed...The heavy angle-mouldings of the shaft suggest direct affinities between this monument and the local group of debased moulded pillar-crosses (Nos. 193, 205, 206, 252)'.

RCAHMW/1976, 57: `Monolithic shaft and ringed cross-head. The shaft and head are formed from one block of coarse sandstone, standing 2.2 m above ground, much weathered and lacking part of the head and a section out of one side. The shaft, tapering slightly towards the base, is 91 cm at its widest and 26 cm thick. The original diameter of the cross-head was 96 cm, its width at the neck being 84 cm, and it is 21 cm thick. On both main faces (E. and W.) of the cross-head the decorated splayed arms with hollowed angles are linked by a ring of plain square section recessed below them; the lowest member spreads out to the full width of the shaft, with a narrow band representing the neck'.

Rhys/1905, 38: `the middle of the stone is worn out...it looks as if the stone had been used as a threshold for a long while some time or other. Another suggestion made to me was that the wear was effected by sheep rubbing against it for centuries. I do not know which theory to prefer: in either case the result is greatly to be deplored'.

Nash-Williams/1950, 154: `wheel-head...(fractured away on the left) and broad shaft (partly fractured away on the right)'.

Westwood/1879, 17: `In the centre is a raised boss, and the limbs are boldly ornamented with triquetrae formed of wide ribbons united together in the centre. Below the cross are two transverse bands with cross raised lines, the middle of the upper band on the back of the cross having a simple ribbon pattern...The back side of the basal part of the stone as well as the edges are ornamented with various patterns formed of incised lines, which have however become as much defaced, especially in the middle, as the inscription itself. It will be seen from what remains that the designs formed quadrangular compartments arranged irregularly, and filled in with the Chinese-like pattern common on these stones'.

Macalister/1949, 161: `The cross has interlacing (triquetra) work on the front, key-pattern on the sides. The back of the stone is covered with key-pattern, as is also the edge of the dexter side of the inscription. The sinister edge has interlacing above, key-pattern below, with a fractured gap between them'.

Nash-Williams/1950, 154--155: `Front. The cross-head, with widely splayed arms and plain recessed wheel, is filled with triquetra-knots (R.A. 798), partly plain and partly double-beaded, interjoined around a central raised boss. The short neck joining the head to the shaft is decorated with vertical ribbing. The shaft is filled with a moulded vertical panel, divided by an incised transverse line into two unequal halves (cf. No. 239), containing a Latin inscription...Right. The arm-end of the cross-head is filled with a panel of irregular straight fretwork (damaged). The neck below is ribbed (as before). The shaft (damaged) is filled with a narrow vertical panel containing a two-cord double-beaded twist (R.A. 501) above and a triangular fret (cf. R.A. 924/6) below. Back. The head is filled with double-beaded triquetra-knots (R.A. 798) interjoined around the central raised boss (as before). The neck below bears a horizontal plain two-cord twist (R.A. 501) of four lobes in the centre flanked on either side by vertical ribbing, which is continued across the angle-moulding separating the neck from the shaft. The shaft is filled with a double vertical band of eight squares of variegated swastika key-pattern (cf. R.A. 1006), partly debased. The angle-mouldings of the shaft are marked off at the bases by triple neck-mouldings with scroll-work or foliated (?) ornament above. Left. The arm-end of the cross-head is fractured away. The neck below is ribbed. The shaft is filled with a narrow vertical panel comprising five (?) squares of nondescript cruciform frets (?)'.

RCAHMW/1976, 57: `The arm-ends were probably all originally decorated with the pattern of straight frets surviving only on part of the N. arm-end. The carved faces of the shaft are framed by upper and lower edge-moulding and by bulbous angle-mouldings with much-weathered traces of carved decoration (? plaitwork) between roll-mouldings of three bands at the head and base.

On the E. face or `front', the decoration of the cross-head consists of double-beaded triquetra knots in the upper arms which in the lower member resolve into plain beading incorporating a single pellet, all linked around the central plain raised hub and double-beaded ring; the edge-moulding is also double-beaded. Below the straight lower edge the brief neck band is ribbed vertically. The face of the shaft forms a panel framed by double-beaded moulding and is entirely filled with a much-weathered Latin inscription...The narrow vertical panel on the S. side of the shaft is much weathered but has traces of a pattern of twists and frets, the lowest section forming a square of diagonal swastika key-pattern.

The W. side of the cross-head has double-beaded edge-moulding, within which the arm-ends contain double-beaded triquetra-knots linked around the central damaged hub without ring. The neck-band has centrally a plain two-cord twist of four lobes set between vertical ribbing. The upper edge-moulding of the shaft has alternate bands of vertical and horizontal ribbing in contrast to the plain double-moulding on the E. side. Eight irregular panels of diagonal swastika key-patterns, framed and divided vertically by a double-beaded moulding, fill the face on the shaft. On the N. side, the arm-end of the cross shows traces of straight fretwork, while the neck is boldly ribbed vertically. The incomplete panel of the shaft has above the missing section a double-beaded loop and twist, and below it a similar loop separated by horizontal ribbing from triangular fretwork at the foot'.

