Roman Military

Clothin (1)
100 Be -AD 2

Fnt publiIMd In Gr.-t 8rlIAon In 2002 ~ ~ F'I.obllII'IWlg.

Elms Cot.rt. ~ W.,. BotIeot. 0 . _ 0X2 llLP. I...Wt.o ~
Em.iI: ~.com

Dedication
This book III dedlclIled to CtlrlI Hairles and all !he pMt and ~
mlln and 'NDfTWI aI the
StrMt Goard. in recognrtlon aI tta c ~ In Ronwl..-ch lind ~ Wooe the
fo..ndabOn a1me .cldely on 1912.

ermr.e

M .... --'.AQIrllrofoo..,_~
...... _ d ~ ~ .

...... ~ _ .. ~ ~ t n d
. ..

-"~01'

"-Act.l"'""lWId_~..,be~_

.........,-.0I'_.. ..,....... 0I'!lJ..,_--..

~

- - - . - - - . _ _ pl.

_"prIlIr

d"~_""""lI'ocIuIllbe

_1D"~

---

WI Qwwllw!;lugll 'MloIll

Ud.

FOR A CATALOGUE OF ALL BOOI<S I"IJ(llJ5I-D BV

OSPREY MlI..ifAAV AND AVIATION PLEASE OOHTACr

TIM MlIf'o:eling "4anagar, ~ ClQc1 UK. PO Boll 1010

......ogtx>oougn. NoManlt. NNe 2FA. Uniled ~
Email:

lnl~oo

...k

TIM Mar\<Itlrlg ~. Oa;ny lhcI USA.

do Mill P'ul:IlIIiNnO. 72ll "-l*:I A......
Emait

."''''4 ...-.-..

_.oopuw

I sholAd lib 10 IhIriI Oft ..--. QwI ~ Tony SegIln

a'ld Martwt 'I"o'hrM: a11h1
StJeM ~. who .. kuld the tme
to COI'I'mliI'lt on .... _ _ of Ihs Iexl. or hIiIJ*:l WIlh
p/wtoo;jI",.. 1n mdiIIon I WlllJd"IO"" MiM BoItIop.
van 0riIIl-~, SimOn.Mrr-., .... ~ ~ ~
Rogers and ~ John , . . . 'Nikllor It.- '*PIA ~ a'lCI
su~ "*"Y of whicto ku'ld Ins WWf InUl IhI fin.- ~ thougt1 a'lY .-rotS wI'kh ,..,..,. _ .mnfy IhI auIhor'a. I .m
grntetulto CIado Ml(lfll..Ul. Masswno Bizzarri. RalfMMI D'AmlIto
and In partiC\AlW 10 PMII Holder, who ku'ld moe! a11he obKuI1I
photogfaphl tInd ,.,~ I .m IIlso gnnefullO !he IollowW'og
Individuals and lns!lIut!or-., who kandIy assisted wtltl!he ~ 01
photographic rtIIIlerIaI: Ff1IIlelll AIifIari, Fnnze. Italy; Alan JetfNy
~.-.cer. TIlIlI8 Walldnland DeArIl Welsby, The BfIIish M~m:
Dr Ur&U1II H&lmburg and Anita Rieche. RheinischM ~ ,

em..

c.a

o.gn,; AIIrl ~
....... b1H1fl~
~ bJ ....... ~ lbcIlridgoa. UK.
~

Acknowledgements

j)' lQ., -*'II oroew-,

.~

o.o.c.. WI ~. USA

Bconn; Andrew Ball. Ml,IHUm of London: Malcolm Chapman. The

Manchester MLlIeUrn: Char\Q Kline and Or JeMlfer H~ 'NegIw.
University 01 Pennsytvlll'lla M~: Robin and Pat Bn~. The
Vlndolandll Muaeurn. Anally I would like 10 thank Andy 8odllIy.
Eliline Ncwt:lU'y and Martin WIndrow. who at helped to mak,elNe
PI'l:Jtecl~.

