AN OFFPRINT FROM
SMALL FINDS AND ANCIENT SOCIAL
PRACTICES IN THE NORTH-WEST
PROVINCES OF THE ROMAN EMPIRE
Edited by
STEFANIE HOSS & ALISSA WHITMORE
Paperback Edition: ISBN 978-1-78570-256-3
Digital Edition: ISBN 978-1-78570-257-0 (epub)
© Oxford & Philadelphia 2016
www.oxbowbooks.com
Published in the United Kingdom in 2016 by
OXBOW BOOKS
10 Hythe Bridge Street, Oxford OX1 2EW
and in the United States by
OXBOW BOOKS
1950 Lawrence Road, Havertown, PA 19083
© Oxbow Books and the individual authors 2016
Paperback Edition: ISBN 978-1-78570-256-3
Digital Edition: ISBN 978-1-78570-257-0 (epub)
A CIP record for this book is available from the British Library
All rights reserved. No part of this book may be reproduced or transmitted in any form or by any means,
electronic or mechanical including photocopying, recording or by any information storage and retrieval
system, without permission from the publisher in writing.
Printed in Southampton by Hobbs the Printers Ltd
For a complete list of Oxbow titles, please contact:
UNITED KINGDOM
Oxbow Books
Telephone (01865) 241249, Fax (01865) 794449
Email: oxbow@oxbowbooks.com
www.oxbowbooks.com
UNITED STATES OF AMERICA
Oxbow Books
Telephone (800) 791-9354, Fax (610) 853-9146
Email: queries@casemateacademic.com
www.casemateacademic.com/oxbow
Oxbow Books is part of the Casemate Group
Front cover: Gladiator glass © The Vindolanda Trust; York gorgoneion (no. 7) © York Museums Trust, Yorkshire Museum;
“R”-shaped brooch from recent excavations in Forum Hadriani, © Restaura.nl.
Contents
1. Introduction: Small finds and ancient social practices
Stefanie Hoss and Alissa Whitmore
1
PART 1: SMALL FINDS, THE BODY AND IDENTITY
2. Iron footed – hobnail patterns under Roman shoes and their functional meaning
Boris Burandt
3. Wearing socks in sandals: The height of Roman fashion?
Barbara Köstner
4. Laying it on thick – makeup in the Roman Empire
Gisela Michel
5. Of brooches and men
Stefanie Hoss
9
16
28
35
PART 2: RELIGION AND RITUAL IN THE ROMAN NORTH-WEST PROVINCES
6. Ordinary objects transformed: the compound natures of material culture
Mara Vejby
7. A Mars with breasts from Weißenburg in Bavaria
Nicole Birkle
8. Metropolitan styling: metal figurines from London and Colchester
Emma Durham
9. Staring at death: the jet gorgoneia of Roman Britain
Adam Parker
57
68
75
98
PART 3: ARTEFACTS, BEHAVIOURS AND SPACES
10. Dining with Mithras – functional aspects of pottery ensembles from Roman Mithraea
Ines Klenner
11. Cloth working in the baths? Site formation processes, needles and spindle whorls in Roman
bathhouse contexts
Alissa M. Whitmore
12. The complexity of intramural and extramural relationships on the northern frontier of Roman
Britain – a Vindolanda case study
Andrew R. Birley
117
128
146
11
Cloth working in the baths? Site formation processes, needles and
spindle whorls in Roman bathhouse contexts
Alissa M. Whitmore
Keywords: Site formation; deposition; bathhouse; cloth working; military; gender; class
The recovery of needles, spindle whorls, loom weights and other cloth working instruments from Roman public
and military bathhouses raises the question of whether sewing, spinning and weaving – activities otherwise
unattested in the baths – took place in these spaces. This paper evaluates the possibility of cloth working in
Roman bathhouses by examining site formation processes, focusing on drains as the most probable context for
artefacts related to the normal use of the baths. A review of the alternative functions and possible owners of
textile implements discovered in bathhouse drains suggests that while these activities were uncommon, sewing
and spinning occurred in a few baths and these needles, spindle whorls and related objects likely offer physical
evidence for the activities of lower class bathers and bathhouse attendants.
Introduction
The last three decades have witnessed an increase in artefact
studies in a wide range of ancient spaces. This research has
greatly enriched our understanding of how people actually
used built space, particularly when comparing artefacts to
information provided in ancient texts. Lisa Nevett’s study
(1999) of small finds in ancient Greek houses indicates that
there was little gender segregation in the use or storage
of men’s and women’s objects, overturning decades of
scholarship which tried to locate women’s quarters in
ancient Greek houses. Penelope Allison’s research (2004)
on artefacts and furniture in Roman houses illustrates the
multifunctional use of many spaces and the dangers of
depending upon ancient texts alone. Studies of small finds
have also suggested the presence of women and children in
Roman forts and barracks (van Driel Murray 1994; AllasonJones 1999; Allison 2006). Caution is necessary, however,
when interpreting artefacts for the presence and activities of
ancient peoples, since object findspots do not always indicate
places of use. Significant questions, for example, have
been raised about the contexts of women’s and children’s
items in barracks, since relatively few were recovered
in clear stratigraphic levels linked to the occupation of
barrack buildings, rather than construction deposits or later
dumps (Hodgson 2014; Becker 2006). Careful attention
to deposition practices is crucial when evaluating small
finds and offers the potential for a better understanding of
ancient spaces and behaviour. This is especially the case
when evaluating unique finds in settings without previous
studies of small finds assemblages, such as cloth working
items in Roman bathhouses.
With this in mind, I seek to critically interpret the
discovery of needles, spindle whorls, weaving tablets
and loom weights found in Roman public and military
bathhouses. While cloth working artefacts make up a very
small percentage of bathhouse finds assemblages (typically
less than 3%, see Whitmore 2013), and they are absent
entirely from many sites, these objects have been found
in a number of baths and provide potential evidence for
an activity that is otherwise unattested in these spaces.
To evaluate whether these needles, spindle whorls and
weaving tools provide evidence for a new activity in
11. Cloth working in the baths? Site formation processes, needles and spindle whorls in Roman
Roman bathhouses, or if the objects are simply intrusive,
this paper focuses upon the Roman construction, demolition
and deposition practices and archaeological site formation
processes behind the stratigraphic contexts in which these
objects were found. First, however, an introduction to
Roman bathing is useful.
A brief overview of Roman bathing
Only the wealthiest individuals could afford spaces in their
homes that were dedicated to bathing, which for the Romans
entailed more than simply cleaning the body. Instead, the
vast majority of the population bathed at the public baths:
large, multi-room structures which offered a variety of
bathing options, including hot and cold pools, dry and wet
sweat rooms and rooms devoted to massaging, cleaning and
grooming the body. Ancient texts suggest that the majority
of bathhouses were visited by people of all genders, ages
and classes, and the presence of bathhouses in ancient
literature, inscriptions and the archaeological record attests
to their popularity in large and small towns and military
communities throughout the Roman Empire (Fagan 1999;
Yegül 2010). In the Northwest provinces, public baths
were associated with military forts and large urban centres
and are often viewed as spaces involved in the creation
and maintenance of Roman identity (Nielsen 1993, 73–84;
Revell 2009, 172–179).
Written sources also attest to the important function of
Roman baths as social centres. It was here that the Romans
met with friends and lovers, caught up on the latest news
and occasionally struck business deals and forged political
alliances (Ovid, Ars amatoria 3.639–640; CIL IV.10677;
Martial 3.36; Digest 50.1.27.1). Passages from Seneca
(Epistles 56), Martial (11.82) and Juvenal (Saturae 6.419–
425) mention a number of activities which occurred in
these lively spaces, including eating and drinking, exercise,
massage, depilation and sex. Many rooms in public and
military bathhouses lack pools or other obvious architectural
markers of hygienic activities, which may underline the
importance of the socialization, and spaces for it, in the
baths (DeLaine 1992; Revell 2007). Artefacts recovered
from bathhouse contexts (Whitmore 2013; forthcoming)
provide material evidence for many of these social activities,
especially grooming, eating, drinking and status displays,
and also indicate others, such as gaming, medical procedures
and, perhaps, cloth working.
Artefact deposition and site formation processes
in Roman baths
Before turning to the cloth working artefacts, it is first
necessary to consider how the architecture of Roman baths,
ancient behaviour and site formation processes, affected the
129
ways in which objects became part of the archaeological
record. Like any other site, the artefact assemblages
recovered in baths provide only a partial picture of what
once occurred in these spaces, and several factors assuredly
limited or prevented many objects from becoming interred.
The majority of bathhouse floors are flagstone or mosaic,
and these solid surfaces would have made it easy for bathers
to retrieve lost possessions and for attendants to clean
the floors, an activity which ancient sources and modern
scholars suggest occurred with some frequency (Seneca,
Epistulae 86.10; Pliny, Epistulae 10.32.2; CIL II.5181;
Zienkiewicz 1986a, 243). Thus, lost items and debris would
not accumulate or be trod into floors and many objects used
in the baths will never be recovered in these spaces.
Furthermore, not all objects recovered at bathhouse sites
reflect the activities of ancient bathers. The most common
findspots for bathhouse artefacts are discussed below, many
of which do not diverge strongly from contexts at other
Roman period sites. While each bath must be evaluated
individually, I offer general guidelines for interpreting these
types of contexts, drawing upon scholarship on Roman
construction and demolition practices, rubbish disposal and
archaeological site formation processes.
