Introduction

Ancient lamps-in particular the Hellenistic and Roman attempted here , given their special significance in Jewish
mould-made types-come in many varieties. No existing traditions. The Iron Age Palestinian lamps fall outside
collection (or publication) embraces more than a fraction the main Mediterranean tradition of lamp making. All of
of the known types and decorative motifs , though some the above specimens, some 650 in number, are reserved
assemblages based on locally excavated material (e.g., for future publication. Lamps and lighting devices in
the finds from the Athenian Agora published by metal , stone, and other materials are also excluded, s as
Howland and by Perlzweig) 1 may come near to are terracotta figurines with small lamp attachments.
illustrating the full range of production in a particular Returning to the Greek and Roman types of clay lamps
region . Only by combining the published items from which form the subject of this volume, one may observe
many collections may we arrive at a reasonably complete that a number of authors over the last hundred years
picture of lamp production in ancient Greek and Roman (notably Dressel , Petrie , Loeschcke, Broneer, Vessberg,
times . In this respect the study of ancient lamps may well Howland, and Deneauve; see Bibliography)" have
be compared with numismatics: as yet, no single offered typologies of ancient lamps and of the motifs on
published work can be said to present anything them. These are generally most successful when applied
approaching an all-embracing view of the subject; in to lamps found or made in a single spot (e .g. ,
fact, the very diversity of the material has lead to a dearth Vindonissa, Athens) . When material of diverse origins,
of general introductory studies . The present catalogue such as is likely to be present in any major museum
does not attempt to fill the gap (any such claim would be collection, is presented, its sheer diversity is liable to
presumptuous); it should be viewed rather as another defy any attempt at classification on the basis of shape or
contribution to the growing corpus of known material , to mot ifs alone . Thus, rather than adding to the already
be taken into account in any future synthesis . General confusing multiplicity of typologies , I shall here content
comments on certain classes are offered where it seems myself with citing those already in existence, where
that they might further knowledge; in this respect the relevant. A more productive approach-and that adopted
present work avoids being merely a bald catalogue . But I here- is to divide the material on a regional bas is and to
have avoided lengthy discussions in favour of presenting attempt to identify fabrics, which largely illustrate
the material at hand. parallel traditions, within each region . However, where
The scope of this volume, which is envisaged as the relative uniformity of fabric occurs in the products of a
first of two embracing the collections of ancient clay whole region (e.g., in Cyprus), I have reverted to a more
lamps in the Royal Ontario Museum , may be defined as typological approach. The degree of classification
lamps of Greek and Roman types, both wheel -made and attempted naturally varies according to the number of
mould-made, from all regions of the ancient classical lamps of given origin in the collection and the amount
world, including their successors in the Roman tradit ion known about them.
down to the 7th and 8th centuries A.D. It excludes, more The techn ical details of production of ancient clay
for reasons of convenience than of logic , categories of lamps have been discussed by a number of writers, 7 and
lamps of Roman date native to Egypt (the so-called are relat ively well known . No more than a brief
" frog lamps" and their relatives)," and Palestinian restatement of certain points is called for here . Broader
lamps in the Jewish tradition and their Iron Age questions relating to lamps and lighting in ancient times,
predecessors. 3 The reason for this may be stated partly as and the references to them in ancient literature, are
lack of precise dating evidence-a situation which touched on by Robins and by Forbes , among others. 8
should be much clarified over the next few years by the Exhaustive bibliographies of modern literature on an-
publication of certain studies of excavated material now cient lamps are presented in several of the more recent
in preparation." The ROM'S collections in these fields stud ies.
will be much better illum inated when this material is The earlier types of ancient lamps" generally have
available for comparison. In addition , the Palest inian wheel-made bodies, and hence share their technical
lamps require a more iconographical approach than is characteristics with ancient wheel-made pottery . By

