Pedestals as 'altars' in Roman Asia Minor Author(s): J. J. Coulton Source: Anatolian Studies, Vol. 55 (2005), pp. 127-157 Published by: British Institute at Ankara Stable URL: http://www.jstor.org/stable/20065539 . Accessed: 30/05/2011 07:03 Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use. Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at . http://www.jstor.org/action/showPublisher?publisherCode=biaa. . Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission. JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. British Institute at Ankara is collaborating with JSTOR to digitize, preserve and extend access to Anatolian Studies. http://www.jstor.org
32
Embed
J. J. Coulton Anatolian Studies, Vol. 55 2005, - Copie
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Pedestals as 'altars' in Roman Asia MinorAuthor(s): J. J. CoultonSource: Anatolian Studies, Vol. 55 (2005), pp. 127-157Published by: British Institute at AnkaraStable URL: http://www.jstor.org/stable/20065539 .Accessed: 30/05/2011 07:03
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unlessyou have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and youmay use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .http://www.jstor.org/action/showPublisher?publisherCode=biaa. .
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].
British Institute at Ankara is collaborating with JSTOR to digitize, preserve and extend access to AnatolianStudies.
The Greek word bomos usually means 'altar', but in inscriptions of the Roman period it sometimes refers to statue
bases and other forms of support, where the meaning 'altar' is not appropriate. Many scholars believe that in addition
to its normal meaning of cult or votive altar and (by extension) funerary altar, bomos could also mean a pedestal, socle
or platform in general. This paper examines the use of the term bomos in Roman Asia Minor for statue bases, for
pedestals for sarcophagi, ash chests and columns, and for other structures which are not altars, concentrating particu
larly on their shapes. It concludes that in all these cases the element called bomos had the shape of a normal type of
altar, and that in many cases (but not all) it also carried some of the symbolic value of an altar.
?zet
Yunanca bir kelime olan bomos genelde sunak anlamma gelmekle birlikte Roma D?nemi'ne ait yazitlarda bu kelime
bazen heykel kaidelerine i?aret etmekte ya da diger destek bi?imlerini de i?ermektedir. Bu durumda sunak s?zc?g?
dogru bir kullamm te?kil etmemektedir. Bir?ok bilim adami bomos s?zc?g?n? k?lt ya da adak sunagi ve -i?erigi
geni?letildigi takdirde- g?m? sunagi olarak yorumlamakla birlikte bu terimlere ilaveten genel olarak kaide, s?tun
kaidesi ya da platform anlamlanni i?erdigine de inanmaktadir. Bu ?ah?ma, Roma D?nemi'nde Anadolu'da bomos
kelimesinin heykel ve lahit kaideleri, k?l sandiklan, s?tunlar ve sunak olmayan diger str?kt?rleri kapsayan kullanimim
inceleyip ?zellikle bu str?kt?rlerin ?ekilleri ?zerinde yogunla?mi?tir. Sonu? olarak s?z? ge?en b?t?n durumlarda
bomos olarak adlandinlan ?ge normal sunak tipi ?eklinde olup, her zaman olmamakla birlikte bir?ok durumda sunagin bazi sembolik degerlerini de ta?imi?tir.
The
word ?coMOc (bomos) normally means 'altar'
(Couilloud-Le Dinahet 1991: 119), but sometimes
appears in epigraphic contexts where that sense seems
inappropriate. Robert (1950: 202-03, no. 204; 1978a:
404, n. 7) argued that in addition to its well-established
extension from cult or votive altar to funerary altar (Drew Bear 1972a: 64), bomos could still in the Roman period mean simply a pedestal, socle or platform. This general sense certainly occurs in Homer (//. 8.441; Od. 7.100), and is supported by the lemma in the Souda (s.v.):
?coiioi- outc?? ??yovTcn ai ?aaeic, crn?aSec
(bomoi: bases (or) heaps? are so called). Etymologically bomos probably derives from the same root ba- as the
more colourless basis or bathron (Chantraine 1968
1980: s.v. ?conoc). Robert's position has been followed
by a number of scholars (Kubiriska 1968: 73; Hellmann
1992: 74; Milner 1998a: 19; Nolle 2001: 490, n. 107;
compare also Drew Bear 1972a: 65-66; 1972b: 190), but
an examination of the shapes of the blocks or structures
concerned suggests that the usage was more limited. It
will be argued that where an element called bomos was
carrying something, it had the shape of an altar, and
might also carry some of the symbolic value of an altar.
Narrowly defined, an altar is the platform on which a
blood sacrifice was made (for the standard Greek ritual see
Burkert 1985: 55-57, 70-71). However, domestic altars
were used for offerings of fruit and food rather than blood, and it is not clear how far all the votive altars to be found
in a sanctuary were intended for regular blood sacrifice.
Some sanctuaries had more votive altars than can
reasonably have been required by the cult, and some altars
were personal offerings set up outside sanctuaries, and
motivated by specific events (for example, Milner 1998a:
no. 150), so that there might be no intention of regular sacrifice. Most votive altars show no provision for the
sacrificial fire. With appropriate protection they could
127
Anatolian Studies 2005
have been used for sacrifice (Fraser 1977: 21, n. 91;
B?schung 1987:48; Couilloud-Le Dinahet 1991:119-20); but even so, regular use would probably have left signs of
fire damage, and these are virtually never recorded. Some
votive altars were indeed rendered unusable by their size,
such as a monolithic altar to Zeus Bronton from the Nikaia
area with a height of 1.58m (?ahin 1981-1982: no. 1511). The shape of others made them useless; several votive
altars from Anazarbos and Tyana have gently domed tops
(for example, Sayar 2000: no. 57; Berges, Nolle 2000: no.
32, here fig. 1 ), which are not well adapted to carry a fire
slab or dish, or to hold a fire directly. Alternatively they
may have dished tops, which are equally unsuitable for a
fire or for a fire slab, but suggest rather that they received
wine or food offerrings. This is particularly clear when
there is a central dome in the dish, identifying it as a phiale
Fig. 1. Tyana. Round votive bomos with domed top
(Berges, Nolle 2000: no. 32, pi. 107.1; photo: D. Berges)
Fig. 2. Burdur region. Rectangular votive bomos with
mesomphalos top (Burdur Museum inventory no. 5586;
Horsley forthcoming: no. 26; photo: R.P. Harper by
permission ofG.H.R. Horsley)
mesomphalos, the characteristic libation dish (for
example, Horsley forthcoming: no. 26; here fig. 2). Thus
the role of many votive altars was probably much more as
votive offerings than as functional objects; that is, they were symbols of piety, but they were still called ?couoi,
they were meant to be understood as altars, and it is
reasonable to call them altars (Couilloud-Le Dinahet 1991 :
119; tables la and lb below). This acceptance of the altar form as a symbol or
representation of reverence probably combined with the
tendency for the deceased to be in some sense seen as
heroes, and their tombs as heroa, to explain the
widespread use of funerary monuments in the form of
votive altars (against the idea that altars imply formal
128
Coulton
heroization see Fraser 1977: 76-81). Like votive altars,
they commonly have flat tops, with no indications of the
protection needed for a sacrificial fire. But, also like
votive altars, some funerary bomoi have convex (for
example, Sayar 2000: nos 529, 531, 536), dished (for
example, Horsley, Kearsley 1997), or mesomphalos tops
(for example, ?ahin 1994: no. 136), which imply either
no offerings or offerings of wine or food. As with votive
altars, some funerary altars were also too high to be used
for offerings, such as a cylindrical grave monument from
Modrene in Bithynia 1.90m high (Marek 1997: 83). In
several cases dowel holes suggest that they carried upper elements which would prevent use. Pine cones are well
attested (Robert 1955: 247-56, pi. 32-34), and one
funerary altar carried a sundial (Wiemer 1998). Others
may have carried busts or statues of the deceased (Milner 1998a: xv, 19), and a damaged bust from the Olbasa area
with a funerary inscription has been suggested for such a
location (Milner 1998a: no. 130). So far, however, this
usage is attested only for funerary columns (as Petersen, von Luschan 1889: 165-66, nos 193-94, 174, no. 223;
compare Corsten et al. 1998: 70-71, no. 13), not for
round funerary altars; the distinction is discussed below.
These altar-shaped funerary monuments, then, were not
actual recipients of blood sacrifice, but were symbols or
representations of an altar, which by that means
conveyed a proper respect for the dead. Nonetheless they are widely and reasonably called funerary altars by
modern scholars, just as they were widely referred to in
antiquity as ?conoi (see tables 2a, 2b below). However this was not the only context in which the
altar form was used symbolically. A monument to
Claudius beside the newly built Roman road from Myra to
Limyra was given the clearly recognisable form and
decoration of an altar (fig. 3; Marksteiner, W?rrle 2002:
549), and passers-by were presumably meant to read it as
such. Being in all 4m high, however, it could never have
been used for sacrifices, and passers-by would presumably realise also that it was symbolic. The size of the
monument (2.44m by 1.54m on top) would suit an eques trian statue, as suggested for the Stadiasmos monument at
Patara (Salway 2001: 56-57); but Marksteiner and W?rrle
do not report appropriate cuttings on the completely
preserved top course of the Myra 'altar', and the pediments on its ends (only) would be abnormal on a statue base.