Topsoil, post-abandonment and demolition
accumulations: the afterlife of baths
Roman period artefacts are often recovered in the upper,
poorly stratified layers of excavations, as at the Ribchester
military baths (UK), where ancient and medieval robbing,
antiquarian excavations and backfill trenches have resulted
in mixed deposits (Godwin n.d.).
At some sites, there is clear evidence of artefacts being
introduced into abandoned baths via natural processes, as at
Mirobriga (near Santiago do Cacém, P), where soil, bone,
Roman ceramics and a glass bracelet have washed into the
courtyard and service yard of the West Baths from a nearby
hill (Biers & Biers 1988a, 73; 1988b, 172, 175; cf. Schiffer
1985, 30). While these poorly stratified finds offer a general
picture of life in the neighbourhood of the baths, they cannot
be used to populate the baths with activities and people,
even if the debris includes bathing instruments.
While the floors of buildings are often considered
occupation contexts (e.g. Gardner 2007, 70), it seems
unlikely that we can interpret assemblages in the majority
of bathhouses in this way. If left standing, abandoned
baths were often reused by squatters or became industrial
workshops during Roman and later periods (e.g. Ellis
2000, 75–78; Gregory 2010, 15), and some bathhouse
features and accumulations can be tied to these activities.
A large quantity of rubbish, mostly cooking pots, pottery
and animal bones dating to the 4th century, was found in
the Vindolanda 3rd century baths (UK), and are believed
130
Alissa M. Whitmore
to have been dumped there by squatters or stone robbers
after the baths were abandoned (Birley 2001, 6–7). While
these contexts and finds provide fascinating insight into the
afterlife of bath buildings and the individuals living and
working in these repurposed spaces, they do not provide
any evidence relevant to the bath’s period of use. Aside
from sites with rapid abandonment and interment, it is
unlikely that objects used in the baths would have remained
in these spaces, since salvagers or squatters presumably
would have collected any valuable, functional or recyclable
materials (cf. Schiffer 1985, 26–28; Zienkiewicz 1986a,
42–45; Philp 2012, 75–84; Keller 2005; Peña 2007,
250–271).
Many bathhouses were demolished during the Roman
period for their architectural materials or land, and
these activities can clearly affect our interpretations of
these spaces. Workshops were sometimes set up within
buildings to recycle construction materials (cf. Munro
2010), which occurred in a late 3rd century abandonment
phase at the Caerleon legionary baths (UK), when a
metalworking furnace was built over the frigidarium floor
drain, introducing slag and architectural materials into the
drainage system (Zienkiewicz 1986a, 253–254). Robber
trenches, both ancient and modern, often cut through
numerous bathhouse features and layers to access stone
foundations and drains, mixing different strata and artefacts
(e.g. Gillam et al. 1993, 1; Barker et al. 1997, 6; Philp
2012, 75–76; Proctor 2012, 68–69). Before, during and
after demolition, abandoned baths could also become local
dumps (cf. Schiffer 1985, 29). The abandoned Baths of the
Swimmers in Ostia (IT) were infilled with ceramics and
small finds which are interpreted as neighbourhood rubbish
(Carandini et al. 1968, 8), and this may also explain the
medical instruments found below a layer of destruction
rubble in a room off the palaestra of the Xanten baths (DE,
Hoss forthcoming). During demolition, bathhouse spaces,
especially hypocausts, were often filled with building
rubble (e.g. Daniels 1959, 92–93; Perkins 2004, 42–43;
Philp 2012, 62, 67).
Interpreting these infills is particularly challenging, as
they may represent rubbish generated in the baths, waste
produced by construction or demolition crews, or materials
which were imported from outside areas as fill. While some
finds might relate to activities in the baths, they are not from
primary occupation layers and could easily originate from
elsewhere. This is also true of objects recovered from the
spoil heaps of earlier excavations.
Occupation floors in rapidly abandoned baths: the
baths of Pompeii and Herculaneum
While most artefacts recovered directly on bathhouse
floors are more likely related to post-abandonment
activities, a possible counter-example are baths which were
rapidly abandoned and buried, since these sites may have
preserved occupation floors and avoided the squatting or
demolition activities which introduce outside artefacts.
The best known rapidly abandoned baths are in Pompeii
and Herculaneum (IT), but as scholars of Campanian finds
assemblages have warned (Allison 1992; Wallace-Hadrill
2011, 272–280), the notion of in situ artefacts at these
sites can be misleading.
The baths at Pompeii and Herculaneum are greatly
affected by their early excavation dates. Data on excavated
artefacts must be recovered from early published sources
(e.g. Fiorelli 1862) and unpublished inventories (Libretti
d’Inventario) and excavation books (Giornali degli Scavi),
and oftentimes, information in these sources leaves much
to be desired. Descriptions of artefact contexts are woefully
uneven, with findspots ranging from very general (‘the
baths’) to more specific (‘between the columns on the
north side of the peristyle’). Rarely is the vertical position
of artefacts from Pompeian bathhouses described, but on
occasion, finds were noted as recovered from a meter or
more above the pavement, presumably in ash and rubble
layers. Whether these objects fell from shelves or terraces
when roofs collapsed, or were swept into rooms from
elsewhere on ash, pumice or lava flows, is unknown, but
this does call into question where precisely many of these
objects originated (cf. Allison 1992, 50).
A particularly significant point when interpreting artefact
assemblages in Pompeii’s baths, however, is that few of
these buildings were open at the time of the eruption.
Possible interruptions to the water supply and evidence
for incomplete repairs following the earthquake of AD 62
suggest that part of Pompeii’s Forum Baths (Eschebach
1982, 319; Koloski-Ostrow 2008, 231) and potentially all
of the Stabian Baths (Maiuri 1931, 574; Eschebach 1979,
70) were closed. The Sarno Baths (Koloski-Ostrow 2008,
240) and Baths of Julia Felix (Parslow 2000) may also
have been under renovation, the Central Baths were not
yet open (Koloski-Ostrow 2008, 224) and the Republican
Baths had been demolished decades prior to the eruption
(Maiuri 1950, 113–116).
Recovering a true occupation floor that relates to
the use of Roman bathhouses is thus difficult, if not
impossible. While the Stabian Baths have a particularly
rich artefact assemblage, the fact that these baths were
closed and still under reconstruction at the time of the
eruption means that these objects might have arrived into
these spaces through any number of ways. The artefacts
might relate to the normal use of the baths, and were
stored here awaiting their reopening, but they could
easily be associated with squatters, an altered use of the
structure, or construction fill or workers. At Pompeii
and Herculaneum, we must be more, rather than less,
observant of context and potential disruptions to the dayto-day functioning of these spaces.
11. Cloth working in the baths? Site formation processes, needles and spindle whorls in Roman
Construction deposits: levelling layers and
foundation trenches
Many bathhouses were in use for several centuries, and
the long lives of these buildings resulted in numerous
construction phases and deposits (Zienkiewicz 1986a,
46–50; Nielsen 1993, 82; Rook 1992, 6; Fagan 1999, 180).
In the later centuries of the High Empire, pools in the
North-west Provinces were regularly closed and infilled
with rubbish while the rest of the baths remained in use
(e.g. Zienkiewicz 1986a, 253–255; Ellis 2000, 68). Waste
was also used to fill up no longer used hypocausts while
the rest of the bath was still in use.
Many of these deposits include artefacts. Ceramic sherds
and broken tile are common ingredients in Roman levelling
layers and floor make-up (Peña 2007, 250; Vitruvius, De
Architectura 7.1.1–5), and construction fills from the Baths
of Caracalla (IT) contained numerous ceramic fragments,
including amphorae, terra sigillata and coarse wares, mixed
with pozzolana, clay, mortar and fragments of brick and
marble (DeLaine 1997, 138). Glass sherds, coins and small
finds are also occasionally recovered in these fill layers.
Beads, fibulae, pins and toilet instruments were recovered in
and around the foundation trenches of the Caerleon Castle
baths (Lee 1850, 17) and intaglios, beads, gaming counters
and a finger ring have been recovered from various make-up
layers at the Caerleon Legionary baths (Zienkiewicz 1986b,
133, 142, 149, 156). Objects recovered from these layers
often provide a date for the construction or renovation of
baths, but since many bathhouses have multiple construction
phases, it is tempting to interpret finds in these contexts
as potentially providing hints about life during the earlier
periods of the baths. We must, however, be cautious when
interpreting the assemblages from such contexts.
Kevin Dicus has presented a compelling argument
for the potential origins of artefacts found in Pompeian
construction trenches and layers. Large-scale construction
projects, such as the creation of levelling layers or infilling
pools, would have required large quantities of sediments and
materials, likely originating from outside the construction
area. At Pompeii, the probable source for such fill appears
to be rubbish dumps outside town walls, which included
construction and demolition materials and refuse originating
from numerous sources dating to various occupation periods.
As a result, artefacts found in levelling layers or infilled
features likely came from the community’s garbage dumps,
rather than within the construction area itself (Dicus 2014,
70–76; cf. Liebeschuetz 2000, 51–54). Undoubtedly, similar
construction practices occurred outside of Pompeii as
well, and this suggests that finds from the vast majority of
bathhouse construction layers cannot provide evidence for
activities taking place during the normal use of these spaces,
even if these objects fit our conception of bathhouse material
culture. Since strigils and bath flasks, the quintessential
131
bathing artefacts, are frequently recovered in settings other
than baths (Wardle 2008, 207–211; Whitmore forthcoming),
presumably because people carried them to the baths from
home (Shelton 1981, 26–28; Carandini et al. 1982, 334;
Dickey 2012, 121–125, 201; Juvenal, Saturae 6.419–420),
any notion of ‘exclusive’ bathhouse material culture seems
overly simplistic.
