CAREW/1

The cross was revisited by Rhys in 1894, and again in1895, and by the Royal Commission in 1923.

Macalister/1949, 169: `This monument...now standing in a recess by the roadside, facing the old castle of Carew. It appears to have been close to its present site since 1822 at least: in recent years it has been set in a recess to guard it against possible injury from motor traffic'.

Macalister/1949, 169--170: `a sculptured cross...The shaft measures 9' 0" x 4' 0" x 0' 10" in maximum dimensions. The cross which it supports is in a darker coloured stone, of smaller horizontal dimensions, and of inferior workmanship: quite possibly it is not the cross for which the shaft was originally intended'.

Nash-Williams/1950, 182--184: `Free-standing composite slab-cross, comprising a combined wheel-head and splayed and shouldered neck, with tenon below, mounted on a tall splayed shaft with expanded butt. 162+" h. overall. Shaft, 115" h. x 50" w. (above butt), reduced at a height of 22" by a double offset to 38" and thereafter diminishing to a final width of 25" x 13" t. at base diminishing to 7--8" at top. Neck, 19 1/2" h. x 24" w. at joint, reduced to 18 1/2" by a double step 8" above and thereafter narrowing to 17 1/2" at junction with wheel-head, x 4" t. Head, 25--7" diam. (overall, with the arms projecting 2" beyond wheel) x 4" t. The wheel-head, of Anglian type, and the shaft are decorated on all faces with carved patterns in low relief (partly defaced by flaking); the shaft is also inscribed...Like No. 360, the monument is of special interest as one of the largest and most elaborate of the Welsh crosses, as well as one of the best preserved. The form represents the culminating development of the composite slab-cross, represented elsewhere in S. Wales, by Nos. 159, 222, and 360, to which the present cross stands in close stylistic relationship. Like most of the larger and later Welsh crosses, the monument is characterized by its faulty proportions, the tiny head being clumsily perched on top of a disproportionately tall and wide shaft. It thus lacks at once the slender grace of the best of the Anglian monuments, and the massive shapeliness of the Irish `high crosses'. The decoration comprises a limited range of patterns -- broken plaits and knot-work, T-frets, and swastika and diaper key-patterns -- of Celtic type but showing indirect Scandinavian influence (looped pattern); the execution is fairly good, but is characterized by coarseness and irregularity, a tendency that becomes increasingly marked in all the later S. Wales crosses. These features combine with the use of late details -- ring-twist, looped knotwork, and pellets -- to place this monument somewhat later than No. 159, and more or less contemporary with No. 360'.

Westwood/1879, 119--120: `On the west side (which has been more weathered) the design is nearly defaced, although a small portion of a ribbon-interlacement can be observed on this side. The two chief faces of the base of the cross are divided into compartments, each with a different style of interlaced ornament, of which an inspection of the figures will give a clearer idea than a detailed description. On the upper part of the east side (fig. B) will be observed the curious pattern formed of groups of four T's arranged with the bottoms of the down-stroke radiating into a geometrical pattern; which also occurs on the west side just below the wheel of the cross. The middle portion on this side has a modification of the curious Chinese pattern, in parts of which the sculptor had made several mistakes in the carving. This design, which also appears on the south and east sides of the Nevern cross (Plate LXII), is in fact a slight modification of a series of filfot crosses united together by straight raised connecting bars.

In the lower dilated part on this west side is a fascia inclosing a very classical fret; and below the middle are two tranverse spaces, each measuring 11 inches by 6, the right-hand one being quite plain, and the left-hand one having an inscription'.

Macalister/1949, 170: `[The head] is decorated on the face with poor frets and zigzags, but has no ornament on the edges. The shaft has a running guilloche ornament on each edge. On the face turned toward the road there are four panels: in the topmost is a ket-pattern founded on a swastika; in the next there are three groups of pairs of oval loops intersecting; and below that are two panels of badly designed interlacement. On the opposite face there are five divisions: in the topmost an irregular, unsymmetrical panel of interlacement; in the next a very badly drawn key-pattern; then come two smaller panels, side by side, the dexter one containing the inscription, the sinister blank; then a plain fret, with one breach in its monotony, and finally a narrow `wall of Troy' pattern'.

Nash-Williams/1950, 182--184: `Front. The head was apparently filled with double-beaded knotwork (mostly flaked away), comprising two-cord twists... ending in Stafford-knots filling the arm-ends (cf. No. 360). The decoration of the neck and shaft is disposed vertically in moulded panels, two on the neck and five on the shaft: (a) two squares of double-beaded swastika T-frets; (b) three squares of diaper key-patterns); (c) fourteen-cord double-beaded plaitwork with irregular breaks and incorporating a row of ring-twist a characteristic Viking Age motif); (d) triple vertical band of irregular diagonal swastika key-pattern, with interspersed pellets; (e) double horizontal panel-left-hand compartment, inscription (see below), right-hand, blank; (f) twelve-cord (changing below to fourteen-cord) double-beaded plaitwork with single break (bottom l.) and single pellet (bottom r.); (g) horizontal band of double-beaded T-frets, flanked by short vertical square frets filling the shoulders of the butt.

Right and Left. The head and neck are plain. The shaft is decorated with a narrow vertical panel of four-cord double-beaded plait.