Turcan (1991) argues that several apparent 'altars' from
Mithraia had no connection with sacrifice either, although their form had (intentionally) specific features of an altar.
J . ,, , , \ Fassadenprofil Giebelprofil
i 1 1
0 1m i-1
Fig. 3. Kakhk near Myra. Roadside monument in the form of an altar (reproduced from Marksteiner, W?rrle 2002:
568, fig. 7, by permission ofT. Marksteiner and M. W?rrle)
129
Anatolian Studies 2005
Fig. 4a. Pergamon. Roman rectangular altar (after
Habicht 1969: no. 120, pi. 35; drawing: author)
Another monument with the form of a normal votive altar
probably served as the terminal of a water supply
(Schwertheim 1987: no. 47; from Hadrianoi), for the
inscription commemorates the construction of a piscina,
and there is a cylindrical hole running through the middle
of the stone, presumably to hold a pipe ending in a water
spout. It was marked out as an 'altar' by akroteria (see
below), but the connection with water supply meant that
this too was symbolic rather than functional.
The two main shapes of votive altar in Roman Asia
Minor are the round and the rectangular, both commonly monolithic. The basic form consists of a cylindrical or
squarish shaft with projecting crown and base mouldings
(fig. 4). Both these simple types go back to the late sixth
century BC (Yavis 1949: 131-37), and were common in
the Hellenistic period (Yavis 1949: 142-65); both
continued in use until at least the Severan period. At
least from the Hellenistic period, the same two shapes were also used as funerary monuments. Sometimes the
shaft has appropriate relief decoration and/or an
inscription which makes the function plain, but examples with neither are known (B?schung 1987: 47-48). Garlands and bucrania appear on a large group of
Hellenistic and early Roman round altars; in the city of
Rhodes these are almost all funerary, but on Delos and
elsewhere they are both funerary and votive (Hermann
1961: 30; Fraser 1977: 25-33; Berges 1986: 26-28; Couilloud-Le Dinahet 1991: 113). Because this paper discusses such monuments in terms of both shape and
function, it will use the terms 'rectangular bomos' and
'round bomos' to refer to the two shapes, without any
implication of function; 'altar', 'votive altar', 'funerary
altar' and 'statue (or other) base' can then be used for the
main functions, without any implication of shape. Where
the usage of the ancient Greek word is discussed, either
italics or the Greek alphabet will be used (bomos or
?conog). In modern scholarship 'bomos' has been
widely used as the term to describe a rectangular bomos
(for example, Mitchell 1977: no. 20), but tables lb and
2b show bomos was as much the normal term for the
round form as for the rectangular, and the variety of
recent terminology suggests that no ideal alternative is to
hand. In recent scholarly literature in English a round
bomos has been called a columnar stele, a round bomos,
a circular pillar, a funerary column, or a column drum
(Levick, Mitchell 1988: nos 49, 51, 56; Milner 1998: nos
41, 140). Similarly inconsistent terminology appears in
other languages.
One general cause of inconsistency is the use in
descriptions of both terms defined by function and terms
defined by form. The forms of an altar and a statue base
may be identical, but they cannot both be described as
130
Coulton
rectangular altars. The problem is exacerbated if the
function of the stone in question is uncertain. The
commonly used terms 'altar' and 'base' may then be
stretched to include stones for which they are
functionally inappropriate; or there is a search for alter
native phrases which do not pre-empt the question of
function, but mask similarities of form. The approach used here is to separate the terminology of form from the
terminology of function.
A special cause of difficulty has been confusion
between a round bomos and a column. Round bomoi are
unfluted cylinders 1-3 diameters high with crown and
base mouldings (fig. 5a), whereas columns properly
speaking are considerably taller (6-10 diameters), may be
fluted, and would originally have carried a capital quite different from the crown mouldings of a bomos. The
individual drums which frequently constitute a column
can be distinguished from bomoi not only by fluting (if
present) but also by the absence of mouldings at the top and bottom (for example, fig. 5b). Naour (1980: 44-45) showed that this formal difference is reflected in ancient
terminology. The examples in tables lb and 2b show that
Fig. 5a. Territory of Keretapa-Dioskaisareia (?). Round
inscriptions consistently refer to even the more slender
round bomoi of the Roman period as bomoi (altars), whereas columns in the architectural sense, which may also be used as funerary monuments, are called kiones or
styloi (columns, pillars; see Hellmann 1992: 214 on the
use of these words). Although the ancient terminology for different monument types was not perfectly consistent
in Roman Asia Minor, it was certainly not random.
Itacism occasionally causes confusion between stylos and
stele (as Milner 1998a: no. 160), and in the territory of
Hadrianopolis/Kaisareia in northern Galatia grave monuments in the form of a round bomos (often
unusually large) are called s/e/a/(Marek 1993: 101). But
I know of no example where a round bomos (as defined
above) is called kion, or a column (as defined above) is
called bomos.
A general problem in trying to determine the
ancient range of meaning of the term bomos (and other
architectural terms in funerary and other inscriptions) is that the stone carrying an inscription rarely consti
tutes a complete monument. Since the form of the
complete monument is frequently unknown, it cannot
Fig. 5 b. Territory of Balboura. Lower drum of funerary column (Milner 1998: no. 6; photo: A.S. Hall)
131
Anatolian Studies 2005
be assumed a priori that a term refers to the stone on
which it was inscribed. For instance, seven funerary stones from the district of Ikonion (McLean 2002: nos
57, 66-67, 77, 80, 83, 88) have the form of a rectan
gular bomos, but carry inscriptions referring only to a
larnax 'chest', either a smaller bone or ash chest (as
suggested by Robert 1965: 240-43, esp n. 4; Kubiriska
1968: 52) or a full-sized sarcophagus (McLean 2002:
no. 120). In theory one might argue for a lax termi
nology, supposing that larnax referred to the bomos.
But many stones of the same type from the same
district record the erection of t?v ?co|jov kc? tt)v
X?pvaKa (the bomos and the chest) as a two-part monument (McLean 2002: nos 50, 52-55, 58, 61-63,
71-72, 76, 78, 87). It is surely preferable to conclude
that the stones of the first group belong to similar
monuments, and that the bomos has in these instances
been omitted from the inscription. The presence of a
funerary bomos standing beside a sarcophagus is
clearly attested at Tyriaion (Naour 1980: 119) and
elsewhere in the territory of Balboura.
Fig. 6. Territory of Laodikeia on the Lykos. Funerary altar on block base, called bomos and thema (Corsten 1997: no 114; photo: T Corsten)
Similarly, when the word bomos appears on a block
other than a round or rectangular bomos with crown and
base mouldings, it need not imply a different type of altar
or a loose use of words. For example, McLean 2002: no.
69, a plain cylindrical block with a diameter greater than
its height (diameter 0.65m, height 0.42m), has an
inscription referring to a larnax and a bomos. This block
was obviously not the larnax, but there is no reason to
suppose that it was the whole bomos either. It has a
dowel hole and pour-channel in the top, and probably carried a normal round funerary bomos. For both round
and rectangular funerary altars usually stood on a plain base of one or more steps (Bean 1971: nos 24, 32; Fraser
1977: 25; Berges 1986: 12, 31-32; 1996: 32-33), and a
funerary monument from Laodikaia on the Lykos
(Corsten 1997: no. 19; here fig. 6) clarifies the termi
nology. It reads: t? Q?\xa kcc\ t?v ?tt' a?Tcp ?coiiov
(the base and the altar on it), and consists of a simple
square lower element (the thema) carrying a taller upper element with plain bevel crown and base mouldings (the
bomos, with the usual bomos shape).
Many bomoi of the Roman period have an additional
element above the crown mouldings, whose interpretation
requires further analysis, but the addition of akroteria at
the four corners of a rectangular bomos must be
mentioned here. The origin of this type, which is
sometimes called a 'horned altar', seems to lie in the
Levant, where altars/incense burners of basically similar
form are known from the Late Bronze Age to the
Hellenistic period. Although many of them are smaller
than a normal bomos, examples from Beersheba and Tel
Miqne-Ekron measure over a metre in all dimensions
from Tel Miqne-Ekron (seventh century BC) has a plain
projecting band below the simple horns at each corner,
which makes its form comparable to a bomos with crown
moulding and akroteria. These Levantine altars must be
ancestors of the horned altars which appear in Egypt from
the third century BC, and on Delos from the second (fig.