There are some scenarios, however, which might allow
for materials from inside or around the baths to be interred in
construction layers during small-scale projects. The Romans
may have set aside soil removed when digging foundation
or water pipe trenches for later backfilling. Any finds in
such soil, however, are still unlikely to have originated from
occupation layers, and subsequently would be mixed with
objects from other strata (Dicus 2014, 72–76). Two levelling
layers and a foundation trench at the Caerleon Legionary
Baths might provide a possible exception (Zienkiewicz
1986a, 244–249). In this case, similarities in sediment
colour and consistency, as well as finds assemblages, suggest
that the materials in these construction deposits originated
from a cleaning of the drainage system, a context which
Zienkiewicz interprets as related to the daily operation of
the baths.
Floor and pool drains: traps for accidental losses
and small rubbish disposal
Although not every bathhouse drain contains an artefact
assemblage, material culture, occasionally in very large
quantities, has been recovered in the sediments of some
bathhouse drains. Objects could enter drainage systems
through bathhouse pools or floor drains, the latter of which
were often covered with slotted stone grates to allow water
(and small objects) to enter (cf. Zienkiewicz 1986a, 35–36,
196–199). At least some of these drain artefacts likely
represent items lost by bathers or small rubbish swept
toward floor drains by bath attendants, thus this context
may represent the closest thing to an occupation or primary
rubbish layer in the baths (cf. Schiffer 1985, 24–25).
When interpreting drainage contexts, it is critical to isolate
the strata and objects that represent the normal functioning
of the baths. Material culture related to construction and
post-abandonment activities can enter drainage systems,
and a mass of lead, iron nails, glass and coins attributed
to renovation activities was found under the Phase IV
frigidarium floor drain in the Silchester Public Baths (UK;
Hope & Fox 1905, 351–352). The infilling of pools surely
could have introduced some material culture into adjacent
drain segments, and rubbish and debris were dumped into
sections of the Caerleon drain on at least two occasions after
the closure of the baths (Zienkiewicz 1986a, 249, 253–255).
The robbing of drains can also result in the removal of
sediments and the introduction of artefacts, which occurred
in the Dover Shore Fort Baths (UK), where 4th century coins
132
Alissa M. Whitmore
are believed to have entered sediments when the drains were
robbed in antiquity (Philp 2012, 76). At sites where detailed
stratigraphic excavation of drain sediments occurred, as at
Caerleon, it is possible to isolate drain strata associated with
construction, dumping and robbing, and eliminate objects
that are likely unrelated to bathers.
Using the Caerleon drain stratigraphy as a model, the
sediments and materials most likely associated with the
normal use of the baths are the horizontal layers found
throughout the length of the drain, often consisting of
fine, silty sand that was presumably gradually laid down
while the baths were in operation (Zienkiewicz 1986b, 13).
Since waterborne materials could also flow through the
drain after the closure of the baths, objects found in lower
drain strata are more likely tied to the use of the structure,
though distinctions between drain layers cannot always be
made, and it is always possible that heavier items could
sink through the fine, viscous drain sediments (Zienkiewicz
1986a, 244–249). Detailed stratigraphic drain excavation
and documentation, however, is absent from many sites,
and it is often possible to only associate artefacts with
specific segments of the drain, or merely with the drain
itself. In such cases, especially when there is evidence that
demolition or salvage work may have affected sediments,
drain assemblages can only be viewed as possible evidence
for bathhouse activities.
Latrines, pits and furnaces: bathhouse rubbish
deposits?
No ancient texts reveal what happened to the broken
ceramics, bath flasks, toilet instruments and animal bones
that were inevitably produced during a typical day in the
baths. Ancient sources and scholarship on Greco-Roman
rubbish practices, however, can provide some suggestions.
Salvagers collected and purchased recyclable broken metals,
glass and ceramics (Keller 2005; Peña 2007, 253; Martial
1.41, 10.3; Statius, Silvae 1.6.73–74; Juvenal, Saturae 5.48),
and surely these individuals would have regularly stopped
at the baths. Other sources indicate that carts may have
removed various types of rubbish from town (Thüry 2001,
7–8; Tactius Annales 11.32; Cicero De Divinatione 1.57; CIL
I2 593). Studies of artefact assemblages and refuse disposal
at South Shields and York, in contrast, suggest that there
may have been relatively little movement of objects from
their original places of use, perhaps indicating that materials
from ditches and dumps may illustrate activities occurring
in spaces nearby (Gardner 2007, 85–87).
Contexts in or around bathhouses may have functioned as
disposal sites for garbage produced in the baths. Artefacts are
often recovered in bathhouse latrines, which were frequently
located at the end of a bath’s drainage system. These objects
may have entered the latrine from bathing spaces further
upstream in the drainage system or were lost or thrown in
as rubbish via the seat openings (Van Vaerenbergh 2011a).
While serving as a rubbish dump wasn’t the primary
function of latrines, excavations of Campanian latrines and
sewers have illustrated that the Romans threw kitchen waste
and broken items into latrines and presumably lost intact
objects down household and commercial toilets (Jansen
2000, 38; Wallace-Hadrill 2011, 282–285; Camardo 2011).
Public latrines could not have been permanent or official
garbage dumps, however, as large objects would remain and
eventually compromise the latrine’s functionality, requiring
that it be cleaned out with greater regularity. Instead, it is
more probable that the majority of rubbish was inserted into
latrines after they fell out of use, and only the lowest latrine
layers might contain objects that entered through bathhouse
drains or were lost down a still functioning toilet (Daniels
1959, 145–157).
In addition, bathers were not the only individuals using
bathhouse latrines. Many were located near entrances or
courtyard areas, somewhat separate from the main bathing
block, suggesting that latrines could be accessed without
entering bathing spaces that required an entrance fee (Van
Vaerenbergh 2011b; Koloski-Ostrow 2015, 11, 21). Since
anyone potentially had access to these public bathhouse
latrines, material culture from this context might better
exemplify items that the Romans lost or threw down the
toilet, rather than activities taking place in baths.
Coins, ceramics and occasionally small finds have
been recovered in furnaces and hypocausts, begging
the question of whether bathhouse rubbish was thrown
into furnaces. Since hypocausts were often infilled with
rubbish during demolition or renovations, and sometimes
flooding introduced sediments and materials into hypocaust
systems (Philp 2012, 67), careful attention must be paid
to stratigraphy in hypocaust layers. Ceramics and coins
have been recovered in ash layers thought to reflect the
use periods of hypocausts and stokeholes at the Bewcastle
(UK) and Dover Shore Fort baths (Gillam et al. 1993, 41;
Philp 2012, 59). In the Late Roman baths at Corinth (GR),
fragments of glass cups, a broken bronze finger ring, a
marble mortarium, ceramics and a coin were recovered
in ash attributed to the use of the tepidarium’s hypocaust,
though these layers also contained building materials
(Sanders 1999, 460–461), suggesting perhaps that there is
some contamination from the demolition layer above.
While the vast majority of hypocaust and stokehole finds
can likely be attributed to infill or destruction layers, it
remains to be explained how objects might enter a stokehole
or hypocaust and become part of an ash layer during normal
use. Broken ceramics and other finds certainly wouldn’t be
suitable as fuel, but perhaps they entered stokeholes with the
wood, chaff and/or charcoal that was used to heat hypocausts
(Nielsen 1993, 19–20; McParland et al. 2009). The long term
disposal of solid debris, such as ceramic sherds or building
materials, in hypocausts would not be practical, since they
11. Cloth working in the baths? Site formation processes, needles and spindle whorls in Roman
would simply accumulate, requiring that the hypocaust be
swept out more often and the rubbish disposed of elsewhere.
The majority of ash layers, in fact, contain little material
culture, suggesting that hypocausts were not regularly used
for the burning or disposal of rubbish. It is possible that
hypocausts simply may have served as a convenient garbage
for bath attendants on occasion, though it seems unlikely
that these objects necessarily represent the possessions or
on site eating and drinking activities of these attendants,
as the heat of the hypocaust furnace would not encourage
workers to linger unnecessarily.
Rubbish produced in the baths may also have been
disposed of in pits nearby. At Wroxeter (UK), excavations in
the portico, courtyard and precinct near the baths recovered
several pits containing jewellery, bath flasks, broken glass
and ceramics and other small finds which may represent
garbage from the baths. It is equally plausible, however, that
these pits contain rubbish from the nearby market, houses
or other buildings (Cool 2000, 162, 185; Mould 2000,
138), and it is impossible to discern to which structures this
garbage belongs. While pits may have provided a means for
bathhouse rubbish disposal, the urban setting of most baths
makes it difficult to associate pits and rubbish exclusively
with bathhouses.
Cloth working items in Roman baths
The artefact data in this paper come from a larger study of
published and unpublished small finds assemblages from
Roman public and military bathhouses, the aim of which
is to reconstruct social life in the baths using these objects
(Whitmore 2013; forthcoming). As discussed above, drains
seem to be the least problematic context for objects which
were actually used by bathers, since the artefacts found
in drain layers definitely entered from within the baths,
likely during the normal functioning of the building, and
this context is less affected by construction or demolition
activities than any other bathhouse space. Thus, I have
concentrated much of my research on sites with excavated
drainage systems. This method of site selection assuredly
has some bias since only a small fraction of bathhouses have
excavated drains with finds assemblages. Large imperial
thermae, for instance, are entirely absent from this study,
since few have excavated drains and most are published
without finds catalogues (cf. Hoss forthcoming). As a result,
the artefacts in my study are best suited to illustrate social
life in specific bathhouses, regions and time periods, and
only tentatively can these data be extrapolated elsewhere.