?, ., . . -

Artist'. Note
Re.dets m.y ~ to not, '"-' the cwlgInel ~ 110m whicto !he
colour plel-. In Ihs l:IOl;*. - . prepa'tld _ ......... tor priYate
sM.M~~~I$""'bylhe
~ M ~ IhouId be ldI
cd to:

The PubIehets NgI'tIt '"-' they c.n

upon_man.

__ into no ~

OMAN MILITARY CLO HI G (1)

ROM CAESAR TO CO MODUS,
100 Be - AD 200
INTRODUCTION
A oldin- nel'd not befr'm"l'd
;11 h' mone')' belt. (HA.
L: H ,eLI I

em

cmned, hod, aII'd and has omf>thing

e and r, 52)

(,REAl DL\L 0

and equipm
basi

Annalus Daverzus, an auxiliary

soldier with Cohors 11I1
Delmatarum. from his gravestone
found at Bingen, Germany - see
Plate B2. This best preserved
example of the Rhineland
tombstones shows the strikingly
curved drapery of the early
Impertal tunic (and also a
legionary-style shield). A
rectangular object seems to be
tucked Into a waistband under
his weapons belt. The fourth
cohort of Dalmatians were later
stationed at Hardknott Fort
guarding the crest of a bleak
pass In Cumbria, north west
England.

if he is clothed,

nL 1I

d b

TUNICS (TUNICA}
Ther i very liLtle evid ne
dLUing the R pu Ii , and
lit raJ ' and arti ti vid n
f Domitiu' Ah n arb I

for the typ

tunic u cd b the Roman arm
recon'UU tion relic' n a ombination of
. W lI-kn wn' ulpLur in luding til
t r
and th
miliu Paullus III num nt, whil

Late Republican/early
Imperial tunics on
bronze statuettes
from Italy:

IA) Telamon,
IB) Villa Glulla, Rome.
Both wear short,
short-sle v d
tunics.

ur and
h R man

Late RepublicanJearly Imperial

tunIcs worn by soldiers from the
time of Julius Caesar.

IA) Mlnutlus, a centurion with

Leglo III Martla from Padua, Italy;
(8) a comlcen from Osuna,
Seville, Spllin.

I~

1_ _

Tunic dimensions, to constant scale: (Al Military tunic

description from BGU 1564 - 1.55m long 111.40m wide.
(B) Child's linen shirt, Nahal Hever, Israel, no.65762 (after
Vadin) - O.38m long II O.45m wide. (C) Young adult's tunic
Nahal Hever no.6-210 (after Vadln) - O.65m long II O.gOm
wide. (0) Mons Claudlanus, Egypt, tunic A (after Mannering)
- O.80m long II 1.07m wide. (E) Nahal Hever, nO.2284 (after
Yadln) - 1.0m long II 1.15m wide. (F) NUbia, Egypt, Grave
Q150 (after Thurman & Wllllams)- 1.27m long 111.40m wide.

earli r peri d
Part of a rare glass vessel from
Vlndolanda, England, showing a
gladiatorial scene. The referee In
the centre wears a typical broad,
sleeveless Roman tunic In a buff
colour, with reddish-brown clavI,
the two vertical stripes from
shOUlder to lower edge; the te""
came from clavus, 'naif.
(Vlndolanda Museum)

a
till
l ,

to

to b 3Y1 ubits I ng

Ofl

luni
dging

ABOVE A relief from Metz,

France, pOSlllbly showing a
merchant and his customer
examining a tunic. Although this
Is not a military tunic, and Is
probably of a type known today
as a 'GalHc coat', It gives an
Impression of the wIdth of
contemporary Roman garmelfls.

Philadelphia
d d alt r-

11

ABOVE LEFT Late Republican

ma, d

and early Imperial tunIcs soldiers from the arch at Susa,

Italy, dating from the reign of
Augustus. tAl appears to be
wearing a '/orics segmentata',
perhaps the earliest representation of this type of armour.
(B) seems to wearing only a tunic
with a fabric waistband.