7; Deonna 1934; Yavis 1949: 165-66; Soukiassian 1983); a late third-century altar to Attalos I at Pergamon already has 'horns' in relief. On some altars the 'horns' are large, but in Roman Asia Minor they normally have the form of
simple akroteria, which may be carved as flame
palmettes, or left as plain surfaces of similar shape. The
same forms appear as corner akroteria on sarcophagi and
stelai, and in other minor architectural contexts. Since the
relationship was presumably recognised in antiquity, they are here called akroteria rather than horns. However, their
origin is quite distinct from the akroteria which, from the
fifth century at least, decorated funerary and other stelai, for these are associated with a pedimental crown, and
include a central akroterion as well as the corner ones
132
Coulton
Fig. 7. D?los, Sarapeion B. Hellenistic altars with
168; 1995: figs 120,123,127,147,150). There can be no
doubt that this feature originated in a cultic or votive
context, but it was already transferred to funerary contexts in the Hellenistic period. The earliest examples are square ash chest lids with akroteria, probably of late
Hellenistic date, which come from Pergamon and
Mytilene; it is usually presumed that they belonged to
altar-like monuments (Pfuhl, M?bius 1977-1979: nos
2229-31, 2234-35). Akroteria continue to appear on
votive altars in the Roman period, and become a common
feature of rectangular funerary bomoi (tables 3a, 3b). By the late first century BC they also appear on statue bases, the earliest instance being for the statue of a local seer at
Hadrianoi in Bithynia; they are common in this context
by the second century AD at latest (table 3c). The two basic forms of bomos used for votive and
funerary altars are also virtually indistinguishable from
the commonest statue base types of the Roman period. The similarity begins, in fact, as early as ca. 525 BC, for
the altar represented on the south frieze of the Siphnian
Treasury at Delphi shares its form with the pedestals of
the two Caryatids of the west fa?ade (Daux, Hansen
1987: 147-48). Statue bases consisting of a rectangular or cylindrical shaft with projecting crown and base
(the bomos and the statue on it; ?ahin 1981-1982: no.
1503; here fig. 9) and a well recorded form. It is a
standard rectangular bomos-shaped statue base, but the
akroteria above its crown mouldings identify it as repre
senting an altar. It carried a statue of Zeus, and the
dedication could be regarded as a votive altar adorned
with a statue or as a statue on a base. The wording of an
inscription from Phrygia is equally clear (Drew Bear
1993: 147-52): t?v ??cc kg t?v ?tt' cc?tco ?conov ke
t?v ?v Tcp ?coiicp 'Atto??ovcc (the [statue of] Zeus
and the bomos below it and the [statue of] Apollo on the
bomos). In the case of L. Ant. Claudius Dometianos
Diogenes at Aphrodisias the wording and the archaeo
logical context work together. The inscription on the
statue base records that Ti. Claudius Ktesias oversaw the
erection of the statue, and 'had its bomos, and the other
things made through his own efforts' (Trorr|Gcc{aEVou 5?
\
0 0.25m
Fig. 8b. Lindos. Hellenistic round statue base (repro duced from Schmidt 1995: no. II. 5, fig. 208, by
permission of I. Schmidt)
kcc\ Tov ?coiaov cc?Tcp Kcc\ Ta AoiTT? Trapee ?auTcp;
Inan, Alf?ldi-Rosenbaum 1979: 210-13, no. 186, pl.
140.3). The statue was set up in the portico in front of the
bouleuterion; this portico has been excavated, and there
is no sign of a separate altar, so that bomos must again
refer to the base itself. The base is of the normal bomos
form, but there are very small akroteria (visible in Smith
1998: 67, fig. 1), which suggest that it could be read as
an altar.
In other cases listed in table 4a the association of
statue and bomos is obviously close, and the interpre tation of bomos as referring to the statue base is plausible but not certain. Unfortunately statues and altars (votive or funerary) were both appropriate in much the same
places, and their juxtaposition in an inscription may indicate two separate elements rather than a single whole.
Wording alone can not tell us whether a phrase such as
tov ?coiaov o?v Tcp ??ovTi (the bomos with the lion;
Robert 1938: 220, n. 10), or t?v ?couov Ka\ t?v
avSpiavTa (the bomos and the statue; CIG: 3776) referred to a bomos base beneath the lion or statue, or a
bomos altar beside the lion or statue. For, as we shall see
below, similar phrases linking bomos and sarcophagus can mean either a sarcophagus on a bomos or a
sarcophagus beside a bomos. Where the inscribed block
134
Coulton
is of bomos form and has cuttings on top for fixing a
statue (as Corsten 1997: no. 51b), it can be taken as effec
tively certain that the bomos was the stone that served as
base. But that physical evidence is not recorded for, for
example, Engelmann et al. 1980b: no. 1266, so doubt
must remain.
Where the form of the bomos is known, it is generally a normal rectangular bomos (with or without akroteria),
which might have served equally as statue base, votive
altar, or funerary altar. But in other cases the text clearly refers to a bomos and statue elsewhere, and there are two
more problematic cases. A stone from Ephesos carries a
relief of a goddess in a pedimented panel, and an
inscription referring to t?v ?coiaov o?v tco ?tt' a?Tcp
suppose that the stone is the bomos and the relief figure is
the xoanon. But the stone is a slab rather than a block,
and a break to the left shows that a significant part of it is
missing. It is not obviously suitable as a statue base, and
the relief figure (for which the term xoanon would be
unusual) is not 'on it' in the way that ?tt' a?Tcp would
normally signify. It is easier to suppose that a statue stood
on a bomos base nearby. Another stone, from the territory of Antioch, carries a relief of two women and the words
Ta ?ya?naTa Kai t?v ?co|iov ?TroirjGa (Jarry 1982:
95-96, nos 41-42; I made the statues and the bomos).
Again the editor supposes that the inscribed stone is the
bomos and the relief figures are the agalmata, but again the block does not serve as a base for the relief figures, neither is it obviously part of an altar. It seems likely that
this is another case where the words of the inscription do
not refer to the stone on which they are written.
The inscriptions on two statue bases from Laodikeia
on the Lykos (Corsten 1997: nos 51b, 65) indicate that
bomos was not used as a general synonym for basis
(base), for they refer to both a basis and a bomos. In the
first case the block carrying the inscription, though not
fully preserved, was a rectangular bomos of the usual
kind; as Corsten observes, if this was the bomos, the
basis would have been another (lost) block which
carried it. The second inscription is on a plain block,
1.20m wide and 0.49m deep. Corsten again suggests that this was the bomos, but it seems more likely that it
was the front block of a pair forming a square or rectan
gular base (the basis) for a taller bomos, similar to no.
51b, which would have been the bomos. Both
monuments would then have matched the funerary altar
from the same city, discussed above (Corsten 1997: no.
114), for which the words thema (rather than basis) and
bomos were used. The use of bomos to refer to a statue
base must have been determined by the form of the
monument rather than its function.
Fig. 9. Territory of Nikaia. Statue base with akroteria,
called bomos (Bursa Museum inventory no. 2550; repro
duced from ?ahin 1981-1982: no. 1503, by permission of S. ?ahin)
135
Anatolian Studies 2005
In the case of dedications, the usual assumption is that
the base is a negligible element, present only to focus
attention on the object carried ? whether statue or other
offering. However, this may be a modern imposition. An
inscribed bomos from Apollonia in Phrygia records the
offering of small figures of oxen to Zeus, but the
inscription also draws attention to the bomos, itself a
significant part of the dedication (MAMA 4: no. 140; Robert 1980: 244-45). Similarly a substantial bomos
dedicated to Zeus from the Nikaia area carries a bust of the
god on top (?ahin 1981-1982: no. 1089). ?ahin rightly calls this an altar, although clearly not for any sacrificial
use; to regard the bomos element as merely a base for the
bust ignores its greater size and equal complexity. Hellmann (1992: 74) has suggested that the
diminutive form, bomiskos, was used to refer to a statue
base in the Hellenistic period, which would imply that
bomos was probably similarly used at this time.
However, in most of the instances listed in her note 4 the
bomiskos is an isolated item, probably a votive in the
form of a miniature altar. Two mid-second century Delian inventories list a bronze statuette of Hestia seated
on a stone bomiskos (ID 1416 A 1.83-84; 1417 B 1.90), but since both Hestia and the bomiskos are distinguished from the stone basis which carried them both, the
bomiskos was part of the image, not the base of the
statuette. Hestia's seat may have been a plain rectangular
block (for this meaning of bomiskos see below), but it
might very appropriately have been a small altar (a sense
which bomiskos could equally carry; Hero, Spir. 1.38,
39), as on a round bomos altar from Ostia (Museo Ostiense 120; LIMC s.v. Hestia: no. 16, pi. 293). If this
explanation is correct, no Hellenistic examples of bomos
or bomiskos referring to a statue base have yet have been
found. On the other hand bathron and bema, normal
words for a statue base in the Hellenistic period, seem to
have dropped out of use later (tables 4d, 4e; LSJ and
Orlandos, Travlos 1986: s.w.; Hellmann 1992: 63, 69). Basis was probably the normal word for a statue base
(table 4c), and was already established by the Hellenistic
period, although not restricted to that meaning (LSJ and
1997: no. 4, pi. 2.2; photo: courtesy of Istanbul Archae
ological Museums)
An inscription from Nikaia with the words t?v
?couov Kai t?iv ?GTO0r|KTiv (the bomos and the
ostotheke) (Kubi?ska 1997: no. 8 = ?ahin 1981-1982: no
1581) was carved on a stone beam with both ends treated
as consoles (fig. 11). Kubi?ska suggests that the
ostotheke would have been carved in the top of a tall
pillar with this beam as its lid, but this does not explain the form of the beam. ?ahin is surely right to suppose that the bomos was a bomos-shaped element carrying this
beam, which in turn carried the ostotheke, as a separate chest. The lower surface of the beam is ca. 0.64m by
0.65m, suitable for a bomos slightly larger than usual,
while its upper surface, 1.13m by 0.65m, could hold an
ostotheke in the form of a small sarcophagus. For, as we
shall see, the Nikaia area has revealed monuments of this
form which carry sarcophagi. A similar monument from
the borderlands of Isauria and Lykaonia consisted of a
rectangular bomos carrying a small sarcophagus (i.e. an
ostotheke?) with a lion on the lid, but no cantilever beam
was identified (Sterrett 1888: 91-93, nos 153-54; Robert
1937: 394-95; for the lid compare Keil et al. 1935: 20,
fig. 17); a cantilever beam from Sparta may have served
the same purpose (Altmann 1905: 31; top 1.77m by ca.