The majority of cloth working tools in Roman baths were
recovered from drains, destruction or demolition layers and
unstratified or unspecified contexts (Table 11.1). Needles
are by far the most commonly recovered cloth working
artefact, followed by loom weights and spindle whorls.
The majority of the objects from demolition contexts were
133
recovered from generalized destruction layers or infilled
rooms and pools. Most of the unstratified or unspecified
finds can only be traced to certain rooms or spaces. In
Pompeii’s Stabian Baths, for instance, spindle whorls,
awls and a sacking needle were found in the palaestra and
nearby rooms, likely those containing hip baths (Fiorelli
1862, 638, 646–647, 650; Libretti d’Inventario). These
finds, however, lack a vertical context and the stratigraphic
details necessary to discern how they might have entered
the archaeological record.
The vast majority of needles, spindle whorls and loom
weights in Table 11.1 are not from contexts that can be
interpreted as bathhouse occupation layers. A few needles
and a loom weight were recovered in the Mirobriga and
Vindolanda bathhouse latrines, but it is impossible to
determine whether these items were used and lost in the
baths. While the description of the silty fill in Mirobriga’s
latrine seems comparable to other bathhouse drain sediments
which reflect normal use, the latrine fill was not excavated
stratigraphically (Biers and Biers 1988b, 179–180), making
it difficult to distinguish which materials may be from the
more promising lower layers. Details on the Vindolanda
latrine drain’s stratigraphy have not yet been published. Both
of these bathhouse latrines, like many others, are also located
at the end of their respective drainage systems (Figs. 11.1
& 11.2). One of the main entrances to the Mirobriga East
Baths is located quite close to the latrine, and the latrine was
entered through the courtyard, rather than directly from a
bathing room. The Vindolanda latrine is detached from the
main bathing block and robbing has obscured the location
of its entrance. While it is probable that this latrine could
have been entered from within the changing room of the
baths (Birley 2001, 7), it is also likely that it could have been
entered from the street. As a result, the objects recovered
in these latrines could have plausibly been introduced
by passersby rather than bathers, and with the currently
available data, it is impossible to determine which latrine
finds, if any, may have originated from the baths.
The only probable bathhouse occupation context that
produced cloth working tools is the drains. In looking closer
at these drainage contexts, however, some finds cannot be
definitively tied to the use of baths. The Cataractonium
(Catterick, UK) needle was found in the rubble fill of a drain
dating to the last period of the baths (Wilson 2002, pt 1, 68;
pt 2, 182), making it a likely demolition context, and four
of the bone needles from Caerleon were recovered directly
under or near the open frigidarium floor drain, a section
which also may have been impacted by demolition activities
(Zienkiewicz 1986b, 173, 201). While the York Sewer
finds (UK) are commonly identified as a bath assemblage
(MacGregor 1976; Cool 2006, 207–210), no sewer segments
have been directly connected with bathhouse spaces.
Furthermore, while the sewer may have drained a bath, it
likely also served other nearby buildings (Whitwell 1976,
134
Table 11.1: Cloth working artefacts in Roman bathhouses, by context.
Unstratified
/ Uncertain
Site
Caerleon Legionary
1st – 3rd c AD
Caerleon, UK
Finds
Needles
Weaving Tablet
Spindle Whorl
Stabian Baths
2nd c BC – 1st c AD
Pompei, Italy
Needle
Spindle Whorls
Loom Weights
Awls
‘Women’s Tool’
1
4
Frauenthermen
1st – 4th c AD
Augst, Switzerland
York Church Street
2nd – 4th c AD
York, UK
Baths of the Swimmers
1st – 3rd c AD
Ostia, Italy
Needles
5
Cataractonium Baths
Late 1st – 4th c AD
Catterick, UK
Needles
Spindle Whorls
Awls
Republican Baths
2nd – 1st c BC
Pompei, Italy
Vindolanda
3rd – 4th c AD
Bardon Mill, UK
Wroxeter
1st – 4th c AD
Shrewsbury, UK
Loom Weights
Contexts
Near Baths
Topsoil /
Post-Baths
Destruction
/ Demolition
2
Construction
Pits in
Courtyards
Latrine
Drains
13
2
1
Total
18
4
4
‘many’
3
6
3
17+
1
1
16
Needles
12
3
1
2
2
1
1
10
1
5
2
1
1
10
7
7
Needles
Spindle Whorls
1
1
2
1
5
Needles
Weaving Tablet
Spindle Whorl
2
1
1
1
5
Mirobriga East
2nd – 3rd c AD
Santiago do Cacém,
Portugal
Needles
Loom Weights
1
2
1
4
(Continued)
Alissa M. Whitmore
12
Needles
Weaving Tablet
Shuttle
Table 11.1: Cloth working artefacts in Roman bathhouses, by context. (Continued)
Total
Finds
Needle
Unstratified
/ Uncertain
Contexts
Near Baths
Topsoil /
Post-Baths
1
Destruction
/ Demolition
Construction
Pits in
Courtyards
Latrine
Drains
Total
1
Needles
2
2
Needles
1
1
2
Spindle Whorls
1
1
2
Loom Weight
1
1
Needle
1
1
Needle
1
1
Needle
1
1
Needle
1
Needles
Weaving Tablets
Spindle Whorls
Loom Weights
Shuttle
Other
11
5
3
19
4
1
3
8
5
7
5
2
11
2
12
4
24
11
1
2
2+
1
15+
2
3
1
28
2
1
1
4
32
1
71
4
16
15+
1
9
116
11. Cloth working in the baths? Site formation processes, needles and spindle whorls in Roman
Site
Mirobriga West
2nd – 3rd ? c AD
Santiago do Cacém,
Portugal
Caerleon Castle
?
Caerleon, UK
Ribchester
2nd – 3rd c AD
Ribchester, UK
Dover Shore Fort
2nd – 4th c AD
Dover, UK
Herculaneum Forum
1st c BC – 1st c AD
Ercolano, Italy
Carsulae Baths
3rd c BC – 4th c AD
San Gemini, Italy
Abbey Villa Baths
1st – 3rd c AD
Minster-in-Thanet, UK
Cramond Baths
2nd – 3rd c AD
Cramond, UK
Bewcastle Baths
2nd – 4th c AD
Bewcastle, UK
Subtotal
135
136
Alissa M. Whitmore
Fig. 11.1: Mirobriga east baths. Note location of latrine (La) at bottom center. A: apodyterium, F: frigidarium, C: caldarium, T: tepidarium,
S: service area/praefurnium, Co: courtyard, e: entrance (after Biers 1988, fig. 75. image: William R. Biers and Jane C. Biers)
Fig. 11.2: Vindolanda 3rd–4th century baths. Note
location of latrine (La) at top left. A: apodyterium, F:
frigidarium, C: caldarium, T: tepidarium; S: service
area/praefurnium, L: laconicum (sweat room), Po: porch
(image: Andrew Birley)
11. Cloth working in the baths? Site formation processes, needles and spindle whorls in Roman
24, 32). As a result, we cannot be certain that these finds
only represent activities in the baths. Although the York
sewer artefacts resemble expected finds from bathhouse
drains (MacGregor 1976, 18; Cool and Baxter 2002), it is
possible – given the rarity of well published Roman sewer
assemblages – that Roman drains, whether connected with
baths or not, simply have similar types of material culture.
After eliminating the problematic drain artefacts, ten
needles, two weaving tablets, one spindle whorl and one
loom weight remain from drainage contexts in three baths,
with the vast majority stemming from the Caerleon legionary
bathhouse (Table 11.2 and Figs 11.3–11.5). The majority
of these finds and contexts are strongly associated with the
normal functioning of the baths. Cloth working artefacts
were spread throughout an 18-metre stretch of Caerleon’s
main drain (from contexts N-1 to 4), with an additional
weaving tablet from Inlet 3, a separate but adjacent section
of the drainage system (Figs 11.6 & 11.7) and several
objects (Caerleon Needles #13 & 36, Spindle Whorl #40
and possibly Ribchester Needle #445) were recovered from
the lower layers of their respective drains.
While these 14 artefacts represent the best evidence for
cloth working in Roman public baths, this is only a small
fraction – 12% – of the 116 bathhouse cloth working finds
from Table 1. The vast majority of needles, spindle whorls
and loom weights come from strata that cannot be strongly
related to activities taking place in a functioning bathhouse.
Furthermore, only 3 of the 19 baths in this sample have cloth
working artefacts from use contexts and they typically make
up a very minor percentage – 1–2% – of the total small
finds drain assemblage (Whitmore 2013). The relatively low
occurrence of these artefacts suggests that cloth working
activities in bathhouse spaces were even less common than
first appeared.
Interpreting cloth working tools from bathhouse
drains
Since artefact evidence for cloth working is so rare in the
baths, and this activity is otherwise unattested in bathing
spaces, it is worth considering whether these objects were
actually used for sewing, spinning and weaving. The most
problematic cloth working object found in a bathhouse
drain is the very fragmentary loom weight recovered from
the Herculaneum Forum Baths. While needles and spindle
whorls are tools for small scale, portable cloth working
activities, loom weights require a large, stationary loom.