Early Imperial tunics showing

drapery effects on tombstones
soldiers from the Rhineland area
of Germany:
(A) Flrmus, of the Cohors
Raetorum, from Andernach, and
(Bl unidentified, from Bonn both wearing paenu/a cloaks;
(C) unidentified, from Bingen.

pa for th ir equiprn nt, and d

clothing w
p r ent ofa oldi r' pa b
a it
pain
c ntury

urn nts Uk

'miliLal

od supply of clothing

OPPOSITE late Republican and

early Imperial tunics In details of
fighting figures from the Arch of
Orange, dating to the reign of
Augustus (31 BC-AD 14). The
soldiers' equipment Is easily
identifiable from other sources,
so It Is InterestIng that none of
them appear to wear any armour.
Just visible on (S) Is a knot
behind the neck, gathering the
neck opening together.

mad fr m wo I,
than on wi th at

him an evening me I l \ hich h mi ht invit as man of hi fri nd as h

lik d. An officer hould receive fift), dra 'hmas a da and twO 'uit'l of
lothe on to w ar a home and one to w ar wb n h W TIt out'.
cording to Mauriki
(tmt. 1.2) >Idi rs in th eastern R man
arm of th 7th ntury w re tunic mad li'om ith r r ugh w I, lin n
I' v n goa . hair. Lin n
p iaily w uld b
pr fi l-abl altem bV in
h tter Iimal and als ft r m r Ii rmal
ion . Th r ar t I
two r fi rence t p ci I 'dina r luni (vestis r tunica cenalar7tL), n
in a I tl r from indol nda (1: b. indo!. n 196), th
ther a dinner
co tume giv n b th Emperor Alexand r
vem (
222-23r;) to
aximinu lh
oun r r ord d in th
criptol'e Hi lmiap Auguslae
I ,30).
xtant pap ru d umen fr m Egypt I' V al that bl ach d or
natural- olour illl n was favoured [or 'umm I' garm n whil w I I th
was worn in wint r. An in ncory f clothe b longin to Z non, a
mana r for Ap lion it ,finan mini l r to King Ptol m n in .257 BC,
gi
a 0 d indi alion f th wardrob s whi h oliciaJ and oldiers
mi I1t own. Zenon p s, ssed 13 Ilmi (1:\ 0 of th m with leev ), nin
cloaks and ft ur pairs of cks.
I

nwry Rhin land

mad [rom a fine

an.

OPPOSITE Relief from Nickenlch,

Germany; the figure may
represent either Hercules, or a
slave-trader. However, thh; Is one
of the best representations of
the draped tunic and fabric
waistband. (Rheinisches
Landesmuseum, Bonn)

Early Imperial tunics on

tombstones from Malnz,
Germany.
(AI Publlus Flavolelus Con:lus,
Legio Xliii; his dress, including
the sagum cloak, Is almost
Identical to that worn by
auxiliaries like Daverzus (see
page 31, and he too seems to
wear a waistband with a
document or purse tucked
Into it.
IBI Unidentified - and
It is impossible to state with
certainty, just from the soldier's
dress, whether he is a legionary
or an auxiliary.

~Illralis - was worn to flatten the bunching around tlle waist, thus
concealing how the drapes had been produced. These i1ll.erpretations
are purely speculative, however, and any interpretations from sculptural
evidence alone are always fraught with difficulties. Unless further
e\~dcncc is forthcoming tlle deliberation over the correct design of these
tlillies may never be resolved.
It is not known LO what extent lhe Rhineland type of ulIlic was wom
elsewhere ,,,itllin tlle empire but, to judge from the available evidence,
this style \\IOIS only popular from around the beginning to tlle lauer half
of lhe lSI cemury, when il disappears entirely to be displaced by a
simpler, perh,lps more utilitarian version, This practical form of tunic
dearly e.XiSICd before tllC Rhineland examples, as it can be secn on
the Arch ofOl'ange (Augustus. 31 Be-AD 14), and on early Principalc
tombstones from Caesarca in Algelia. Il is also tlle tunic made familiar
by its depiction on Trnjan's Column (Trajan r.AD 98-117), and the
Chatswonll relief (Hadrian, r.AD 117-138), and is dearly tlle slyle wom
by the Camomile Streel soldic.' frolll London. The diffcrences between
this lllnic and lhose shown on lhe early Rhineland tombstones arc
readily appm'cnt: il has a su-aighllower hem and very few folds at all.