0.60m, under face 0.90m by ca. 0.60m). Another rectan
guiar bomos from the Nikaia area (Kubi?ska 1997: no. 6) has its upper part concealed in the earth, so when the
inscription speaks of 'this bomos with the ostotheke', it
remains open whether the ostotheke was carried on a
beam (as Kubi?ska 1997: no. 8), or was directly on the
bomos (as Kubi?ska 1997: no. 4), or was placed beside
the bomos.
At Ikonion the interpretation of the comparable
phrases 'the bomos and the chest (larnax)' or 'the bomos
with the chest' (see above, and McLean 2002: nos 50,
53-55, etc. [ko\], 61, 76, 78 [oi/v]) depends on the
nature of the larnax. Robert (1965: 240, n. 4) suggested that it might be a small round ash urn that could be
placed on top of the bomos concerned, and McLean
2002: nos 50 and 55 both have a cylindrical drum above
the bomos proper, which might support a round
ostotheke like that from Nikomedeia discussed above
(Kubi?ska 1997: no. 4). However, McLean reports no
dowel hole or clamp cutting to hold such an urn in place, so Robert's suggestion remains hypothetical. As noted
above, at least one sarcophagus was called a larnax
(Mclean 2002: no. 120; sows was the commoner term:
nos 180-83), and a larnax of this size must have been
placed beside its bomos, not on it, since the bomos
would certainly be too small to carry it. Although they do not include the word larnax in their inscriptions, chests in the form of small sarcophagi are commonly called larnakes today (Mclean 2002: nos 185-90; lids
for comparable chests: nos 191-94), and these might have been carried on the larger bomoi. But no dowel
holes are reported in the top of the bomoi or bottom of
the chests, and they may equally have been placed beside their bomoi. It might be simpler to explain Kubi?ska 1997: no. 10, from Amastris, in a similar way. The phrase 'the bomos and the ostotheke' appears on an
ostotheke which was re-used separately. Kubi?ska
supposes that it was cut from its bomos base for re-use,
but the re-use would have been easier if the ostotheke
had been a separate block from the start, perhaps origi
nally standing beside the bomos, not on it. Kubi?ska
1997: no. 9 is also on an ostotheke, but the existence of
a related bomos is hypothetical.
Fig. 11. Territory of Nikaia. Cantilever beam to carry ostotheke (Bursa Museum inventory no. 2742; repro
ducedfrom ?ahin 1981-1982: no. 1581, by permission of S. ?ahin)
137
Anatolian Studies 2005
For the present purpose the important thing is that in
all the cases cited by Kubi?ska where a bomos is
mentioned, the 'base' element of the monument has the
same bomos form as a funerary or votive altar. Some
have akroteria (Kubi?ska 1997: no. 5, from Nikomedeia =
?ahin 1981-1982: no. 1318); others do not (Kubi?ska 1997: nos 2-4). The simple upper element containing the
ostotheke of Kubi?ska 1997: nos 1-3 is found on a
number of funerary altars from Bithynia and elsewhere,
but without being hollowed out to hold ashes (?ahin 1981-1982: nos 1242, 1319, 1354, 1450, 1460, 1469,
from Nikaia; MAMA 1: nos 72, 154, from Laodikeia). Further research is required to establish whether it origi nates in altar morphology or is a potential ostotheke
which was not always used as such. Kubi?ska 1997: nos
12-13, 15-18, all from the Ankara area, also consist of
apparently bomos-shaped blocks, though most are
unillustrated and incompletely described. Only no. 18
specifies the relationship of bomos and ostotheke: t?v
?conov K? t?\v ?tt' a?Tcp ?GTO0r|KTiv (the altar and the
ostotheke on it); it is not clear from the description whether the ostotheke was cut from the upper element of
the same block, or was formed by a separate block. The
less specific formulae 'the bomos and the ostotheke' or
'the bomos with the ostotheke' would cover an ostotheke
set beside the bomos, rather than on it, as suggested above. In Kubi?ska 1997: no. 14 the noun associated
with the ostotheke is not bomos but an unintelligible word. The stone carrying the inscription is nevertheless
in the form of a simple bomos (with bevelled crown and
base mouldings) below a tapering upper element whose
narrow top perhaps carried a separately made ostotheke
in the form of an urn.
A different connection between altar and ostotheke is
illustrated by three round bomoi from western Asia
Minor. The first (Kubi?ska 1997: no. 20), once in a
private collection at Izmir and probably from Smyrna, is
best known from a 19th century description by Baumeister. It was ca. 1.20m high and ca. 0.40m in
diameter, with garlands round the drum, which was
hollowed out to form an ostotheke. The second, from
Kyme (Yavis 1949: 149; Berges 1986: no. 98), is similar
in form, but the cavity for the cremated ashes is
concealed in the cone of leaves which comes above the
crown mouldings (fig. 12). The third altar, from Aphro
disias, has a rectangular cutting in the upper part of the
bomos, probably intended to hold the ashes (Berges 1986: 21, K102).
One of the inscriptions on the monument from
Smyrna refers to it as an ostotheke, and Kubi?ska
maintains that it is wrong to call it an altar, as some
scholars have. In terms of function she is right, but all
three monuments are representations of altars; their form
and decoration is exactly like those which are normally called bomoi in inscriptions, and funerary altars by
epigraphists. The pine cone which crowned the lid of the
piece from Smyrna can be matched on rectangular
funerary altars from Nikomedeia (D?rner 1941: no. 89,
pi. 35), Akmonia (Robert 1955: 247-56, pi. 32-34) and
elsewhere. The leaf-covered cone on the monument from
Kyme can be compared with omphaloid cones on
Rhodian funerary altars (Fraser 1977: 40, n. 229, pis
110a-d), or the imbricated omphalos on a square altar
from Phrygian Apollonia (MAMA 4: no. 181, pi. 43). It
seems likely that all three monuments were envisaged by
contemporaries as funerary altars containing an
ostotheke, and this conception is expressed by the
Fig. 12. Kyme. Round funerary bomos with leaf-covered
omphalos serving as ash container (Istanbul Archaeo
inscription on a monument from Ankyra which reads t?v
?coiaov o?v Tr? ?v a?Tcp ?GTo6r|Kfl (the altar with the
ostotheke in it; Kubi?ska 1997: no. 11). There is no
detailed description of this stone, but its early editor
called it a cippus, which often means a round funerary altar. Rectangular funerary altars at Rome also
frequently have an ash container, usually cut in the top, but sometimes in one of the sides; yet other features show
that they were conceived as, and meant to be read as, altars (B?schung 1987: 38, 47-48).
The style of the monument from Kyme is late
Augustan, and that from Aphrodisias has been dated to
the early first century AD, so this association of altar and
ostotheke precedes, and may explain, most of those
discussed above. Placing an ostotheke on a funerary altar
is not a large step away from cutting an ostotheke in the
top of a funerary altar.
Sarcophagi on bomoi
Bomos-shaped bases could also carry sarcophagi. Three
monuments from the Nikaia area consist of a
sarcophagus carried on a stone beam cantilevered above
a bomos-shaped pillar (fig. 13; ?ahin 1981-1982: nos
1231-33). The inscriptions on these monuments use
only the general terms of\[ia (marker), |avf)|aa or
livrmE?ov (monument), but the pillars of the two
monuments which have been illustrated have all the
trappings of a monolithic altar with akroteria, although on a gigantic scale. The height of the bomos element of
?ahin 1981-1982: no. 1232 was 5m and its width 2m.