While it is hard to imagine this activity in the baths,
loom weaving is more efficient in humid environments
(Kemp & Vogelsang-Eastwood 2001, 327). It is possible
that the moist, warm bathhouse air may have facilitated
this activity, which could have been performed by bath
attendants in side rooms or after hours. The context of the
Herculaneum loom weight sheds little light on its use, as
137
it was uncovered not during a stratigraphic excavation, but
a cleaning of the drain, whose entry points are uncertain
but likely include the palaestra, men’s apodyterium and
the latrine (Pagano 1999, 181).
More problematic than its context, however, is that only
one loom weight was found. The Roman warp-weighted
loom most likely had two separate rows of loom weights,
with a number of threads tied to each weight (Wild 1973,
62–64). The total number of weights used on a loom at a
given time would depend upon the diameter of the yarn, the
weight and thickness of the loom weights, and width and
desired thread count of the fabric being woven (Mårtensson
et al. 2009). Archaeological finds of loom weights in rows,
as well as hoards of weights presumably in storage, suggest
that all but the narrowest fabrics would require anywhere
from 30 to 70 weights (Hoffmann 1964, 311–315; WallaceHadrill 1996, 112; Allison 1999, 70). In an experiment using
loom weights based upon Bronze Age examples, one weaver
used as many as 100 light, narrow loom weights (each
weighing 177 g and 2 cm thick) to successfully produce a
1 sq. m fabric (Mårtensson et al. 2009, 394–396).
It seems probable that isolated finds of loom weights,
such as this one from the Herculaneum Forum Baths, might
reflect chance loss or use as simple weights, rather than
on-site weaving (Wild 1970, 63 n.1). Weights may have
been used in the sale of foods, drinks and perfumes in the
baths, though admittedly selling these products in containers
would be easier. Heavier loom weights may have been used
to automatically close bathhouse doors, a practice known
from the more recent 19th century (T. Fischer 2015, pers.
comm.). While loom weights could have been used for other
functions at any time, the fact that warp-weighted looms
were beginning to wane in popularity when all these baths
were in operation (Seneca, Epistulae 90.20; Wild 1970, 67),
might make it even more probable that loom weights had
additional functions.
The weaving tablets are also challenging to interpret.
Only two broken weaving tablets were recovered at
Caerleon, and while this small number is characteristic
for the recovery of these objects at other types of sites,
additional tablets, perhaps as many as 16–150 more, would
have been required to weave a border (Wild 1970, 73;
Knudsen 2010, 153–156). While tablet weaving can be
more mobile than loom weaving, it still requires a fixed
warp of some kind to stabilize the threads, and smaller
tablets, such as those from Caerleon, may have been
used with a warp-weighted loom, rather than individually
(Knudsen 2010, 153). Regardless, while in use, the tablets
would have been attached to threads, likely making them
more difficult to lose. The fact that the broken Caerleon
tablets were found in a drainage section that only received
materials from pool and floor drains may suggest that these
objects were damaged in the baths, perhaps while someone
was weaving a band or border. It is equally plausible,
138
Alissa M. Whitmore
Table 11.2: Cloth working artefacts from probable use contexts in Roman bathhouses.
Description
Context
Date
Source
Caerleon Legionary Baths
Needles
Broken bronze needle
L 62 mm; T 1–1.5 mm
Bronze Obj. #13
Lowest layer of early drain silts
between bath building and drain
curve (1302:C)
AD 75–90
Zienkiewicz
1986b, 173
Bronze shaft of possible sacking needle
L 94 mm; T 0.5–2 mm
Bronze Obj. #116
Upper drain silts between bath
building and drain curve
(DG 4 [-5])
AD 160–230
Zienkiewicz
1986b, 181
Complete bone needle, rectangular eye
L 112 mm; T 1–4.5 mm
Bone Needle #32
Upper drain silts under frigidarium
floor between inlets from the north
central pool and the NW pool (DG
4 [4])
AD 160–230
Zienkiewicz
1986b, 201
Head of bone needle, rectangular eye
L 22 mm; T 4 mm
Bone Needle #35
Upper drain silts between bath
building and drain curve
(DG 4 [-4])
AD 160–230
Zienkiewicz
1986b, 201
Head of bone needle, rectangular eye
L 17 mm; T 4 mm
Bone Needle #36
Middle layer of upper drain
silts, just outside bath building
(DG 4 [1] / 936:1)
AD 160–230
Zienkiewicz
1986b, 201
Complete bone needle with figure 8 eye
L 68 mm; T 1–4 mm
Bone Needle #39
Upper drain silts between bath
building and drain curve
(DG 4 [-4])
AD 160–230
Zienkiewicz
1986b, 201
Complete bone needle with figure 8 eye
L 69 mm; T 0.5–4 mm
Bone Needle #40
Upper drain silts between bath
building and drain curve
(DG 4 [N-1])
AD 160–230
Zienkiewicz
1986b, 201
Broken bone needle with figure 8 eye
L 59 mm; T 4.5–5 mm
Bone Needle #41
Upper drain silts between bath
building and drain curve
(DG 4 [N-8])
AD 160–230
Zienkiewicz
1986b, 201
Head of needle with prob. figure 8 eye
L 19 mm; T 4.5 – 5 mm
Bone Needle #42
Upper drain silts, near drain curve
(DG 4 [N-5])
AD 160–230
Zienkiewicz
1986b, 201
Broken triangular bone weaving plate
T 1 mm
Bone Obj. #12
Fill of Drain Inlet 3 from NE
Apodyterium or Basilica (1182)
AD 160–230
Zienkiewicz
1986b, 207
Near complete triangular bone weaving plate.
L 39 mm; T 1.5 mm
Bone Obj. #13
Upper drain silts between bath
building and drain curve
(DG 4 [N-7])
AD 160–230
Zienkiewicz
1986b, 207
Lower layer of upper drain silts,
near drain’s exit from bath building
(DG 4 [2] / 939:2)
AD 160–230
Zienkiewicz
1986b,
214–215
Clay in unrobbed portion of Drain
157 (687)
2nd–3rd century AD
Ribchester
Inventory
Drain fill near entrance to Men’s
Baths
1st century BC–1st
century AD
Herculaneum
Depot
Weaving plates
Spindle whorl
Stone spindle whorl
D 34 mm; T 5 mm; Perf. D 6.8 mm
Stone Obj. #40
Ribchester Military Baths
Needle
Broken bone needle
Inv. #445
Herculaneum Forum Baths
Loom Weight
Broken loom weight with suspension hole
L 40 mm; W 25 mm; T 20 mm; Hole D 5 mm
n.b. For the Caerleon finds, the bracketed number indicates which main drain segment the object was found in (see Fig. 11.7)
11. Cloth working in the baths? Site formation processes, needles and spindle whorls in Roman
Fig. 11.3: Bone needles from Caerleon frigidarium drain (© National Museum of Wales. Photo: A. Whitmore)
Fig. 11.4: Bone weaving tablets from Caerleon drains (© National Museum of Wales. Photo: A. Whitmore)
139
140
Alissa M. Whitmore
Fig. 11.5: Stone spindle whorl from Caerleon frigidarium drain (© National Museum of Wales. Photo: A. Whitmore)
Fig. 11.6: Location of Caerleon frigidarium drain. Smaller drain segments and inlets are numbered, with1 representing the position of the floor
drain. 7 indicates the position of a a small inlet or manhole of uncertain purpose in the drain roof, which has been obscured by robbing. A:
apodyterium, F: frigidarium, T: tepidarium, B: basilica, e: entrance (after Zienkiewicz 1986a, fig. 70, 196–199. © National Museum of Wales)
11. Cloth working in the baths? Site formation processes, needles and spindle whorls in Roman
141
Fig. 11.7: Segments of Caerleon frigidarium main drain . The drain segments are divided in two ways. The labels A–C divide the early
drain sediments into three large, differently oriented segments. The later drain sediments (Drain Group 4 or DG 4) were excavated in 27
smaller, roughly metre-long segments, labeled n–12. (Zienkiewicz 1986b, fig. 1, 15. © National Museum of Wales)
however, that these tablets could have been broken while
stored in the apodyterium with their owner’s clothing, and
subsequently thrown away or abandoned in the baths. The
limited evidence for this activity makes tablet weaving in
the baths only a possibility.
Needles and spindle whorls may also have had alternative
functions. Needles were used in toilet and medical activities
(Juvenal, Saturae 2.93−9, Celsus, De Medicina 7.7.4, 8),
though it is more probable that metal, especially iron,
needles longer than 150 mm were preferred for suturing
or medical activities (Jackson 1994; 2011, 250). Since the
majority of needles from bathhouse drains are bone, with
the exception of two broken bronze needles from Caerleon
(Table 11.2), it seems improbable that most were used for
medical activities. Janet Stephens has suggested that some
bone needles, especially those longer than 90 mm, may
have been used to create women’s hairstyles (Stephens
2008). While one needle from Caerleon fits this criterion,
and could potentially have been used in hair styling (#32,
112 mm), other complete examples fall short of the required
length (Caerleon #39, 68 mm; #40, 69 mm). Spindle whorls
may also have functioned as items of adornment. In the
Romano-British cemetery at Poundbury, shale spindle
whorls were recovered near and under the heads of three
female skeletons, which may suggest that they were used
as hair ornaments (Farwell & Molleson 1993, 85–86,
100). However, since needles and spindle whorls would
be interpreted as evidence of cloth working if found in
any other context (cf. Crummy 1983, 65; Manning 2011,
85–86), seeking alternative, non-cloth working explanations
for these objects may be overly conservative. These artefacts
can be reasonably interpreted as evidence for sewing,
spinning and perhaps tablet weaving in the Caerleon, and
possibly Ribchester, military baths (Zienkiewicz 1986b, 20;
Allason-Jones 2011, 240).