Decoration -

clavI

An inslalllly recognis.'l.ble fealUre of many survl\'mg tLInics from the

Roman period are the lWO contrasting bands of colour known a.'i clavi
which run from the shouldcl1i to lhe IXHlom edges. Altllough senatorial
and c<luesuian officers would be entitled 10 purple clavi on their tunics, it
is not known for cel1ain if ordinary soldiers during lhe early Ptincipatc
were allowed lO display them, The use of purple davi by soldiers from tllC
3n:1 celllury onwards. howcver, is attested bolh in thc Historill AlIg'llstM
(SHA, ClamlillS, XVlI, 6) and on wall paintings from Dura Europos,

RIGHT Early lmpernol tunk: on

the relief of lin IIU~llary 'rom
Cae.....a - modem Cherchell,
Algeria - dating from the eariy
1.t century AD. The tunic doe.
not display the characterl.tlc. of
the 'draped' Rhineland eumple.,
.nd I, much plainer.
FAR RIGHT Trelan" Column In

Rome, celebrating that empero....

Daclan ware at the end or the 1.t
century AD, ha. been highly
Influential Ifl term. of modem
perception. of Roman mHltary
d..... TIl.. I. particularly so In
the ca.. of Ea,tem archers, who
.....lmo.t Inv.rnobly depicted
today they are app.rently
.hown on the Column, with Iont
flowing robe.

Fuemes beli~'ed lhal some musicians in a gladiatorial scene depicted

on a mosaic from Zlilen in North Africa mar represent members of a
military band; these men wear military style yellow-brown cloaks and
have short while wnics wilh black dfWi. (llowever. the presence of a
female member of the band playing a w',tlcr-pow('red organ may cast
some doubt 011 the pl'csumed milillll)' stalliS of the others.) A couple of
the portrait paimings from Egypl - discussed below - may also show
COllllllon soldiers, who have tllnics which al'e decorated with either red
or black clavi.
TIle rklvi were wo\'cn ilHo the fabrics during manufacture. At Mons
Claudianus, Egypt. the majOlil)' of clamonlexLile fragments were between
Inn and 4cm in width. bllllhcre were al~o a Illll'nber of fragmenLS lhal
had IwO and three stripes. The dUll; themselves are someUlllt::s decorated
and some of these examples can be around 7crn wide.
The dropped shoulder and 'bunched' neck

On Trajan's Colullln some scenes seem to illusl.nlle !.hat legionaries

involved in conSlnLCUOI1 dUl.ies wore a special mnic which ",,'as let
do\,'n from the right shoulder, A similar appeamllce is found in
civilian sources, e.g. representations of blacksmiths. However, modern
re-enactors expericllce liuJc difficul}' in working in the basic tunic; and
it should ,llso be remembered that soldiers wcre expected to work in full
annou,. if neccssal)', so should have been accustomed to any minor
inCOllV('nience causcd by the wnic. (Might there have been a Roman
equivalellt orthe 'Pioneer' Corps' who adopted lhis fashion? The literary
sources seem to contradict the idea.) It seems that under normal
circumstances the wide neck opening of the tunic \'~d.S g<uhered and
fastened together in a small bunch of sUI'Plus malCrial at the rear of the
neck. This feature is noticeable in a number of scenes on Trajan's
Column, and on the mid to late 2nd celllury Belvedere Sarcophagus in
the Vatican, Rome, where the detail sun~ves dead)' enough to strongly
suggest tlmt the bunch was tied with a thong nnher than pinned
together b)' jilmltl~ (brooches). An almost identical arrangement can still
be obsen'erl l<>day wom by nomadic tribesmen in Mallretania.
There arc two apparent disadvantages to wearing a tllnic tied into a
bunch at the back: it might seem to require assistance in holding and

All the. . tombstone Images of

arche.. show them wearing the
same type of c:lothlng lUI wom by
other .aldie..: (A) Pantera, of
Coho.. t Segltterlorum, from
BlngerbrGc:k, Germany;
(B) Monlmus, of Coho.. I
ltul1Ieorum, from Tehtbaeh
near Mal~ Oermany; Ie) 'rom
HouNS:lead., England. (D) 15 an
l~ul1lr archer from the Column
01 Marcu. Aurellua, mldtate 2nd

cenhJry, Rome.