A different tomb type, found at Nikomedeia, consisted of a large rectangular platform (about 1-1.5m
by 3m in plan) with a bold projecting moulding around
the top and bottom, often standing on a stepped base (fig. 14; D?rner 1941: 22, nos 63, 78, pi. 4). The inscription
on one of these refers to t?v Tro?aXov a?v tco
UTTOKEiiiEvcp ?concp (the sarcophagus and the bomos
beneath it), so this large platform was also known as a
bomos (D?rner 1941: no. 78). Tombs of the same type are numerous at Hierapolis, where a tall platform carried
the sarcophagi of the leading members of a family, and a
chamber within the platform served as a burial place for
the lesser members of the family (Humann 1898: 16-17; Kubi?ska 1969: 78-79, pis 7-11). Again, inscriptions make it plain that this large platform was known as a
bomos (Robert 1950: 202-03). In these monuments (see table 6a) the structure called
bomos did not resemble the common rectangular bomos
or monolithic altar, but altars were built in a variety of
shapes and sizes. Whereas monolithic altars were
commonly personal dedications arising from some
specific prayer or vow, the main public sacrifices in a
sanctuary would need a larger altar, which Yavis (1949:
Fig. 13. Territory of Nikaia. Sarcophagus carried on a
giant bomos with akroteria (?ahin 1981-1982: no. 1232;
reproduced from Athenische Mitteilungen 17, 1892, pi.
5, by permission of the Deutsches Arch?ologisches
Institut)
Fig. 14. Territory of Nikomedeia. Sarcophagus carried
on rectangular platform inform of large altar (named as
bomos); note that lower half is concealed by vegetation
(reproduced from D?rner 1941: pi. 4.3, by permission of the Deutsches Arch?ologisches Institut)
139
Anatolian Studies 2005
95-105, 177-83) classified as either a 'ceremonial altar'
or a 'monumental altar', depending on size. Both types
normally consisted of a platform from 2m to 20m long, 1
2m wide and l-2m high, with a projecting moulding or
cornice at the top and a base moulding at the bottom.
Some were rectangular, with a step or steps at one side for
the officiating priest (the prothysis); in others the prothysis was built between spur walls or antae, projecting from the
main altar mass (Yavis 1949: 183-87; Etienne 1991).
Fairly well preserved examples of the rectangular type have been found in the sanctuaries of Asklepios at
Messene (12.62m by 2.03m; Orlandos 1970: 133-35, figs
8-9; here fig. 15) and of Apollo at Cyrene (22.08m by
4.95m; Pernier 1935: 61-70), and a clear representation of
one appears on a Hadrianic relief from the Arco di Porto
gallo at Rome (Nash 1961: pi. 88). The structure set up by the Milyadeis and others in honour of Roma and Augustus is a possible example in Roman Asia Minor (Hall 1986:
139-40, no. 1, n. 2). Hall reasonably suggests that it was
an altar, at least 3.5m long, and the associated masonry, re
used in an Ottoman bridge abutment, includes a suitably
large cornice moulding (Hall 1986: pi. 12). These larger altars would always have been less numerous than the
votive bomoi, and, not being monolithic, they have rarely
preserved their full form to the present day. But in the
Roman period their higher status would compensate for
their fewer numbers, and they would naturally come to
mind in connection with the word bomos. The tomb
platforms at Nikomedeia and Hierapolis with their cornice
and base mouldings, often set on steps, resemble such
large rectangular altars, as seen from the back, the side
which the public would face. That means that the term
bomos here too means (at the least) 'altar-shaped
platform' not just 'platform in general' or 'funerary
platform'. And here too the symbolic value of the form
and of the term may have been intended and recognised.
Fig. 15. Messene. Reconstructed drawings of the altar of
Asklepios from northwest (above) and southeast (below)
(reproduced from Orlandos 1970: fig. 9, by permission of the Athens Archaeological Society)
In some other cases the monuments are lost, but the
inscriptions alone suggest monuments of similar form, with the bomos being a large built platform beneath a
sarcophagus. At Aphrodisias, for instance, some
inscriptions show that sarcophagi were carried on bomoi
containing several burial recesses (MAMA 8: nos 545
46, 554, 556, 570). An inscription from Hypaipa (near
Ephesos; Meri? et al. 1981: no. 3866) speaks unambigu
ously of 'this bomos and the chamber within it'
(toutov t?v ?couov Ka\ TOV EV a?Tcp oTkov). In
principle, phrases such as 'the bomos and the
sarcophagus upon it' (B?rker, Merkelbach 1980: no.
1637; Merkelbach, Nolle 1980: 2222c, 2228, 2241,
2266a, 2306a; Meri? et al. 1981: no. 3138) or 'the
sarcophagus and the bomos below it' (Merkelbach, Nolle 1980: no. 2304) could refer to gigantic monolithic
bomoi like those supporting sarcophagi in the Nikaia
district. But since the inscriptions are carved on smaller
blocks, they probably refer to built platforms like the
bomoi of Nikomedeia and Hierapolis. At Termessos a platform measuring 6.22m by 7.40m
and at least 1.4m high (TAM 3.1: 623) carries an
inscription claiming t?v ?coiaov Kai tt\v KprjTTE?Sa
(the altar and the platform) for the sarcophagi of the
owner and her children. This platform is much larger than those at Nikomedeia and Hierapolis, so it was
probably the krepis; the bomos would have been a more
restricted structure upon it, but still big enough to carry more than one sarcophagus. Another Termessian text
(TAM 3.1: 814) curses anyone who moves the owner's
sarcophagus from the bomos; the size and form of the
bomos are unknown, but a monument similar to those at
Nikomedeia and Hierapolis ? and altar-like for the same
reasons ? is again plausible.
Three earlier monuments support the idea that a tomb
and a large altar could be seen as related. The earliest
survives only in description. Pausanias (3.19.3) says that
the base (basis) of the statue of Apollo Amyklaios was in
the form of an altar (bomos), in which the hero
Hyakinthos was said to be buried; a door in the side gave access to the tomb chamber. Clearly this is meaningless if bomos meant just 'base or platform in general'; as at
Laodikeia on the Lykos (above) basis and bomos were
distinct terms, and bomos conveyed a form. The tomb
chamber was in the bomos platform, as at Nikomedeia
and Hierapolis. Tomb Kl at Messene, which probably dates to the second century BC, has the form of an in
antis altar of the type mentioned above, except that
slightly stepped base courses take the place of the
prothysis between the antae (Themeles 2000: 114-19,
figs 99-107; Ito 2002: 4-15; here fig. 16). A door in the
rear wall leads into a chamber with seven burial cists
below the floor. The form of the top is uncertain. It has
140
Coulton
Fig. 16. Messene. Restored plan of Hellenistic tomb Kl
(reproduced from Ito 2002: fig. 13, by permission of J. Ito)
been restored as a stepped pyramid, but flat slabs are
equally possible, resulting in an altar-like monument
comparable to the platform bomoi at Hierapolis and
Nikomedeia. The sarcophagus of L. Cornelius Scipio Barbatus (consul 298 BC) at Rome embodies the
connection between altar and tomb more clearly but in a
different way. It was given the unmistakeable form of a
monumental triglyph altar (Yavis 1949: 138-39),
complete with volute end barriers (Yavis 1949: 181-82,
fig 46; Coarelli 1972: 43-49, figs 4-9). A series of later
tomb monuments from Rome, Pompeii, Ostia and
elsewhere are variations on this theme. They take the
form of a normal monolithic altar (with projecting crown
and base mouldings), but many times magnified (fig. 17; Kockel 1983: 22-26; for the distribution, 25-26); the
'altar' component is roughly square with sides 1.5-3.5m
and a height of 2.5-3.5m, so comparable in scale to the
platform bomoi at Nikomedeia and Hierapolis. Those at
Pompeii are clearly identified as representations of altars
by their cylindrical end barriers (pulvini); although the
Ostian examples lack pulvini, they presumably carried
the same significance. Many of these altar monuments
were solid, but one at Pompeii (North 1; Kockel 1983:
111-15) contains a burial chamber like the platform bomoi at Nikomedeia and Hierapolis. In all these cases,
however, the burials were within or beside the altar
monument. There is no evidence that any of them carried
a sarcophagus, as the platform bomoi in Asia Minor did.
The change is the same as that from the early ostothekai
where the round bomoi contained the ashes (Kyme and
Smyrna) to the later ones where the bomos carried the
ash container (Nikomedeia and elsewhere). Four inscriptions from the area of Kibyra use the
adjective ?coiaiKOc (altar-like; Corsten 2002: nos 147,
funerary platform or statue base), ?coniKOc 'altar
shaped' would mean 'like any altar-shaped thing'; there
is no need to derive the adjectival form specifically from
?coiioc =
funerary platform (as Drew Bear 1972b: 190). Two inscriptions from Perge use ?com's, a diminutive
form of ?cojios, in conjunction with a sarcophagus
(Mansel, Akarca 1949: nos 4, 13; table 6b). Kubi?ska
(1969: 79) again supposes a built platform carrying the
sarcophagus, but Herodotos (2.125) used ?coiiioEc for
the steps making up the form of a pyramid before the
outer facing was added, so at Perge ?conic might equally have described a stepped base rather than an altar-shaped
platform of the Hierapolis type. However, the excava
tions at Perge revealed neither a platform nor a step beneath any of the sarcophagi, and since the inscriptions do not say that the sarcophagus was on the ?coiiic, a
normal monolithic funerary bomos beside it may be
meant, ?conic in the sense of bomos-shaped statue base
is attested at Pogla (Pisidia; Bean 1960: 61-62, no. 105).