J. W. Hayes - 9789004663534
Downloaded from Brill.com 11/11/2024 07:40:52AM
via Durham University
Roman times , however, mass production in two-part indicates that the two mould sections were of rather
moulds was normal ; some actual examples of these different sizes .
moulds are listed in Appendix I. The mould-made series The two main sources of the lamps in the ROM are
had its origins in Greece and Egypt in the 3rd century purchases made by Dr. C.T. Currelly (the first Director
B.C. , and persisted until the 7th and 8th centuries A.D. in of the Museum) in Egypt-particularly the Fayoum
various Mediterranean lands . The analog ies here are with region-when he was working there alongside W .M .F.
the production of plastic vases and terracotta figurines, Petrie in the years around 1903 to 1905, and the large
and similar terminology is generally employed. I 0 purchases of Palestinian and other lamps from Vester
The use of moulds encouraged the development of (Jerusalem);' " made around 1910. Many of the former
relief ornament, and once a decorative type had been and a few of the latter (most of which are of non-Roman
created it could be reproduced virtually ad infinitum . The Palestinian types) are listed here. More than 120 of the
exchange of moulds between manufacturers, or the ROM's "Fayoum" lamps were illustrated by Petrie in
taking of fresh impressions from existing lamps, results Roman Ehnasya (published in 1905) prior to their arrival
in the appearance of identical types on the products of in Toronto (see page 145 for list). 13 This fact was lost
different lamp-makers and indeed on lamps from sight of when they were entered in the Museum's
different regions. The latter may be established on the inventory some years later , and has only come to light
basis of fabric (i.e., body clay and surface treatment), again quite recently by a close comparison of the
the former by means of the makers' stamps and originals with the published photographs. The Toronto
signatures, II where these occur. But even the appearance lamps are not finds from the Ehnasya excavations, but
of a stamp or signature is no sure guide to the place of represent supplementary material included by Petrie in
manufacture, since a stamp could be " pirated" as easily his corpus for completeness. Almost all bear ink and
as a decorative scheme, and no laws on the use of pencil markings-including what seems to be a prelimi-
trademarks existed to prevent the practice . An example is nary classification of types superseded(?) in Roman
the stamp FORTIS on the two rather similar lamps , 256 Ehnasya-which show that they were a single collection.
and 262, one a North Italian original, the other (as The type numbers also appear on other "Fayoum" lamps
proved by its fabric) a "provincial" product. in the collections not published by Petrie ; the provenance
Three features peculiar to mould-made lamps (and of these lamps might occasionally be open to doubt (i.e .,
terracottas also) may be noted here . One common they may come from elsewhere in Egypt) . Further lamps
characteristic is the progressive reduction in size in each come from subsequent Egypt Exploration Society exca-
"generation" of copies, the inevitable result of the vations, from the Sturge Collection (acquired in 1918,
shrinkage of clay during firing; where clay rather than mostly from Italy) , and from the Loch Collection (from
plaster moulds are used this shrinkage is accelerated. Cyprus, donated in 1965). The rest represent individual
Since lamps derived from a single original were donations and purchases made over the years.
sometimes produced over a period of a century or more, The dates proposed for individual lamps in this
the possibilities of differences in size are considerable. catalogue can be no more than approximate , since none
But the smallest lamp in a series is not invariably the of the lamps listed comes from a closely datable
latest ; size depends on the number of times that copy ing archaeological context (even those from excavations are
occurred. The second common characteristic is modifica- generally unstratified). Some of my datings are , frankly ,
tion of the original patterns, whether by the touching up subjective. Where good parallels are known from dated
of worn details by incision in a freshly taken mould or deposits, these are indicated in the list of comparanda,
cast before it hardened or was fired , or by the addition of and the date cited for the lamp in question can be
an extra feature (see 434, 442) , or by the insertion of a regarded as relatively precise. If other examples of a
replacement for a broken element (normally the particular motif are known, an approximate date is
discus-see 326, 433). The Cypriot series 33~ shows arrived at by estimating the position of the catalogued
both features (reductions in size and pattern modifica- item within its series, and by using typological criteria,
tions) in combination, illustrating the innumerable minor such as the shape of the lamps, to estimate how long a
variations possible within a single decorative type. particular series survived . Where the parallels cited are
Thirdly, the top and bottom moulds did not always form not precisely datable by independent means (as is usually
matching pairs; these moulds could be interchanged- the case), or where no comparanda exist , a possible date
hence the appearance of different makers' marks on is determined by a complex consideration of shape,
lamps with the same motifs. In some instances an offset decorative details , and fabric , on the assumption that
or marked paring along the junction of the two halves some sort of consistent typological development can be