ABOVE SOldle.... from TnJan..

Column, wearing the type of
tunle whleh eould be let doWn
from the right shoulder Or tied up
behind the nec:k In a knot. The
loose nature at the garment Is
dearly evkient.
ABOVE RIGHT A depletion of
Vulcan, blackamlth of the goda,
from Bonn, Germany. ahowing
the tunle dropped from tM right
shoulder, and another renef from
Yortl, England, shows the aame
etfec:t. As both communities had
aubalan1ial military preaenee,
tM figures eould have been
modelled on blaekamlths working
for the army.

tying, and it is an unnecessary encumbrance

when worn beneath annolll. TIlt' first objcction
disappears as SOOI1 as one \....Ilches a woman
casually tying her hair up behind without benefit
of a mirror. The second seems more serious:
nevenhelcss. the existence of this peculiar fashion
is confirmed by il~ appearance in more than one
source, and when reconstructed U1C bunching at
the rear produces all the types of folds observed in the sculptural
renditions of military tunics.
Alu10ugh there arc many texlile fragmenlS from military sites no
complete examples of positively idelllilied military tunics exi!'il. If thcy
were similar in manufacture and design to surviving tunics from this
period. e.g. those from Nahal Hcver, then another problem presenlS
itself. As thc survi\ing tunics are broader than they are long, this lcaves a
lOt of sU'1,lus material belleaU1 the arms. Indeed, this phenomenon can
be observcd on the ChaISworu1 relief where the soldicrs are shown in
their undress uniform of tunic and belts (see pagc 46). This surplus or
material .....ould appear to be an inconveniencc. especially when the
soldier wore any fonn of body anllour. Modern re-enactors have favoured
.. narTOW tunic, which may therefore be incorrect bill which does atlcast
lit comfortably under the armour. This might seelll to lend support to the
idea that Roman soldiers not onl)' wore a special tllnic when they
weill illlo action. but lhat it W".t.s of a difTerenl design. However, in
contradiction of this, recent experiments carried OUI by members of
the Ermine Street Guard indicate that UIC wide Hmic can be worn
comfortably beneath al1110Ur - indeed, it provides additional padding.
Another feature which can be detected on the Chatsworth relief is
what appears to be a narrow strap decoratcd llIiul studs which passes
under lhe righl arm and crosses over the left shoulder. Fuentes
confusingly termed this a 'pOlich bell' by analogy WiU1 those used by
18th/19th celllury cavalrymen, but suggestcd lhat its function \\",lS to
restrict any blousing of the material "..hich might interfere ....,th the
drawing of the sword. Even when \\caring an undress uniform soldiers

\ r

,till

quipp d with

id arm,

n rail a

auxiliaries
f th
arl Prin ip

nd
m

Relief of an olive harvester from

Cordoba, Spain. He wears the
short-sleeved tu.nic of the type
which could be let down off the
shoulder, identical to those worn
by some soldiers on Trajan's
Column, with the wide neck
opening restricted by a large
knot tied at the back.