Bomoi as column pedestals The final context where bomos appears as a bearer is
architectural (table 7). The usage is clearest at Aphro disias, where the architectural context survives. In listing
improvements to the gymnasium of Diogenes, CIG 3:
2782 refers to columns with ?coiiooTTEipa (bomospeira,
'altar-bases') and capitals. Since speira is a well-estab
lished term for a column base of Ionic type (Ginouv?s 1992: 70), Franz, editor of CIG 3, followed Otfried
M?ller in interpreting this compound as 'Ionic column
bases (spirae) placed on tall square pedestals in the form
of an altar'. De Chaisemartin (1989: 44-45) has argued that the gymnasium of Diogenes is what is now called the
Portico of Tiberius, and that the inscription relates to its
west end, where the columns do indeed stand on
pedestals (Erim 1986: 98; fig. 18). Their form, a square die with projecting crown and base mouldings, is
virtually the same as the basic type of rectangular monolithic altar.
The same compound, or its parts, occurs in two
inscriptions from the territory of Hypaipa in the Kaystros
valley (Meri? et al. 1981: nos 3851-52), which refer to
two-columned monuments with 'altar-bases' and capitals.
One uses the compound bomospeira, the other breaks it
down into bomoi, and speirai. The same sense (bomos =
column pedestal) would be appropriate, although the
architectural elements referred to have not been found.
The simple 'bomos' occurs in an equally clear context in
an inscription from Ephesos (Wankel 1979: no. 20.70-1; Rumscheid 1999: 54, n. 121): ke?ove? ?' ... o?v toi?
?TTOKEin?voi? ?coiiois (two columns with the bomoi
below them ? obviously pedestals). The same meaning
of bomos is probable, though unproven, in an inscription from Tyana (Berges, Nolle 2000: no. 39): t? TrpOTru?ov g?v Ta?? TrapaoT?oiv Ka\ to?? ?conoic (the propylon with the pillars and the bomoi). The context makes
pedestals a more likely meaning than votive altars. In the
fourth century AD a letter of Gregory of Nyssa discussing the design of a small church (Epist. 25, Migne, PG 46:
1100A) specifies that the eight main columns should have
bomoeideis speiras (altar-like bases) and Corinthian
capitals, and in an elaborate poetic description of an
ambon, Paul the Silentiary (Ambon 150-60; ca. AD 575) refers to bomoi beneath the columns. Since the archi
tecture relevant to these cases has not been found, the
proof is not conclusive, but all the contexts support the
sense of 'column pedestal' for the bomo- element.
Fig. 18. Aphrodisias, west stoa of south agora (?=
gymnasium of Diogenes?). West end with columns on
pedestals (photo: R.R.R. Smith)
142
Coulton
However, another inscription from the Hypaipa district cannot be easily reconciled with this expla nation. Meri? et al. 1981: no. 3828 records that 'the
sarcophagus with the bomospeiron' belongs to someone.
Although the bomo- element of the word bomo could
refer to a built platform beneath the sarcophagus as at
Nikomedeia or a monolithic pillar as at Nikaia there is
no obvious place for the -speiron element. It is possible that bomospeiron had come to be used of the column
pedestal alone, ignoring the much smaller column base
element; if so, the Hypaipa monument more probably followed the Nikaia form. Alternatively, bomospeiron
might have no direct connection with the sarcophagus, but referred to (for example) the base for a statue of the
deceased with the same profile as an Ionic column base
on a pedestal, as found frequently at Termessos (TAM 3.1: pi. II, types e-f).
Some later scholars have tended to blur the sharpness of the CIG interpretation of bomospeiron. Ebert (1911:
26) translates it as Untersatzring (base-ring), under
standing the first element as a lower element in general. Orlandos and Travlos (1986: 55-56, s.v. ?conooTTEipov) do not refer specifically to pedestals, and Ginouv?s
(1992: 70, n. 106), although translating the word as 'base
in the form of an altar', comments that it returns to the
sense of 'moulded column base', rather than linking it
with a pedestal. Nolle, discussing Berges, Nolle 2000:
no. 39 and Nolle 2001: no. 162 (where he restores bomos
in a phrase including column and base-and-capital),
suggests that bomos refers to the low square plinth which
forms the lower element of an Asiatic Ionic column base.
This does not fit with the usage of bomos which we have
seen in connection with ostothekai, sarcophagi and statue
bases, and the more usual sense of a taller element with
crown and base mouldings (i.e. a column pedestal) would
be equally appropriate here. On the other hand de
Chaisemartin (1989: 45), in reconsidering CIG 3: 2782,
rightly returns to the perception of M?ller and Franz that
bomospeiron must mean 'base with pedestal', and
Rumscheid (1999: 54, n. 21) saw that the bomoi in
Wankel 1979: no. 20 must also refer to column pedestals.
Performers on bomoi?
Weiss (1981) has suggested a fifth context where bomos
was used as a 'bearer' unrelated to an actual altar. He
supposes that it was used to describe the pedestal on
which trumpeters and heralds stood to compete in
festival games. The monument from which he started
this argument is a stepped block from Side in Pamphylia, decorated with reliefs. The inscription on it says that the
donors dedicated the altar (bomos) which they had
provided and gilded, together with the base (basis). Weiss supposed that this stepped block was the bomos,
with the figures of the relief once gilded. However, W?rrle (1988: 191) argues more plausibly that the gilded bomos was a portable metal altar, like the silvered altar
used in the Demostheneia festival at Oinoanda. The
stepped block would then be the basis, comparable to the
simpler steps that we have noted under other altars and
related monuments (see above). The parallels which
Weiss cites for bomos used in this sense also need further
examination. The strongest case is where Pausanias
(5.22.1) describes a bomosby the entrance to the stadium
at Olympia, 'where the Eleians sacrifice to no god, but it
is set up for the trumpeters to stand on, and the heralds, when they compete.' This monument was obviously used as Weiss supposed, but Pausanias felt bound to
explain that the Eleans did not sacrifice on it, which
shows that he was not using bomos as the normal word
for a herald's platform. Rather he was using it to
describe a monument which looked so much like an altar
that a visitor would naturally have supposed that it was
used as one.
The other two instances are quite different. The
Certamen Homeri et Hesiodi (315ff) tells how Homer
sailed to Delos and, standing on the KEpcmvo? ?coiaoc, recited the hymn to Apollo. However the keratinos
bomos (altar of horns) was a functioning altar, not
intended to be used as a reciting platform (Bruneau 1970:
19-29). Similarly, when Philostratos (Vitae Soph. 495) tells us that Gorgias 'proclaimed the Pythian speech from
the bomos ... in the Pythian sanctuary', there is no
reason to believe it was anything other than the cult altar
in front of the temple of Apollo, which had a stepped
prothysis from the top of which a speech could effec
tively be delivered to a festive crowd below.
Discussion
There are then four contexts in which something serving as a pedestal or base is referred to as a bomos: it may
carry a statue, an ostotheke, a sarcophagus,
or a column.
At the start of this paper it was noted that Robert and
other scholars have supposed that bomos in these
contexts meant a base, pedestal, socle or platform in
general, as it did for Homer. However, this sense is not
apparently attested between Homer and the first century AD, and original meanings may sometimes be forgotten
(as that of English 'nice', derived from Latin nescius =
ignorant). Examination of the form of the pedestals described as bomoi shows that most had the bomos
shape, that of the standard rectangular monolithic altar, while the remainder had the shape of a larger altar. Thus
bomos was used in these contexts not for a 'base or
platform in general', but for a 'base or platform in the
shape of an altar'. The instances where bomos and basis are used as mutually exclusive categories show that the
143
Anatolian Studies 2005
words were not seen as synonyms, and this is exemplified most strikingly by Pausanias' description of the basis of
the statue of Apollo at Amyklai as having the shape of a
bomos. Pausanias must have used the word bomos
because it conveyed a specific shape. The architectural context is the one where the use of
bomos is most likely to have been limited, in the Roman
period at least, to a definition of shape. Picard (1927:
255-60) suggested that a symbolic connection of column
pedestal and altar underlay the much earlier use of the
sculptured pedestals below certain columns in the
temples of Artemis at Ephesos and Sardis; the altar
shaped pedestals would have marked certain columns out
as having a higher status than the others. However,
Wannagat (1995) argues that the Hellenistic use of
column pedestals derives from the custom of placing
funerary columns on bases or pedestals, and that these
were conceptually and formally equivalent to the bases or
pedestals placed beneath statues and stelai. Like statue
bases they raised the object set on them physically and in
status, but the similarity to altars is coincidental, or at
least indirect. In the Roman period column pedestals never carry the akroteria or other features that might mark them specifically as representing altars. The
earliest instance of bomos used for a column pedestal is
in relation to a fish market at Ephesos (AD 54 and 59; see
above). This context would make a conscious connection
with altars inappropriate. In the architecture of Roman
Asia Minor, column pedestals were used as much in
ordinary civic porticoes as in temples. Gregory of Nyssa felt that 'altar-shaped bases', like 'carved capitals of the
Corinthian type', would raise the status of the architec
tural design, and that may have been a general under
standing, as Wannagat has suggested. It is unlikely that
either Gregory or Paulus Silentarius understood their
column pedestals to have a specific reference to pagan
altars, for such a reference would have been highly
inappropriate. In these cases bomos simply conveyed a
specific form.