Back. The cross-head is filled with a combined outline and linear cross with fretted expansions to the arm-ends. The decoration of the shaft is disposed vertically in panels, two on the neck and four on the shaft: (a) incised quadruple triangular knot or cruciform panel; (b) narrow horizontal band of opposed moulded triangles; (c) double horizontal row of three squares of plain swastika T-frets; (d) row of three pairs of interlinked plain oval rings; (e) fourteen-cord double-beaded plait with irregular breaks and ring-twists; (f) irregular eight-cord double-beaded looped plaitwork with interspersed pellets, a characteristic Viking Age pattern, found also on No. 190. The lateral expansions of the butt are filled with short vertical square frets'.

Nash-Williams/1950, 182--184: `Front...(e) double horizontal panel -- left-hand compartment, inscription (see below)...The inscription (Fig. 195) on the front of the monument is in three lines reading horizontally'.

Radford/1949, 254: `The documents copied into the Book of Llandaff show that one of the normal forms of signature used by the kings of Morgannwg as witnesses to grants was: A rex filius B. A less usual variant was A rex B filius, and in this particular source it appears only in documents purporting to be earlier than 900. The same forms of the royal title occur in the text of the documents. The spelling of the names on the Carew cross may be compared with those of witnesses in the same series of grants, but a closer parallel to the first is to be found in the pedigrees in Harley 3859, a twelfth century copy of a tenth century MS'.

margit

eut re

cett f...

Possibly one should read Margiteut Recet g F...The former would make Margiteut Recett fecit, or `Meredyd of Rheged made it''.

Rhys/1896, 107: `We could see no reason for reading n instead of u in Margiteut; but he suggested that at the end one should read F X. for fecit crucem; and I should be inclined to do so, but I should be glad to be assured that x ever occurs in that capacity'.

Macalister/1949, 171: `The inscription is in badly drawn letters, and reads:

MARGITEUT EIT RECEN

In the accompanying diagram the letters of the three words are shaded differently, for purposes of distinction. They show the same fault of lay-out that we have already seen on the Tecuri Faenor stone (331), explicable in the same way, though I did not follow out the process completely till I had seen Mr. Nash-Williams's drawing of the inscription (Arch. Camb. 1939: 13) and compared it with my own photograph. This drawing revealed to me the last letter as an ill drawn N, which I had never been able to trace to my own satisfaction till then. The scribe must have been given a model on a writing tablet set out thus:

MARGITEUT

ET RE

CEN

Then by some mental confusion he mistook the EUT for an ET which he ought to have written, and proceeded to write the RE following. This filled his second line, and then he discovered that he had left out ET. He added it underneath, spelling it with double T, either from mere illiteracy or under the influence of the EUT written first above. The last three letters he filled in where he could, the C in the blank before the ETT, the EN in the blank after it. The N in the model must have been composed of two minims, concave in opposite senses [thus, )( ]; and the scribe brought them into contact back to back, turning the letter into an X, adding, as a freak of his own, a wavy line below, to decorate it with a serpentine flourish...Not the least remarkable fact about this inscription is the existence of several copies of it...Two other copies exist at or near Fethard in Co. Wexford'.

Radford/1949, 253--255: `The end of the lettering is cramped and obscure, but the interpretation generally accepted has been that the panel contains the maker's name, Maredudd of Rheged. This solution is not free from difficulties...The drawing...reveals the following transcript: --

MARGIT

EUT . RE

X . ETG(UIN) . FILIUS

Line 1. All the letters are clear, and the reading has not been disputed. The small final T, crushed in the corner, shows that the designer was already in difficulties with the spacing and foreshadows the expedients he will be forced through having started the inscription on a scale too large for the space available. The R, with a wedge-shaped serif at the top of the upright and a line linking this to the base of the loop, is an elaborate, confused form, of which the essential features are repeated in the fourth letter in the next line.

Line 2. All the letters are clear and the reading has not been disputed. The displacement downwards of the R, to avoid the loop of the G in the line above, increases the confusion of the arrangement.

Line 3. The first letter has been generally read a C; the strokes running left from the loop, though poorly cut and shallow, are clear; it is an X. The second and third letters have generally been accepted as ET. The fourth letter has often been read as T; the bend downwards of the end of the loop is clear, and the backwards sweep of this tail, though shallow, is unmistakable; it is a G, a cramped and rather simpler version of the form in l. 1. It would appear that the cutter of the X and G, fearing to break through the margin of the panel, contented himself with a very shallow indication of those parts of the letters, which approached most nearly to the frame. The next symbol is a ligature FIL, represented by an L pendent from the lower bar of the F. The IU, also ligatured, are set rather above the line, the I continuing upwards the left stroke of an angular U; the junction is marked by a slight projection on the right giving the I the same form as in line 1. The S is of the Roman capital form, squeezed thin, the upper loop coalescing with the base of the U. The use of this type of S in Insular majuscule inscriptions is not unusual; and example may be cited from the slightly later memorial at St. David's.

The end of each word is marked by a dot; the first two tend to coalesce with the loops of the preceding letters and can only be distinguished on a close examination of the original; the last two are clear. This method of punctuation occurs on other contemporary inscriptions, including that already mentioned at St. David's.

A blank panel remains on the shaft beside the inscription. This was intended for a second name, either that of Maredudd's brother Hywel or of some ecclesiastical witness to the grant which the cross commemorated. The fact that this was never cut may be perhaps the result of the sudden death of Maredudd at the hands of his enemies'.

Nash-Williams/1950, 184: `Round half-uncials...The looped form of the R's is distinctive'.

For line 3, Radford goes on to offer a different reading from that usually offered.