MILITARY CLOAKS
from Imperial s ulptllre the common oldiers appear to hay
favour d twO distin t typ of cloaks - the paenula and th agum - whil
ffi r' f lh rank
c mun nand ab v had lh pLi n of waring a
mor formal
1 kn wn as a paludamenlUm. In R publican Lim
at rding
ivy ( I, 4.9), ther wa 1 arl a i ibl diEfi r n
etwe n th loa' w rn b om e and heir m n. H des ribe h w lh
Triblill Publiu D iu and' In f hi centurion wrapp d th m lye
in ommon oldi'r cloaks to di gui their rank when th w n OUlon
reconnaissance.
he prumula wa aIr ad e LabJ.i h
but i origin
may hay b en ev n arli r.
no unrlVlng xampl e. i from a
p iti 1 military nl xt il i nJ P ibl t r on LrU tide ign b
11 jud

Jibed as

, had

Praetorian Guardsmen shown in

sculptures from Italy; all wear
paeflula cloaks. (A) & IB) Arch of
Domitlan, Puteoll; (C) tombstone,
Belglolosoi (D) Cancelleria relief,
Rome.
Auxiliaries from Trajan's Column.
Points of Interest are the fringed
cloak, left, and the shortness of
both tunics, worn just below the
waist. Both men also wear
breeches, probably of wool.

Camomile Street soldier had two different types of fastener on his cloak,
though the sculplOr certainly felt it was worthy of recording for posterity.
Both the sagum and its shorter derivative, the sagululII, I\lcrc simply
rectangular pieces of hea\')' woollen material. The papYnls document
BGU 1564 includes an order for fall I' cloaks cach 6 cubits (2.66111) long
and 4 cubits (1.17m) wide. These measuremcnts can be compared with
a surviving cloak from Nubia which ....'as 1.75rn long x 1.05m wide; one
from Nahal Ilcvcr (noo43) \\'hich '\~dS 2.iOm long x Io4Om wide; and
anOlher example from Thorsbcrg in Schleswig-Holstein, Germany,
which measured 250m long x 1.68m wide. The document records that
the cloaks ShOllld weigh 3Y. minae (1.6kg), and each cost 24 drachmae
(6 denarii).
The cloaks in BGU 1564 werc identical in size to a blankel recorded
in the S<lIne document that W'"<lS destined for t,he hospiull, although the
blankct was heavier and cost 28 drachmae (7 denarii). Cloaks of this si1_e
\\lere therefore quite broad and cOlild easily be wrapped around the
bod)' for extra warmth. A further suggestion of large size is provided by
the legend associated with 81 Martin (cAD 371-397) when he was
senring as a tribllne at Amiens. One bitterly cold day Martin saw a beg~rar
outside the city gates and. mo\'ed b}' his plight, cut his own cloak in tWO
and offered half lo the beggar.
In the lSI CClllury AD the pact r-,'!;:lr1ial (EIJig, VlII, 58) lampooncd it
man called Artemidonls, saying that as he wore thick cloaks he should
havc been named sagmis'. In contrasl there were also liglllcr versions
of the $llgllll/ which cQuld be worn in summer or indoors. TIle cloaks
seen on the Rhincland tombstones have almost as many folds as the
tunics, which might suggest thaI the}' were of a relatively light malerial.
Tht: sU1,.''1wt was fastened at the right shoulder by means of a brooch,
and this is one reason for the large numbers of brooches found on
military sites. The Roman hislOrian Varro (D.L.,
V, 167) claimed that the sagum was of Gallic
origin and had been adopted by the Romans
during the Celtic wars of the 4th celllury BC; but
il is also present in German and Spanish
contexts. The Sllgllll/ I\'as therefore another
example of the Romans borrowing what was
originall)' a barbarian form of dress and making
it their own,
It secms that the sagum was especially populal'
with the military and was worn by all l<lnk.~ from
COlTllllOn soldier to emperor. It is by far the most
common t}'pe of cloak represented on Trajan's
Column. In facl the sagulII became so associated
with the army that the phrase 'putting on the
miliuuy sagum' was analogous with 'going to war'.
It \II<IS because of Ihis militalistic associm,ion that
the pacific Emperor Marcus Aurelius lalcr tried
unsuccessfully to ban its use (SHA, Vl, 27). A
recently translated document discovered in
Carlisle refers lo tcn military cloaks, saga l1Iililan'a,
This is far from a unique reference, as 'militalJ'
cloaks' are also specifically mentioned by Appian

OPPOSITE Tombsto...e of a toldier

lrom Corinth, Greece. He wean
the paenul. fattened, anet partl~
folded up over the thoulden,
giving the characterlttlc 'W'
shape "t the lront. By tlmply
pultlng the tides of the clOlilk up
and flipping them onto hrt
thoulden the aoldler h"d e"t~
accett to hrt tldearmt - which
at the same time could be
proudl~ dltplayed, Here he
carries a fusfis Of" cudgel, and
what seems to be " wrlUng
tablet. Beneath the cloak there
appean to be a tubttantlal
tcarl, though thlt could allIO be
Interpreted at a folded-down
h....