This purely formal sense is attested much earlier in
bomiskos, the diminutive of bomos. In solid geometry bomiskos meant a cuboid, which differs from a cube in
having length, width and height unequal (Hero, Def 114;
in Hero, Stereom. 68 it also tapers). However this
geometrical use of bomiskos is probably based on the
specific shape of a common object, i.e. the rectangular monolithic altar, rather than on a loose meaning of bomos
as a 'base or platform in general'. Other terms for
geometrical solids derive similarly from everyday objects which have the relevant shape
? so kubos (cube) from a
dice, kulindros (cylinder) from a roller, sphaira (sphere) from a ball, dokos (a cuboid shape with length much
greater than height or width) from a beam, and so on.
Where Hero used bomiskos for the base element serving as the air and water container of a water organ (Spir.
1.42), he may have envisaged a simple cuboid as above; but given the decorative character of his models, he
might well have envisaged it with crown and base
mouldings, which would make it look more like a small
altar. Elsewhere in Hero's works bomiskos certainly means a miniature altar, functioning as part of a
mechanical model, but intended to represent an altar
(Spir. 1.38, 39; compare bomos in Hero, Spir. 1.12, 2.3,
2.21; Autom. 3-4, 12). The meaning of ara in Latin was extended in just the
same way as that of bomos. It usually means an altar, but
Varro uses ara for a statue base (Scholion on Vergil, Eel.
6.31: arae in quibus stant signa); an inscription from
Udelfangen uses it to refer to the pedestal of a column
monument (Esp?randieu 1907-: 6, no. 5230; CIL 13:
4117: . . . ]cum columfna e]t ara posuit); as Hero used
bomiskos, so Vitruvius (10.8.1,2) used ara for the air and
water container of a water organ ?
perhaps with crown
and base mouldings like an altar; and Cato (Agr. 18.6)
probably used it for the press bed of an olive press (some editors emend ara to area, the word used for a threshing
floor, but an olive press bed is usually a solid block, i.e.
it is a bomiskos as defined by Hero, Def. 114). These
extensions of meaning cannot be explained by
etymology, since the root of ara probably relates to ashes
(Glare 1982), and the word never meant a 'base or
platform in general', as bomos originally did. They can
only be explained as a transfer of the extended meanings of the normal Greek word for altar to the normal Latin
word for altar, on the basis of a shared understanding of
the normal form of an altar.
The comments of Pausanias (3.19.3) on the platform for trumpeters and heralds at Olympia suggest that a
bomos not only had a recognisable shape, but that this
shape was normally associated with the function of
sacrifice. That raises the question whether the same
association was also felt in connection with the other
altar-shaped pedestals discussed in this paper; that is, were they called bomoi only because of their shape? Or
did the form and terminology also carry some symbolic
meaning? We have seen that in the architectural context
the answer to the second question is probably 'no'. But
in the other contexts the transfer of meaning is suggested not only by the use of the term bomos (whose meaning is arguable), but also by the indisputable similarities in
form, both in general, in the transfer of specific features
from votive altars to statue bases, or vice versa. In
addition to the akroteria already discussed, a strong formal link can be seen between the statue bases and
funerary altars of Roman Macedonia. From the first half
of the second century AD a pediment was often added
144
Coulton
above the crown mouldings on three sides of a normal
rectangular bomos, both for altars and statue bases
(Adam-Velene 2002; Gounaropoulou, Chatzopoulos 1998: nos 75, 367 etc.). This addition may derive from
earlier funerary stelai with pediments and akroteria
(Adam-Velene 2002: 59), perhaps combined with the
pedimented wind breaks at the ends of rectangular Classical and Hellenistic altars (for example, Yavis
17, fig. 15; Smith 1991: fig. 214). It is rather rare in
Italy and Asia Minor (Adam-Velene 2002: 59), but
appears on Augustan statue bases from Teos (Herrmann 2000: 91-93). Although one Macedonian statue base of
this form is dated to the first half of the second century AD (Adam-Velene 2002: no. 239), most of the early datable examples in Macedonia are funerary altars
(Adam-Velene 2002: nos 330, 299, 301, 304, 118, 93); most of the statue bases follow about 50 years later (for
example, Adam-Velene 2002: nos 229-30, 232-36). It
is possible that there the motif was transferred from
altars to statue bases; but at the very least it shows that
the two categories were regarded as equivalent and
developed in parallel. The difficulty, sometimes the impossibility, of distin
guishing altars from bases has often been noted (for
example, Altmann 1905: 4; Hermann 1961: 30, 60-73;
Schraudolph 1993: 23-27; Adam-Velene 2002: 28), but
it is normally regarded as just a puzzle for archaeologists,
arising perhaps from differences between ancient and
modern terminology (as Hermann 1961: 10). But, as
noted above, the similarity goes back in some cases to the
sixth century, and in the Hellenistic period it becomes
both closer and more widespread, with a large majority of statue bases taking the standard forms of round and
rectangular altars. A similar convergence of the forms of
marble well heads (putealia) and round altars occurred in
the late first century BC (Dr?ger 1994: 29-30). The long continuation of this similarity between altars and
pedestals, and the recurring transfer of detailed formal
features between them, must imply some convergence of
ideas relating to these objects over a period of several
centuries. Dr?ger is one of the few scholars to see that
the similarity of forms needs some explanation, and he
too suggests that when the altar form was used as a base, it carried over some of its symbolic meaning.
The use of altar forms as pedestals in a funerary context may not be so surprising. In addition to the
widespread use of funerary altars, the treatment of
sarcophagi as funerary monuments to be displayed above
ground also became widespread in Hellenistic Asia
Minor and the Aegean. Given the heroising ideology revealed by funerary inscriptions, the idea of placing the
ashes or body of the deceased in or on a funerary altar,
rather than beside it, does not seem a difficult step to
conceive or accept. Such changes would often make it
impossible to use the 'altar' for sacrifice, but, as we have
seen, funerary altars were frequently symbolic in intent, even when there was nothing in or on them.
The suggestion that statue bases might be given a
form with meaning is unremarkable in itself, for statues
on columns are well known, and the ship's prow beneath
the Nike of Samothrace has never been disputed. The
choice of the altar forms to carry statues is less straight forward. In a funerary context one might suppose that
placing a free-standing image of the dead 'hero' on top of
a funerary altar would be considered little different from
placing a container for his ashes or body on the altar, and
that this usage led to statues on 'altars' in other contexts.
But there is no evidence that this was the sequence
actually followed, and a first step in a sacred or civic
context is less easily understood. Although both sacri
fices and votive statues might seem to be offerings to a
divinity, different verbs were characteristically used for
blood sacrifice (6?eiv) and dedication (?cvaT?0Evai), and
in principle blood sacrifices were offered at a bomos, while other offerings were made at a trapeza (table).
However, there is evidence that the clear distinctions
to be found in the late lexicographers were not always maintained. The inscription on a rectangular altar from
2 the Asklepieion at Athens (IG 2 4986; third to second
century BC) uses the word B?eiv for cakes, while cult
regulations at Pergamon require the god's share of the
blood sacrifice to be put on the trapeza (Habicht 1969:
no. 16.6-7). An inscription from Termessos records that
a trapeza was placed on top of an altar (t?v ?conov o?v
Trj ?TTiKEin?vri TpaTr??ri; TAM 3.1: no. 916). A
catalogue of victors in the Romaia games at Xanthos in
Lycia (Robert 1978b: esp. 282-83) records that when
there was no contestant, the victor's wreath was
dedicated on the bomos of Roma. The practice of
worshipping living individuals as heroes had already
begun in the early fifth century (Currie 2002), and
continued in the Hellenistic period for non-royal as well
as royal benefactors (for example, SEG 1997: 1218; Gauthier 1996); in such cases an altar might seem an
appropriate base for the living hero's statue. But openly divine or heroic honours for the living were later
restricted to emperors; since the placing of statues on
recognisably altar-shaped bases continued, it can not
have carried such a strong message. The earliest
plausible instance of bomos used for a statue base is for
a statue of Zeus in the first century AD, but the termi
nology is not conclusive. The first more or less certain
usage is in the inscriptions relating to the silver figures of
the gods donated to Ephesos in the early second century
(table 4a). This may be an accident of the evidence,
145
Anatolian Studies 2005
however, for statue bases were already marked out by
akroteria as symbolising altars in the late first century BC
(table 3c). Vase paintings showing divinities (or their
symbols) on altars occur as early as the fifth and fourth
centuries BC (De Cesare 1997: for example, figs 36, 50,
71, 105, 124). These were often intended to show the
divinity in person on the altar, not a statue of the divinity, so they do not actually represent the phenomenon of
statue bases as altars. However, they show that a divinity could already be envisaged as present on the altar, and
the statue of a divinity might equally be thought to show
'the divinity in person', and so to be appropriately placed on an altar. Thus one might suppose that the practice of
putting statues on (symbolic) altars began with statues of
divinities on altars dedicated to them, and was extended
first to deified rulers, then to other local benefactors. But
such a chronological sequence has yet to be established.