Sewing, spinning and weaving in Roman military
baths
If we conclude that sewing and spinning occasionally took
place in a few Roman military bathhouses, what additional
information does this provide on life in the baths? One
of the challenges with interpreting drain assemblages as
occupation ‘floors’ is that the majority of drain contexts
cannot be associated with specific bathhouse rooms, thus
drain artefacts reflect only activities occurring somewhere in
the baths. Robbing and modern disturbances have eliminated
direct connections between drains and bathing spaces at
Ribchester, so we can only speculate where materials may
have entered. The Caerleon baths and drainage system,
however, is better preserved. The majority of the cloth
working finds – all of the needles, the spindle whorl and
one of the weaving tablets – were recovered in the main
frigidarium drain (Fig. 11.6). Inlets to this drain include
a floor drain in the frigidarium, the cold plunges in the
frigidarium and presumed floor drains in the north-east
apodyterium and the basilica. A more specific point of origin
can be given to a broken weaving tablet (Caerleon #12),
which was found in a drain segment (Inlet 3) coming from
the north-east apodyterium or basilica. While only part of
the Caerleon baths have been excavated, leaving us without
comparable data from heated rooms, these drain entry points
permit the location of sewing, spinning and perhaps tablet
weaving at Caerleon in unheated bathing spaces, some of
which included pools (the frigidarium) and some without
(the apodyterium and basilica). Notably, Caerleon’s basilica
and frigidarium are both very large rooms, and the pools
take up only a fraction of the total available space in the
frigidarium, suggesting that these spaces would be well
suited, and perhaps designed, to accommodate socialization
and other activities, as well as bathing (DeLaine 1997;
Revell 2007).
142
Alissa M. Whitmore
Who would have been sewing, spinning or tablet weaving
in the Caerleon and Ribchester military baths? Roman
men and women were involved in sewing and weaving
(Moeller 1969; Allason-Jones 1999), but spinning was more
strongly associated with women alone (Allison et al. 2005).
While both sites have military associations – the Caerleon
legionary bathhouse was located within the walls of the fort,
while the extramural Ribchester baths were situated in the
surrounding settlement – the presence of women in military
spaces is not unexpected (van Driel Murray 1997; AllasonJones 1999; Allison 2013), and material culture associated
with Roman women, such as beads, hairpins and bracelets,
have been recovered in the drains of several military
bathhouses (Whitmore forthcoming). While the gender of
the individuals sewing and weaving in the baths cannot
be plausibly surmised, the spindle whorl from Caerleon
suggests that at least one woman brought a spindle whorl
into, and perhaps spun in, the baths (Allason-Jones 1989,
179). Beyond gender, however, who were these individuals?
At Caerleon, where the largest and most diverse
assemblage of cloth working artefacts was found, it is
suggested that bathers (Allason-Jones 2011, 240), soldiers
or craftspeople might have been involved with cloth working
activities (Zienkiewicz 1986b, 20). Since needles were
regularly part of a soldier’s equipment (van Driel-Murray
1994, 357; Allason-Jones 1999, 1), it is possible that soldiers
were sewing in military baths. But would bathers – soldiers
or otherwise – have brought their chores into the baths, a
space predominantly associated with leisure? Lower class
bathers – a designation which may fit many soldiers – might
multi-task by mending in the more enjoyable atmosphere
of the baths, but it seems improbable that upper class
individuals were engaged in cloth working activities.
It is possible that these needles, weaving tablets and
spindle whorl provide evidence for the activities of slaves.
In addition to slaves that were owned by the baths, domestic
slaves came as bathers and attendants (Fagan 1999, 199–
206). While some slaves actively assisted their owners
with bathing (Martial 7.35, 11.75; Ammianus Marcellinus
28.4.6–9), others may have been left guarding their clothes
elsewhere in the baths (Fagan 1999, 200; Lucian Hippias
5), and these individuals may have been expected to
complete various tasks, such as spinning wool or mending
the household’s clothing.
While ancient texts focus on the experiences that were
offered to bathers, it is important to recall that there were
labourers, slaves and free, who provided these services. In
addition to those who kept the baths in operation (Nielsen
1993, 125–130), sausage makers, confectioners and other
peddlers advertised food, hair pluckers removed unwanted
body hair, and masseurs offered massages (Seneca, Epistulae
56.1–2). Other commercial services may also have been
offered in the baths, and a darning amenity might have been
especially well suited for this space, since bathers would not
need their clothing during their stay. While Roman soldiers
were entirely capable of mending their own clothes, some
likely had the resources to take advantage of such a service.
Conclusions
It often takes an unusual find in an unexpected place for
scholars to critically evaluate the impact of deposition
practices and site formation on archaeological contexts,
processes that affect every site regardless of how typical
and mundane the assemblages appear. The necessity of
critically appraising the impact of these forces is evidenced
here: very few cloth working tools in Roman baths likely
represent activities that actually occurred in these spaces.
Caerleon, with its ten needles, two weaving tablets and a
spindle whorl from the drain, provides the strongest case for
this practice. The virtual absence of cloth working artefacts
from other bathhouse drain assemblages is interesting, but
difficult to interpret. The Caerleon legionary bathhouse
is one of the largest baths in this sample, and has – by
far – the most plentiful and diverse small finds assemblage
recovered in any bath. This makes comparisons with
other assemblages, especially those from smaller baths,
problematic, as it is difficult to judge the significance of
the absence of artefacts from an already small assemblage.
Given the current evidence, we can only speculate whether
cloth working was a unique occurrence only at Caerleon
and Ribchester, or a more common, but invisible, activity
within other bathhouses and bathing spaces.
Cloth working is absent from ancient texts and artistic
depictions of life in the baths. When comparing these
sources with artefact assemblages, significant areas of
overlap exist, particularly the presence of jewellery, status
displays and eating and drinking. Other activities that are
never mentioned in ancient texts appear either regularly
(gaming) and occasionally (medicine) in the archaeological
record of baths. Conversely, the quintessential Roman
bathing instrument, the strigil, is absent from the vast
majority of bathhouse artefact assemblages, most likely
due to a combination of ancient behaviours and taphonomic
processes (Whitmore forthcoming). If sewing and spinning
did occur in some baths, and was performed by slaves, the
lower classes and women, its absence from ancient texts
wouldn’t be surprising, given the relatively few sources that
offer detailed information on daily life in the baths (Fagan
1999, 7–8). Furthermore, the probable low status of those
sewing and spinning, and the fact that these individuals –
especially women – were engaged in socially acceptable
pursuits, would make these individuals and activities
unlikely to attract the attention of male authors. Though
cloth working doesn’t neatly fit with our conception of
the Roman baths as a leisure space, these sewers, spinners
and tablet weavers may have been craftsmen, servants or
slaves, rather than paying customers, and these finds may
11. Cloth working in the baths? Site formation processes, needles and spindle whorls in Roman
provide material evidence of the countless, largely invisible,
individuals for whom the baths were a place of work.
Acknowledgements
I could not have completed this research without the access
granted to unpublished reports and finds at several of
these sites, and I am especially grateful for the assistance
of Stephen Bull (Lancashire County Museum Service),
Andrew Birley and Barbara Birley (Vindolanda Trust),
Tamara Tännler (Augusta Raurica Research Archives), the
Soprintendenza Speciale per i Beni Archeologici di Napoli
e Pompei, Jane Whitehead (Valdosta University) and Mark
Lewis (National Roman Legion Museum). This research
was funded in part through the Hugh Last and Donald
Atkinson funds, the assistance of which I greatly appreciate,
and I would also like the thank the staff at the Des Moines
Area Community College and Perry Public libraries. My
thoughts on bathhouse contexts have benefited greatly
from discussions with the participants of the Thermae in
Context Congress (Luxembourg 2013), and the current
paper is much improved by comments from Stefanie Hoss,
Barbara Köstner, and Molly Swetnam-Burland. All errors
remain mine alone.
Ancient sources
Ammianus Marcellinus, History, vol. 3. Trans. J. C. Rolfe
(1939). Loeb Classical Library 331. Cambridge MA, Harvard
University Press.
Celsus, De Medicina (On Medicine), vol. 3. Trans. W. G. Spencer
(1938). Loeb Classical Library 336. Cambridge MA, Harvard
University Press.
Cicero, De Divinatione. In On Old Age. On Friendship. On
Divination. Trans. W. A. Falconer (1923). Loeb Classical
Library 154. Cambridge MA, Harvard University Press.
Corpus Inscriptionum Latinarum, vol. 1 2, pars II, fasc. 1:
Inscriptiones Latinae antiquissimae. Ed. A. T. Mommsen & E.
Lommatzsch (1918). Berolini apud G. Reimerum.
Corpus Inscriptionum Latinarum, vol. 2: Inscriptiones Hispaniae
Latinae. Ed. E. W. E. Hübner (1869). Berolini apud Georgium
Reimerum.