Paen"l" cloak from

Eg~pt, of
uncerUlln d.ate. It Is tn a 1.2 twill
weave; a cUrYed hem hils been
tumed back and f1nlthtKl with
braid. (Unlvenlty of Penntylvanla
Museum, E1&8031

, 100) and -acsar ( ivil War, r. ~ .6). io assius (E.]Jil, 50A) a

iliat eta-vian" tr p' put 11 their milit,H1' cloaks imm dial I
after war had I n d clar d again t I p tra ev n Ihough Ih
w r till in ROland lh en 111 wer n wh r n r, Thi,
o Jallon f particular gann nt with parti ular ireum tan
I' info r
. tJl
id a til t 'oldi rs \ r difli r nl garments at
diffi I' nt tim . S th agwfI W uld b mor pra tical in battl
than th pflPnu[a \ hich auld mpl t I o' r the arm.
it is perhaps not Irpri ing ilia' this
lhe garm fit
referred to
tll 'milita
loak'.
h r IV re oth r unction' or th agul/I apart [I' m
Ih obvi us ne f k ping Ih wearer warm. It \
d ubtl
u ed b
I Ii r
n ampaign
a
gr lind h t r blank t, and a cording t
g tiu
( :.flit., IV.) uld v n b trun LIp n b ttl m nts
wab orb th impactofal'rm '. a itll (Hisl.. ,22)
r late that when rebel 'oldier brok into a
R man amp during th rev It 0 til Bata\~an.
th Roman weI' S) tak n b urpli e that th
had" wrap 1o. around th ir r r arm. t a t
hild-.Taciu'aL writ (I{t. ,23) that
on
an th I' oc a ion
the
Batavian
th m el . had v n r on d to u ing h ir
cloaks a ails.
It' a matter of debat a to wh ther loaks
c uld b worn in battl . Juliu'
a ar
ercainl did, whi h wa. on wa Ihat h was
instantl
gni abl ; and thi ma iJ lpl'
that Ih r Sl f hi' men did n l. urin
hi fli ht fr m th battl for th Pharo
I -andri., .ae ar 10 t hi
I ak, \ hi h wa'
sub eqllenu calTi d off b the Alexandrians as a hi hl
prized u'oph' (Appiall, n. 90).
n raj an ,
olumn
,oldi r ngag d in baltl fight \ ith ut do' .. H w v r.
the R man, Idi r kill d \ h n a unl llllin
lIap d during
th i g of Dura Eur p
2 wel' \ aring cJ a at the
tim " th ir I ak bra
\lrvi d.
ltunn and I wll I'
A number of I ks d pi ted n rajan'
cl arty have at Ie ton fring d edg . It i not certain if !hi was a
ign of 'tatus but it do app ar to e as ociated \ 1m higher Tad
troop' in luding Praerorian. bellifiriarii. aval m n, 'Land rd
b ar r, nd
nior ffi rs u h
u'ibun s.
ther
uJptur
indicat Ula.t m
I ak w r dora d with
I at the b ttom
com
(
p g 47 and Pia
1).
ni r m
nLiLl d t w ar the p
paludam ilium. hi "'-"l W rn drap dol' Lh 1 fL hould rand
partjail ov r th left ann.
u h it wa vid nu more a mark of
rank rath r Ulan a practicaJ campaign garment, and affic r' are quail
like} to be
n \ earing th _mor fun Ii nal sag-um. Th paludame-nlu1/1
appear' t have b en r ctangular lik Ule sagum, although it ha. b n
ugg l d Ii' m ulptural
idellc Lhat th low r
m r ma hav
b n CUL a\Va and r LInd d II.

