Archaeology Magazine - December 1951 - 6

tops of the letters is close to 1.29 m. in length.
The inner inscribed band, measuring 0.055-0.060
m. in width, is half again as wide as the outer inscribed
band; the tallest vertical strokes of the letters run to
0.035-0.042 m.; there is an even more comfortable
margin for them, and they are carefully formed. The
inscription in the inner band is at the bottom close to
0.72 m. long, at the top only ca. 0.47 m.
Text. The inscriptions are retrograde throughout.
Following Heiss and Huebner, I have arbitrarily begun the numbering of letters at 6 o'clock, and have
given every element that surely is, or could possibly be
construed as, a separate symbol. Thus each large letter,
each small letter, and each interpunct is numbered, so
as to be specifically referred to. The numbering should
not prejudice the question whether, for instance, 72
and 73 are one letter (I think they are meant to be) , or
again 74 and 75 (more doubtful: a compound symbol
apparently; compound symbols, when referred to as
one united whole, are printed with a slant, e.g.,
74/75) , or whether some of the smaller symbols were
thought of by the forger as punctuation.
On this basis the numbering of symbols in the outer
band runs from 1 to 95. Since totals are of little meaning, the inner band begins at 101, and runs to 143.
This saves the awkwardness of Huebner's
italicized
letters, as in a 23, b 23 (where I have 23 and 123).
The starting-point, however, is precisely that of Heiss.
Relation of Outer and Inner Bands. Out of a list of
twenty-five symbols which occur on the plates more
1The
longest is No. XXII, the Lammina Castellanensis,a
lead (curse?) tablet found in a grave; about 20 letters longer.
2
167-170;addendum, page 214; region, page 166.
3Pages
Conceivablythe SandwichPlate could be one of those acquired by Engel, but the wrappings on the Sandwich Plate
were very old and motley, suggesting rather that it had
passed through humbler hands prior to any date for the likely
dispersal of the Engel collection.
The Segoviaand Sandwich Plates are certainly not identical. Heiss speaks of the Baron plate as being "recouvert d'un
vernis noir," whereas the Sandwich Plate is covered with a
glaze which is not black in any area larger than a thumbprint, but is rather, over almost the whole area, various shades
of dirty brown. Further, in Heiss' photograph the letters
from 5 to 6 o'clock, and from 12 to 1 o'clock, are obscure, as
if from rubbing; whereas in the Sandwich Plate they are all
clear. The central medallion of the Baron Plate seems unobscured,whereas in the SandwichPlate it is heavily encrusted.
There are differences also in preservation. Most notably,
the Heiss photograph shows a crack at 9 o'clock running
from the outer edge through the outer rim, and through
the second rim. The Sandwich Plate is intact in this area.
Similarly, a number of tool-marks in the floral band of
198

than once, sixteen occur in both the outer and the inner
bands. The remaining nine are confined to the outer
band. Five of the compound symbols (i.e. simple symbols united by ligatures) occur only in the inner band.
Apart from these five compound symbols, and apart
from five small point-like symbols of uncertain relationships, all the symbols of the inner band, thirtythree, are found also in the outer band. At least one sequence (51-52, 119-120) is identical in both bands,
and one of the very few elements clearly repeated in
any two ligatures is identical in 7 and 102.
Ligatures. The symbols which are compounds, i.e.
involve ligatures, are at most seventeen in number:
5(?), 7, 26(?), 28/29, 39/40, 58, 61 (?), 65(F),
66(P), 68(P), 74/75, 77, 102, 108, iio(?), 112/113,
Of these, eight are certain; nine are unii7/ii8(?).
certain. Of the whole seventeen, four (7, 28/29, 39/4°»
112/113) can be analyzed into known components.
Conclusion. To judge from the general ignorance
about so conspicuous an object as the Patera Segovinensis, Celtiberian studies - at least studies of the inscriptions, alphabet, and language - offer a considerable opportunity for scholarly research. One obvious task is
to disinfect Huebner's
version of the alphabet, rethose
not
(I think,
moving
numerous) shapes which
he had admitted from the Segovia Plate. The larger
task is a new collection of Celtiberian inscriptions, preceded by a search for additional texts throughout Spain,
and abroad. One unpublished text, on a hand of bronze
in the Cabinet des Médailles, Paris, is known to me.
the Sandwich Plate, made before the glazing, and a few
nicks in the outer edge of the SegoviaPlate, find no duplication in the other plate.
4
Forgeries of Iberian inscriptions are known. Heiss, who
was well acquainted with conditions, commentedupon forgeries made in Spain of Spanish pottery, and rejected the
notion in the present instance (pages 312-313). Huebner's
Monumenta has a section ("Falsae vel suspectae",pages 205212) in which he collectsall the forgeries known to him of
Iberian inscriptions.There are 62: all are much shorter; none
is a terracotta plate in any respect comparable; none is from
Huebner treated the Segovia Plate as genuine.
Segovia.
6Such tests as I have been able
to make, working without
special knowledge, appear to reveal some quite impossible
sequencesof consonants. The outer band as a whole could
hardly be a direct copy, even if all the shapes could be substantiated. The inner band has far more invention.
6This is done with the kind
permission of Dr. John O.
Brew, Director of the PeabodyMuseum.The author is grateful to him, and to the other scholars named supra; to Professor Joshua Whatmough for particularly helpful advice;
and to Mr. F. P. Orchard for technical assistance.

ARCHAEOLOGY

Archaeology Magazine - December 1951

Table of Contents for the Digital Edition of Archaeology Magazine - December 1951