Corpus Inscriptionum Latinarum, vol. 4: Inscriptiones parietariae
Pompeianae Herculanenses Stabianae. Ed. C. Zangemeister
and R. Schoene (1871). Berolini apud Georgium Reimerum.
The Digest of Justinian, vol. 4. Trans. A. Watson (1985).
Philadelphia, University of Pennsylvania Press.
Juvenal, Saturae. In Juvenal and Persius. Ed. and Trans. S. M.
Braund (2004). Loeb Classical Library 91. Cambridge MA,
Harvard University Press.
Lucan, Hippias. Trans. A. M. Harmon (1913). Loeb Classical
Library 14. Cambridge MA, Harvard University Press.
Martial, Epigrams, vol. 1 and 2. Ed. & Trans. D. R. Shackleton
Bailey (1993). Loeb Classical Library 94 & 95. Cambridge
MA, Harvard University Press.
Ovid, Ars amatoria (The Art of Love). Trans. James Michie (2002).
New York, Penguin Random House.
143
Pliny, Epistulae (Letters), vol. 2. Trans. Betty Radice (1969). Loeb
Classical Library 59. Cambridge MA, Harvard University Press.
Seneca, Epistulae (Epistles), vols 1–2. Trans. R. M. Gummere
(1919/1920). Loeb Classical Library 75 & 76. Cambridge MA,
Harvard University Press.
Statius, Silvae. Ed. & Trans. D. R. Shackleton Bailey (2015). Loeb
Classical Library 206. Cambridge MA, Harvard University
Press.
Tacitus, Annals, vol. 4. Trans. J. E. Jackson (1937). Loeb Classical
Library 312. Cambridge MA, Harvard University Press.
Vitruvius, De Architectura (Ten Books on Architecture). Trans.
I. D. Rowland (1999). Cambridge MA, Cambridge University
Press.
Bibliography
Allason-Jones, L. (1989) Women in Roman Britain. London,
British Museum.
Allason-Jones, L. (1999) Women and the Roman army in Britain.
In A. Goldsworthy & I. Haynes (eds) The Roman Army
as a Community, 41–51. Journal of Roman Archaeology
Supplementary Series 34. Portsmouth RI, Journal of Roman
Archaeology.
Allason-Jones, L. (2011) Recreation. In L. Allason-Jones (ed.)
Artefacts in Roman Britain: their Purpose and Use, 219–242.
Cambridge, Cambridge University Press.
Allison, P. M. (1992) Artefact assemblages: not the ‘Pompeii
Premise.’ In E. Herring, R. Whitehouse & J. Wilkins (eds)
Papers of the Fourth Conference of Italian Archaeology 3:
New Developments in Italian Archaeology, 49–56. London,
Accordia Research Centre.
Allison, P. M. (1999) Labels for ladles: interpreting the material
culture of Roman households. In P. M. Allison (ed.) The
Archaeology of Household Activities, 57–77. New York,
Routledge.
Allison, P. M. (2004) Pompeian Households: an Analysis of the
Material Culture. Cotsen Institute of Archaeology Monograph
42. Los Angeles, University of California.
Allison, P. M. (2006) Mapping for gender: interpreting artefact
distribution inside 1st and 2nd century AD forts in Roman
Germany. Archaeological Dialogues 13(1), 1–20.
Allison, P. M. (2013) People and Spaces in Roman Military Bases.
Cambridge, Cambridge University Press.
Allison, P. M., Fairbairn, A. S., Ellis, S. J. R. & Blackall, C. W.
(2005). Extracting the social relevance of artefact distribution
in Roman military forts. Internet Archaeology 17. http://dx.doi.
org/10.11141/ia.17.4.
Barker, P., White, R., Pretty, K., Bird, H. & Corbishley, M. (1997)
The Baths Basilica Wroxeter. London, English Heritage.
Becker, T. (2006) Women in Roman forts – lack of knowledge or
a social claim? Archaeological Dialogues 13(1), 36–38.
Biers, W. R. (1988) Mirobriga. British Archaeological Report
S451. Oxford, British Archaeological Reports.
Biers, W. R. & Biers, J. C. (1988a) The baths. In W. R. Biers
(ed.) Mirobriga, 48–117. British Archaeological Report S451.
Oxford, British Archaeological Reports.
Biers, W. R. & Biers, J. C. (1988b) The baths, locus summaries and
catalogue of finds, catalogue of architectural blocks. In W. R.
144
Alissa M. Whitmore
Biers (ed.) Mirobriga, 164–205. British Archaeological Report
S451. Oxford, British Archaeological Reports.
Birley, A. (2001) Vindolanda’s Military Bath Houses.
Northumberland, Vindolanda Trust and Roman Army
Museum.
Camardo, D. (2011) Case study: Ercolano: la ricostruzione dei
sistemi fognari. In G. C. M. Jansen, A. O. Koloski-Ostrow &
E. M. Moormann (eds) Roman Toilets: their Archaeology and
Cultural History, 90–91. Leuven, Peeters.
Carandini, A., Fabbricotti, E., Gasparri, C., Gasparri Tatti, M.,
Giannelli, M., Moriconi, M., Palma, B., Panella, C., Polia, M.
& Ricci, A. (1968) Le terme del nuotatore: scavo dell’ambiente
IV. Rome, De Luca Editore.
Carandini, A., Ricci, A., De Vos, M. & Medri, M. (1982)
Filosofiana: The Villa of Piazza Armerina. The Image of
a Roman Aristocrat at the Time of Constantine. Palermo,
Flaccovio.
Cool, H. E. M. (2000) The Roman vessel glass. In P. Ellis (ed.) The
Roman Baths and Macellum at Wroxeter, 162–185. London:
English Heritage.
Cool, H. E. M. (2006) Eating and Drinking in Roman Britain.
Cambridge, Cambridge University Press.
Cool, H. E. M. & Baxter, M. J. (2002) Exploring Romano-British
finds assemblages. Oxford Journal of Archaeology 21(4),
365–380.
Crummy, N. (1983) The Roman Small Finds from Excavations
in Colchester 1971–9. Colchester, Colchester Archaeological
Trust.
Daniels, C. M. (1959) The Roman bath house at Red House,
Beaufront, near Corbridge. Archaeologia Aeliana 37,
85–176.
DeLaine, J. (1992) New models, old modes: continuity and
change in the design of public baths. In H. J. Schalles & P.
Zanker (eds) Die römische stadt im 2. Jahrhundert n. chr.:
der funktionswandel des öffentlichen raumes, 257–275. Bonn,
Rheinland-Verlag.
DeLaine, J. (1997) The Baths of Caracalla: a Study in the
Design, Construction, and Economics of Large-scale Building
Projects in Imperial Rome. Journal of Roman Archaeology
Supplementary Series 25. Portsmouth RI, Journal of Roman
Archaeology.
Dickey, E. (2012) The Colloquia of the Hermeneumata
Pseudodositheana, vol. 1. Colloquia Monacensia-Einsidlensia,
Leidense-Stephani, and Stephani. Cambridge, Cambridge
University Press.
Dicus, K. (2014) Resurrecting refuse at Pompeii: the use-value of
urban refuse and its implications for interpreting archaeological
assemblages. In H. Platts, J. Pearce, C. Barron, J. Lundock &
J. Yoo (eds) TRAC 2013: Proceedings of the 23rd Theoretical
Roman Archaeology Conference, 65–78. Oxford, Oxbow
Books.
Ellis, P. (2000) The Roman Baths and Macellum at Wroxeter.
London, English Heritage.
Eschebach, H. (1979) Die Stabianer Thermen in Pompeji. Berlin,
Walter DeGruyter.
Eschebach, H. (1982) La documentazione delle terme del foro
a Pompei. In La Regione sotterrata dal Vesuvio: studi e
prospettive, 313–328. Naples, Università degli Studi di Napoli.
Fagan, G. (1999) Bathing in Public in the Roman World. Ann
Arbor, University of Michigan Press.
Farwell, D. E. & Molleson, T. I. (1993) Excavations at Poundbury,
1966–80, vol. 2: the cemeteries. Dorchester, Dorset Natural
History and Archaeological Society.
Fiorelli, I. (1862) Pompeianarum antiquitatum historia. Naples.
Gardner, A. (2007) An Archaeology of Identity: Soldiers and
Society in Late Roman Britain. Walnut Creek, Left Coast Press.
Gillam, J. P., Jobey, I. M. & Welsby, D. A. (1993) The Roman
Bath-house at Bewcastle, Cumbria. Kendal, Cumberland and
Westmorland Archaeological and Antiquarian Society.
Godwin, E. G. (n.d.) Unpublished preliminary report following
1979 excavation of Ribchester baths. Preston, Lancashire
Museum Service.
Gregory, R. (2010) Ribchester Roman Bathhouse: Desk-based
research and archive assessment. Unpublished Site Assessment.
Oxford Archaeology North and Lancashire Museums.
Hodgson, N. (2014) The accommodation of soldiers’ wives in
Roman fort barracks – on Hadrian’s Wall and beyond. In
R. Collins & F. McIntosh (eds) Life in the Limes: Studies of
the People and Objects of the Roman Frontiers, 18–28. Oxford,
Oxbow Books.
Hoffmann, M. (1964) The Warp-Weighted Loom. Norway,
Universitetsforlaget.
Holmes, N. (2003) Excavation of Roman Sites at Cramond,
Edinburgh. Edinburgh, Society of Antiquaries of Scotland.
Hope, W. H. S. J. & Fox, G. E. (1905) Excavations on the site
of the Roman city at Silchester, Hants, in 1903 and 1904.