Nash-Williams/1950, 155: `The shaft is filled with a moulded vertical panel, divided by an incised transverse line into two unequal halves (cf. No. 239), containing a Latin inscription (partly defaced---Fig. 170) carelessly set out in 15 (?) lines reading horizontally'.

RCAHMW/1976, 57: `The face of the shaft forms a panel framed by double-beaded moulding and is entirely filled with a much-weathered Latin inscription, apparently in sixteen lines divided (six and ten) by a horizontal chace'.

Rhys/1899, 157--158: `So far as one can judge, this monument was free from the blunders characteristic of the cross of `Samsoni Apati', and is rather to be compared in point of language with Houelt's Cross; for I should put grefium for grefio down to the charge of late Latin. But we have probably a touch of the local pronunciation in grefium: that, I think, is the reading, and not grafium, but I am not certain. This word grafium or graphium in Welsh Latinity meant a writing, charter or document (see Seebohm's Tribal System, pp. 211, 212, 217), and the changes of sound which it underwent in Welsh to become graiff may be represented by graphium, grephium or grefium, greifium, greif or greiff, graiff. Of course if one reads santdi we have another touch of Welsh pronunciation. At all events, this cross probably belongs to the same period as that of Houelt.

Lastly, the words in proprium and in grefium seem to show that in this inscription we have a reference to a formal document commemorating some such transaction as the gift of a piece of land to a religious community. Compare a similar document on the Llanllyr Cross, Arch. Camb., 1896, pp. 119--25.

As to the formula I(n) nomine Dei Patris et Fili Spiritus Sancti, we have had it already on the Cross of Houelt, and it seems to mean, `In the name of God the Father and of the Son the Holy Spirit'. have shown both to Mr. F. C. Conybeare, who has devoted to them a brief notice in his Paulician work called `The Key of Truth'. Among other things he says: `This formula takes us straight back to The Shepherd of Hermas, in which the Son of God is equated with the Holy Spirit, and it also exactly embodies the heresy of which Basil deplored the prevalence in the eastern regions of Asia Minor. These inscriptions, therefore, rudely disturb the ordinary assumption, that the early Celtic Church was `catholic in doctrine and practice; as if Bede had meant nothing when he studiously ignored St. Patrick and denied that the British bishops even preached the Word'. He says of the other formula, In nomine Dei Summi, that it seems to be both praetrinitarian and connected with the series of inscriptions in honour of [some Greek], found in Asia Minor, and referred by Schurer and Franz Cumont to Jewish influence. He ends his note with the words: `The survival of such formulae on these old Welsh crosses explains why Bede rejected the baptism of the British Christians, and why Aldhelm (A.D. 715) denied that they had the Catholicae fidei regula at all'.[1] The absence in our inscriptions of anything explicitly trinitarian is remarkable, and I should like to see the question more fully discussed. Lastly, I am reminded by the Editor of the remarkable fact that no representations of the Trinity, or symbols of the Trinity, occur in Celtic art.

[1] Mr. Conybeare has since had a paper on the `Welsh Church' printed in the Transactions of the Hon. Society of Cymmrodorion'.

Rhys/1905, 40--41, discusses the metre of the inscription and then goes on to state: `Lastly, as to the nature of the transaction referred to as committed to a written document, here called grefium, that is graphium, I have no doubt that it was a legal acquisition by the deceased of the plot of ground which was to be his burial place, where this cross, probably of the seventh or eighth century, was originally set up. Plenty of illustrations occur in the Christian epitaphs of the Continent, such as those of Rome collected by De Rossi. We have another Celtic instance, to be mentioned shortly, in the case of the Llanllyr stone, where a saint is also made a party to the transaction. The country round Merthyr Mawr was decidedly within the sphere of the influence of St. Iltutus or Illtud, but the state of the legend does not enable one to define the part which the saint was supposed to act through his successor at the time here in question: probably the transaction would not have been considered valid without the latter's express approval'.

Macalister/1928, 309--310, discusses the implications of this inscription for the theology of the Welsh church.

Macalister/1949, 161: `The legal terms at the end, indicating a perpetuity of appropriation of the site of the monument, are interesting'.

Nash-Williams/1950, 155: `The purport of the inscription is apparently to dedicate some property by deed to a church or monastery (cf. No. 255)'.

RCAHMW/1976, 57: `The stone would thus seem to have recorded the transfer of land to a religious body, as No. 926.[2]

[2] Cf. similar phrases in charters to Llancarfan, Vitae Sanct., pp. 126, 128, 132'.

From the effects of time or ill-usage the greater portion of this inscription is illegible, a few words at the top as well as at the bottom being only decipherable, although I cannot but think that with a good cast held to the light at different angles, an additional portion might be made out. I can only however distinctly read --

i nomine di pat

ris et fili speri

tuf...

...

...

...

...

...a...eus

post...caisto...gre (a?)

ciam aD pro

prium...

in diem iudici

The latter part of the third line is I think intended for the word sancti, written in long ornamental letters, as was usual in some of the early MSS. of these countries, copied in my Palaeographia Sacra and work on the Miniatures and Ornaments of Anglo-Saxon and Irish MSS.