2 / Introduction

J. W. Hayes - 9789004663534
Downloaded from Brill.com 11/11/2024 07:40:52AM
via Durham University
traced (here in the form of a nozzle, there in the probable 7. E.g., F. Fremersdorf, Romische Bildlampen (Bonn/leip-
life span of a particular production centre, and so on) . zig, 1922); Perlzweig (1961); D.M. Bailey, Greek and
This process is inevitably rather subjective, and depends Roman Pottery Lamps (London, 1963); Bailey, A
on which features are considered by the author to be Catalo gue of the Lamps in the British Museum I (London,
1975) 2-8; Provoost (1976) 14-18 . For the basic
chronologically significant ; I may have made some
terminology (in various languages), see Provoost (1976)
wrong choices. Some idea of the probable accuracy of
7-10.
any date proposed in this catalogue may be gained by a 8. F.W. Robins, The Story of the Lamp (O.U.P. , London ,
glance at the list of comparanda cited, either for an 1939); R.J. Forbes, "Light", in Studies in Ancient
individual lamp, or for the series or type to which it Technology VI (Leiden, 1958) 119-93 . For a good brief
belongs . introduction to the subject, see D.M. Bailey, Greek and
Roman Pottery Lamps (British Museum , London, 1963;
2nd ed. 1972). A short survey is also given by H. Menzel
NOTES in Enciclopedia dell' Arte Antica IV 707-18. One of the
most comprehensive surveys is that presented by A. Hug
I. R.H. Howland, The Athenian Agora IV (Princeton, 1958); in Pauly/Wissowa (eds.) , RE, though this is now more
1. Perlzweig, The Athenian Agora VII (Princeton, 1961). than 50 years old and lacks any illustrations. For the earlier
2. For recent statements on these, see L.A . Shier, "The Frog literature on the subject, see J. Toutain, "Lucema,
on Lamps from Karanis", in (ed. Sami A. Hanna), Lychnus " , in (eds. Daremberg/Saglio/Pottier) Diction-
Medieval and Middle Eastern STUdies in Honor of Aziz naire des antiquites grecques et romaines 111.2 (publ,
Suryal Atiya (Leiden, 1972) 349-58, with 4 pis.; M. 1904) 1320-39.
Michelucci, La co llezione di lucerne del Muse o Egizio di 9. Historical survey: A. Hug, in Pauly/Wissowa, RE 13.2,
Firenze (Firenze , 1975) 65-71 , tipo XXIX. 1606-13 (now partly superseded by Bailey [1963] and
3. For a survey of these, see a series of three articles by R.H . others).
Smith in The Biblical Archaeologist (27.1 [1964] 1-31 ; 10. On this, see especially R. V. Nicholls, "Type, Group and
27.4 [1964] 101-24; 29.1 [1966] 2-27). Series; A Reconsideration of some Coroplast ic Fundamen-
4. One may cite the Karanis, Kellia, and Alexandria (Polish tals" , BSA 47 (1952) 217-26, pis. 44-45.
Mission) finds from Egypt, those from various Nubian II . For a summary listing of the known stamps and signatures,
sites investigated in the 1960s, and the mass of material see L. Mercando , in Enciclopedia dell' Arte Ant ica ,
discovered in excavations in Israel (of which only a tiny Supplemento 1970 (publ. 1973) 419--42.
proportion has so far appeared in print). 12. J. Vester & Co., American Colony Store , Jerusalem ;
5. For publication of these, see (C. R. Wason), Bulletin ofthe lamps collected in Palestine and neighbouring regions by
Royal Ontario Museum of Archaeolog y II (March 1932) J .D. Whiting. Another large batch of these lamps was
13, with fig. in text (Minoan stone pedestal lamp); N. acquired by Yale University, and is published by C.A .
Leipen, " A Bronze 'Wall -Bracket ' ", 1960 Annual. Art Kennedy in Berytus 14.2 (1963) 67-115, pis. XX·XXXI.
and Archaeology Division . Royal Ontario Museum 21-26 , 13. The rest of the " Ehnasya" lamps are scattered fairly
pis. IV-V (Cypriot). widely. While the bulk of them remain among Petrie's
6. For a recent attempt at an overall typology for ancient finds at University College, London, others are in the
lamps, see A. Provoost, L'Antiquite Clas sique 45 (1976) British Museum (see D.M . Bailey, BMC Lamps I and
21-39, and 550-86. This inevitably suffers from over- forthcoming) and in the Oriental Institute, Chicago (nos.
generalization, and is biased towards finds from Italy. It 8372-8418; some now dispersed); the whereabouts of
unfortunately lacks proper cross-referencing to the existing others are unknown to me.
classifications.

Introduction / 3

J. W. Hayes - 9789004663534
Downloaded from Brill.com 11/11/2024 07:40:52AM
via Durham University