Archaeologia 59, 333–370.
Hoss, S. (forthcoming) Small finds in Roman thermae: an
introduction. In A. Binsfeld, S. Hoss, & H. Pösche (eds)
Thermae in context: Roman bathhouses in the town and in
daily life. Archaeologia Mosellana 10. Luxembourg, Musée
national d’histoire et d’art.
Jackson, R. (1994) Medical instruments in the ‘Antiquarium’ at
Pompeii. In L. J. Bliquez, Roman Surgical Instruments and
Other Minor Objects in the National Archaeological Museum
of Naples, 189–207. Mainz, Philipp von Zabern.
Jackson, R. (2011) Medicine and hygiene. In L. Allason-Jones (ed.)
Artefacts in Roman Britain: Their Purpose and Use, 243–268.
Cambridge, Cambridge University Press.
Jansen, G. C. M. (2000) Systems for the Disposal of Waste
and Excreta in Roman Cities. The Situation in Pompeii,
Herculaneum, and Ostia. In X. D. Raventós & J. A. Remolà
(eds) Sordes urbis: la eliminación de residuos en la ciudad
Romana, 37–49. Rome, L’Erma di Bretschneider.
Keller, D. (2005) Social and economic aspects of glass recycling.
In J. Bruhn, B. Croxford, & D. Grigoropoulos (eds) TRAC
2004: Proceedings of the 14th Theoretical Roman Archaeology
Conference, 65–78. Oxford, Oxbow Books.
Kemp, B. J. and Vogelsang-Eastwood, G. (2001) The Ancient
Textile Industry at Amarna. London, Egypt Exploration Society.
Koloski-Ostrow, A. O. (2008) The city baths of Pompeii and
Herculaneum. In J. J. Dobbins & P. W. Foss (eds) World of
Pompeii, 224–256. New York, Routledge.
Koloski-Ostrow, A. O. (2015) The Archaeology of Sanitation in
Roman Italy: Toilets, Sewers, and Water Systems. Chapel Hill,
University of North Carolina Press.
11. Cloth working in the baths? Site formation processes, needles and spindle whorls in Roman
Knudsen, L. R. (2010) Tiny weaving tablets, rectangular weaving
tablets. In E. B. A. Strand, M. Gleba, U. Mannering, C.
Munkholt, & M. Ringgard (eds) NESAT X: The North European
Symposium for Archaeological Textiles, 150–156. Oxford,
Oxbow Books.
Lee, J. E. (1850) Description of a Roman Building and Other
Remains Lately Discovered at Caerleon. London, J. R. Smith.
Liebeschuetz, W. (2000) Rubbish disposal in Greek and Roman
cities. In X. D. Raventós & J. A. Remolà (eds) Sordes urbis: la
eliminación de residuos en la ciudad Romana, 51–61. Rome,
L’Erma di Bretschneider.
MacGregor, A. (1976) Finds from a Roman Sewer System and an
Adjacent Building in Church Street. Archaeology of York 17/1.
York, Council for British Archaeology.
McParland, L. C., Hazell, Z., Campbell, G., Collinson, M. E. &
Scott, A. C. (2009) How the Romans got themselves into hot
water: temperatures and fuel types used in firing a hypocaust.
Environmental Archaeology 14(2), 176–183.
Maiuri, A. (1931) Pompei: pozzi e condotture d’acqua nell’antica
città. Scoperta di un antico pozzo presso “Porta Vesuvio.”
Notizie degli Scavi di Antichità Ser. 6, Vol. 7, 546–576.
Maiuri, A. (1950) Scoperta di un edificio termale nella Regio
VIII, Insula 5, nr. 36. Notizie degli Scavi di Antichità, Ser. 8,
Vol. 4, 116–136.
Manning, W. H. (2011) Industry. In L. Allason-Jones (ed.) Artefacts
in Roman Britain: Their Purpose and Use, 68–89. Cambridge,
Cambridge University Press.
Mårtensson, L., Nosch, M.-L. and Andersson Strand, E. (2009)
Shape of things: understanding a loom weight. Oxford Journal
of Archaeology 28(4), 373–398.
Moeller, W. O. (1969) The male weavers at Pompeii. Technology
and Culture 10(4), 561–566.
Mould, Q. (2000) Small finds from the portico pits. In P. Ellis
(ed.) The Roman Baths and Macellum at Wroxeter, 137–141.
London, English Heritage.
Munro, B. (2010) Recycling in late Roman villas in southern
Italy: reappraising hearths and kilns in final occupation phases.
Mouseion Ser. 3, 10(2), 217–242.
Nevett, L. (1999) House and Society in the Ancient Greek world.
Cambridge, Cambridge University Press.
Nielsen, I. (1993) Thermae et balnea: the Architecture and Cultural
History of Roman Public Baths. Aarhus, Aarhus University
Press.
Pagano, M. (1999) Ufficio Scavi di Ercolano. Rivista di Studi
Pompeiani 8 – 1997, 180–183.
Parslow, C. (2000). The hydraulic system in the balneum venerium
et nongentum of the Praedia Iuliae Felicis in Pompeii. In G.
C. M. Jansen (ed.) Cura Aquarum in Sicilia, 201–214. Leiden,
Stichting BABESCH.
Peña, J. T. (2007) Roman Pottery in the Archaeological Record.
New York, Cambridge University Press.
Perkins, D. R. J. (2004) The Roman villa at Minster-in-Thanet.
Part 1: introduction and report on the bath-house. Archaeologia
Cantiana 124, 25–49.
Philp, B. (2012) Discovery and Excavation of the Roman Shorefort at Dover, Kent. Dover, Kent Archaeological Rescue Unit.
Proctor, J. (2012) Faverdale, Darlington: Excavations at a Major
Settlement in the Northern Frontier Zone of Roman Britain.
London, Pre-Construct Archaeology Monograph 15.
145
Revell, L. (2007) Military bath-houses in Britain: a comment.
Britannia 38, 230–237.
Revell, L. (2009) Roman Imperialism and Local Identities.
Cambridge, Cambridge University Press.
Rook, T. (1992) Roman Baths in Britain. Princes Risboro, Shire.
Sanders, G. D. R. (1999) A Late Roman bath at Corinth: excavations
in the Panayia Field, 1995–1996. Hesperia 68(4), 441–480.
Schiffer, M. (1985) Is there a “Pompeii Premise” in archaeology?
Journal of Anthropological Research 41(1), 18–41.
Shelton, K. J. (1981) The Esquiline Treasure. London, British
Museum.
Stephens, J. (2008) Ancient Roman hairdressing: on (hair)pins and
needles. Journal of Roman Archaeology 21, 111–132.
Thüry, G. E. (2001) Müll und marmorsäulen: siedlungshygiene
in der Römischen antike. Mainz am Rhein, Philipp von
Zabern.
van Driel Murray, C. (1994) A question of gender in a military
context. Helinium 34(2), 342–362.
Van Vaerenbergh, J. (2011a). Flush water for toilets in and near
the baths. In G. C. M. Jansen, A. O. Koloski-Ostrow & E.
M. Moormann (eds) Roman Toilets: their Archaeology and
Cultural History, 78–86. Leuven, Peeters.
Van Vaerenbergh, J. (2011b). Location of toilets within baths. In
G. C. M. Jansen, A. O. Koloski-Ostrow & E. M. Moormann
(eds) Roman Toilets: their Archaeology and Cultural History,
115–119. Leuven, Peeters.
Wallace-Hadrill, A. (1996) Engendering the Roman house. In D.
E. E. Kleiner & S. B. Matheson (eds) I Claudia: Women in
Ancient Rome, 104–115. Austin, University of Texas Press.
Wallace-Hadrill, A. (2011) Herculaneum: Past and Future.
London, Frances Lincoln.
Wardle, A. (2008) Bene lava: bathing in Roman London. In J.
Clark, J. Cotton, J. Hall, R. Sherris & H. Swain (eds) Londinium
and Beyond, 201–211. York, Council for British Archaeology.
Whitehead, J. K. (2006) Excavation of the baths at Carsulae 2006.
Etruscan News 7, 9–10.
Whitmore, A. (2013) Small Finds and the Social Environment of
Roman Public Baths. Unpublished doctoral thesis, University
of Iowa.
Whitmore, A. (forthcoming) Artefact assemblages from Roman
baths: expected, typical and rare finds. In A. Binsfeld, S. Hoss,
& H. Pösche (eds) Thermae in context: Roman bathhouses
in the town and in daily life. Archaeologia Mosellana 10.
Luxembourg, Musée national d’histoire et d’art.
Whitwell, J. B. (1976) The Church Street Sewer and an Adjacent
Building. Archaeology of York 3/1. York, Council for British
Archaeology.
Wild, J. P. (1970) Textile Manufacture in the Northern Roman
Provinces. Cambridge, Cambridge University Press.
Wilson, P. R. (2002) Cataractonium: Roman Catterick and its
Hinterland. York, Council for British Archaeology.
Yegül, F. (2010) Bathing in the Roman World. New York,
Cambridge University Press.
Zienkiewicz, J. D. (1986a) The Legionary Fortress Baths at
Caerleon: the Buildings. Cardiff, National Museum of Wales
and Welsh Historic Monuments.
Zienkiewicz, J. D. (1986b) The Legionary Fortress Baths at
Caerleon: the Finds. Cardiff, National Museum of Wales and
Welsh Historic Monuments.