In Iolo Morganwg's MS. reading (communicated by J. C. Nicholl, Esq.) the three top lines are read --

I nomine di sum c

rux FiLIus eri

cus

and the four bottom lines --

I . PosUit...coisto...

eiUS m . in pro

prium

inem . IUdIs

Taking Filius Ericus as a proper name, Mr. E. Williams (Iolo M.) referred the cross to `a prince of Glamorgan named Eric who lived in the second century, whence it may be fairly inferred that the above are the oldest British inscriptions the dates of which can be clearly ascertained of any hitherto known'.

If the fifth line from the bottom could be read `I . posuit' there is space for the letters i n l o, or at least for l o, which would read `posuit in loco isto', but I fear that the inscription will not bear out this suggestion.

Dr. Petrie's MS. reading of the top and bottom of the inscription, also communicated by Mr. Nicholl, is as follows --

I nomine di~ pat

ris et Filii G SPI

tus

...

....

i~.post le .l.

Soisto . IRGCV

s.ibm.in pRo

prium . iUfn

In dIem lUdici'

Rhys/1873, 8--9: `The reader need not be told that all this is very uncertain, and some of it decidedly incorrect, but I can make nothing better of it'.

Rhys/1899, 156--157: `The inscription is on the front, filling a large panel, which is subdivided by a slight groove across, though the legend appears to be continuous, as in the case of the Great Wheel Cross at Margam. But the whole is so far gone that we could not feel sure even of the number of the lines in the middle.[1] This is all I can make of the lettering, even with the aid of the squeezes which Mr. Thomas took:--

...The d of di seems to have a mark of contraction above it, and the traces of letters in the third line after tus, according to Westwood's drawing, seem to make the spelling Santdi, which we have had already for Sancti on the Cross of Houelt; but the stone itself suggests to me santi. The fourth line from the bottom seems to read saisto or caisto, which seems to make isto and the end sa or ca of a preceding word: we thus have isto in grefium for isto in grafio -- try terra inclusa isto in grafio. The word usque is partly abbreviated usq: but the reading is pretty certain. What can be read runs approximately thus: I(n) nomine di~ patris et fili speritus santi...in grefium . in proprium . usque in diem iudici.

[1] Assuming the lines to be equally spaced throughout, each line would be 2 ins. high and the space between the lines 3/4 in., and there would be six lines in the upper panel and ten lines in the lower one -- ED'.

Rhys/1905, 38--40: `My guesses are as follows, with the lines numbered for the sake of reference; but I must confess that I am not absolutely certain of the original number of them -- the doubt is as to those below the dividing groove and the line which I have prefixed 11...It is hardly worth the while to discuss these guesses in detail, but the following notes may prevent them misleading anybody:-- The s is everywhere of the angular gamma form: the contraction of dei has the line over the d. fili speritus is certain, and the adjacent seems to be spelt sagti, there is no room for an n. The ha at the end of line 3 is fairly certain, but not so nc cruce of the next line, which those letters would rather crowd; perhaps if I have hit on the right words the spelling was hancrucem, just as we have inomine for in nomine. pro an is a guess. The t in line 5 may be a g. Line 7 seems to begin with n or si, but the only letter I could feel certain about is the e. Line 8, fuit was not sought for, but suggested by the guesses which I jotted down. Line 9, the lle seemingly belongs to ille, and Iltut is fairly certain, but I was unable to detect the case ending, which I regret, for the point after ille does not favour ille Iltutus. Line 10 is all uncertain, but it should end with a word whose ending i begins line 11. Posuit. se. suggested to me posuisse and possit esse, but I could not fit either of them in. Loco has the peculiarity that the o is attached to the rounded l, or is rather continued from it: the same thing happens to the o following the c, to which I may add that in this instance the o is an oval placed horizontally, so that I took it formerly for an a. In line 13 the u of grefium, more usually graphium, is a sort of intermediate form between u and o. In line 14 usque was written usqe~. In the last line the c is distinctly angular, and the lower part of the e is angular also throughout the inscription, while the top is more that of a minuscule'.

Anon/1928, 371: `The stroke over the `T' at the end of the first line serves as a hyphen: I have seen such a mark elsewhere. I do not think that the horizontal line cutting the inscription in two parts after the sixth line is intentional -- though the similar line on the companion stone certainly is. On this stone I take it to be a mere flaw'.

Macalister/1949, 161: `After the sixth line there appears to be a horizontal stroke, but this is merely a flaw in the stone: it does not cross the whole panel, not does it correspond to any break in the sense. The whole inscribed panel measures 4'1" x 2'3"'.

Nash-Williams/1950, 155: `Round half-uncials, with normal abbreviations (by contraction and suspension). The lettering is punctuated with stops formed of single medial points'.

RCAHMW/1976, 57: `The inscription in rounded half-uncials with squared E-s and abbreviated `et''.

Macalister/1949, 161: `Some of the writing in the middle of the inscription is worn beyond hope of recovery'.

RCAHMW/1976, 57: `much-weathered Latin inscription'.