However, the aim of this paper is not to explain how
the altar form came to be used as a pedestal for statues
and other things, but to show that the form of such
pedestals was recognisably that of an altar, and to argue that the use of the word bomos to refer to pedestals or
platforms reflects that form. Thus in addition to its
normal meaning of 'altar', bomos does not retain (or return to) an original meaning of 'pedestal or platform in
general', but rather it extends from the meaning 'altar' to
cover 'altar-shaped thing (regardless of use)' ? which
may include a pedestal or platform. Given the number of
bases produced in Roman Asia Minor, the conscious
reference to an altar would tend to become diluted, and
was no doubt sometimes absent (as in the case of column
pedestals). But the fact that the connection continued to
be conveyed by word or by form over several centuries
suggests that the formal similarity was often recognised, and that the symbolic value of the form could sometimes
still be felt. If so, it should affect our understanding of
one of the commonest and most prominent artefacts of
Greco-Roman city culture.
Acknowledgements
Many people have assisted my investigations on altars
and statue bases, some perhaps without knowing it. I
should like to thank particularly R.R.R. Smith and N.P.
Milner, who started the train of thought and helped
beyond any call of duty. I am also grateful for comments
and suggestions to T. Corsten, J.R. Green, J. Ma, S.
Mitchell, C. Rouech?, M. W?rrle, and to those who
contributed to the discussion at seminars in Oxford and
London where I presented preliminary ideas related to
this paper. It should be said that several of those named
disagreed with my argument, so although it is much
improved by their contributions, none of them is in any
way responsible for its shortcomings.
Note on the tables
Tables la, lb, 2a, 2b, 3a-3c and 4c-4e are not intended to
be exhaustive, but to provide a sufficient body of examples to sustain the argument. Tables 4a, 4b and 5-7 are
intended to be comprehensive, but no doubt fail in that aim.
Locations are normally given in terms of ancient city
(including its territory) or, where that is problematic, in
terms of geographical areas or reasonably well-known
modern settlements. The aim is to indicate the general
spread of a usage, rather than the specific find-places of
individual monuments. Dates are taken, where possible,
from the publication cited. Where no greater precision is
possible, 'Imp.' indicates broadly first to third centuries
AD, usually before AD 212 (non-'Aurelian' names); Aur.
N. indicates the presence of 'Aurelian' names (so
generally post-AD 212); ND means no date offered (but in most cases probably equivalent to Imp.).
Table la. Rectangular votive altars called ?coiioc
Nikomedeia, 1C BC/AD, to Zeus Soter. Halfmann, Schwertheim 1986: 132, no. 3
Hierapolis, 27 BC-AD 14, to Augustus. Ritti 1979: 186
87; to ava0E|ia Kai t?v ?couov
Doliche, AD 57/58, to Theos Dolichenos. Wagner 1982:
162-63, no. 5
Mopsuestia, AD 64/65, to the God and Demos. Dagron, Feissel 1987: 132-33, no. 85
?coiacp ?v?6r|KEv; on lower part of rectangular bomos. Marek 1985: 133-34, no. 2
Laodikaia/Lykos, after AD 137, Hestia; 'EoT?av . . . o?v
Tr? ?aoEi Kai tco ?coiacp; on a rectangular block,
perhaps the step below a bomos. Corsten 1997: no.
65
Ephesos, Antonine?, Artemis Ephesia; ... to
?ya?iaa . . .o?v TravTi tco TTEpi a?T? K?oiacp
Ka\ t?v ?coiaov; on a round bomos; a separate altar or the statue's base? Engelmann et al. 1980b:
no. 1266
Thyateira, mid-2C AD, local ?lite; KaTaoKEu?oaoa
t?v ?coiaov Kai ?-rrioKEuaoaoa t?v ?vBpiavTa;
shape unknown. TAM 5.2: no. 974 (CIG 2: no.
3488) Laodikaia/Lykos, AD ca. 150-200, local ?lite; av?oTr|OE
[statue] o?v Tr? ?aoEi Kai tco ?coiacp; on a rectan
gular bomos. Corsten 1997: no. 51b
Aphrodisias, AD 150-200?, Asklepios and Hygeia; t?v 'AoK?riTTi?v Kai t?iv 'YyE?av o?v to??
?coiaoTc; on a rectangular bomos. Nutton 1977:
192, no. 1
Antioch, 2C AD?, relief of two women; t? ?ya?|aaTa Kai t?v ?coiaov ?Tro?r|oa; on a block with a relief, not appropriate to either of the items referred to.
Jarry 1982: 95-96, nos 41-42
Nikaia, AD 210, Zeus?; t?v ?coiaov Kai to ett' a?Tcp
aya?jaa; on a rectangular bomos. ?ahin 1981
1982: no. 1503
Aphrodisias, early 3C AD, local ?lite; Troir|oa|a?vou Se
Kai t?v ?coiaov a?Tcp Kai t? ?oitt? Trapa
?auTcp; on a rectangular bomos. Inan, Alf?ldi
Rosenbaum 1979: 210-13, no. 186
Ephesos, Imp., unknown; tt\v ??ETripiav Ka[\] Ta
tt?vte ?ya?jaaTa o?v to?? ?coiaoTc . . .; on a
rectangular bomos. Engelmann et al. 1980b: no.
1139
Ephesos, ND; t?v ?coiaov g?v tco ?tt' a?Tcp ?uavcp; words clear, but on a slab with a relief, not appro
priate to either of the items referred to. Meri? et al.
1981: (IEph 7.1) no. 3228
Nikaia, ND, Zeus?; t?v ?coiaov o?v tco ?ya?iaaTi; on
an 'ara'. TAM4.1: no. 57
Afyon ili, ND, Zeus and Apollo; t?v A?a ke t?v ?tt'
a?Tcp ?coiaov k? t?v ?v tco ?coiacp 'Att?aaovo
?v?6r|KEv; on the plinth of a statuette. Drew Bear
1993: 147-52
Apollonia, AD 222, small oxen dedicated to Zeus;
yEiap?Ta? Boio?? to?oS '?0?|ar|v ... o? |a?ya
Bcopov ?ycb t?v ?coiaov E?rjKa; on a rectangular
bomos, itself a significant part of the dedication.
MAMA 4: no. 140; Robert 1980: 244-45
Table 4b. Statue bases called ?coiiic
Pogla, AD 222-235, local benefactor; ccvSpi?oiv Tp?Giv [Kai ?co]|a?ioiv
. . . tcov ?v5piav[Tcov Ka]\ tcov
?coiaEi'Bcov; on a rectangular bomos. Bean 1960:
no. 105
Table 4c. Statue bases called ?aoic
Delos, 157/156 BC, Hestia; 'EoT?av . . . etti ?coiaioKou ?ioivou Ka6r|iaEvov Kai etti ?aoEcos ?ioivris; on
a stele; Hestia was probably shown sitting on an
altar (the bomiskos). ID: 1416 AI.83-84, 1417
BI.89-90
Delos, 146-144 BC, child; TraiS?ov xa^Kou[v] laiKpov etti ?aoEcos ?iBivris Kai ocpov5??[ou xa]^Ko?; on a stele. ID: 1442 A.79
Ankyra, ND; t?v ?cou.ov Kai tt]v ?oto?t?ktiv. Kubi?ska 1997: nos 15-17
Thessalonike, AD 224/225; t?v ?cojaov Kai tt]v
?oto?t?ktiv. Kubi?ska 1997: no. 9
The form of nos 4-5 is known; no. 18 is unambiguous; nos
9-10 lack the bomos; no. 8 has, and no. 11 suggests, a different form from nos 4-5; and nos 6-7,12,15 17 are unillustrated. Nos 1-3 have a form compa rable to nos 4-5, but do not mention a ?coiaos.
Table 6a. Large funerary platforms called, or
probably called, ?couoc Note. All instances from Aphrodisias, Ephesos and
Hypaipa are on plain blocks, but the sense demands
a structure rather than a monolithic bomos. All
instances from Hierapolis are on large altar-shaped structures except nos 85,162 (on marble slabs), 100,
103 (on sarcophagi) and 193 (on a small bomos).
Aphrodisias, Imp.; ? ?coiaos Ka\ ai ?v a?Tcp ioc?oTai