DYES AND DYEING

h
niqu

LEFT Soldier in a peanuts cloak apparently hooded - and scarf, In

a detail from the Adamkllssl
monument.

d in

nn
on tilll nt

c mm n r d d luff in Lh
madder,
unkn wn at that

RIGHT Detail from Scene LXXXVI

On Trajan's Column, showing
unarmoured soldiers on the
march. This man wears a
paenuta clOak with the hood
visibly thrown back and both
sides thrown up onto his
shoulders.

Paenuta cloak pattem,

1.95m long x 3.1 m wide.
(After H.Granger-Taylor)

<l

IIlg l eh-

an

v rLh 1
of a numb r

ver
bil,

A typical etIvalryman depleted on

TraJan's Column. He wears a
fringed sagum cloak over his
dagged mall shirt, short tunic
and calf-length breeches,

CATALOGUE OF THE EVIDENCE FOR

THE COLOUR OF MILITARY TU ICS
AND CLOAKS
b developm nt of practi al history' roup 'in

g neral int r

incr ased

While Lhere is no doubt lhaL Lhe Romans had Lbe ciipabilit), of

producing all these colours, lhe issue is whelher any or all \,'cre adopted
b}' the arm}'. It should be remembered that it is impossible to specifically
idelllify textile fragmellls as belonging to military nmics, including those
from supposedly military sites. There is a tendency to forget that Lhere
were many civilians presel1L in and al'Ound forts; and, perhaps more
basically, that the Romans had as many uses for textiles Other Lhan for
clothes as we do today.
An anempt has been made here to address this imbalance and to
presem without bias all the evidence that is currently available. By
and large, readers are ultimately left lO dra\\' Lheir own
conclusions as lO how this evidence is best interpreted; but a
sumlllary is included in the forthcoming second part of
this nudy, r-.knal-Arms 390.

The Emperor Marcus ,t,urellus

(r.AD 1111-16O).s commander-lo
chIef, from. stlltl,le now In tn.
B.ltlmonl Art a.llery. He hllS
pIIludamentl,lm f.stened on his
left shoulder, draped around hIs
blick, cllrrled f_lIrd IICntllS

the midriff 110<1 d.-ped ovar the

left IIrm.

(I) Historical fresco, tomb on the Esquiline, Rome;

Republic-cUt period
This fnlgmem of a lomb fresco is probably among lhe olden
surviving examples of Roman paillling, and provides us with
the earliest clues for the colour of Roman military tunics. It
appears 10 illustrate an e\'elll in c.326 BC during the Samnite
wars. Arollnd lhe main celllral group of figures can be seen a
number of warriors in short .....hite tunics and kilts. Becalu dates this
fresco to the lst celllu'l' Be, bllt adds thal it appears to be a copy of
a mllch earlier work. It may depict tlle historical event described but this
is nOl certain.
According to Livy (IX, 40, 2-3 & IX. 40, 9), one division of I,he
Samnite army wore bleached white linen tunics. adopted because the)'
had consecrated themseh'es and ,,,hite was the colour of the priesthood.
Unfortumuely Livy neglected to leave liS a description of contemporary
Roman anny tunics. An Etruscan fresco dating to tlle 4Lh century BC
may provide a clue. The so-called 'Francois' tomb shows a legendary
battle possibly between tlle ElfU.SCans and the Romans. or at least their
Latin allies. A soldier in a bronze Illuscle cuirass and a red llmic is being
killed by a naked opponent. As the defealed warrior is unlikely to be an
Etruscan this could reasonably be seen as evidence for tlle colom' of
RorD<m tunics at tllis period.

(2) Historical fresco, tomb of the Statilii, Rome; Republican period

This may also show Roman soldiers from the early Republic. The details
of tlle costumes, such as the short. \\'hite kilts, are similar to the earlier
Esquilille fresco, so lhis too may be a copy of a mllch earlier work. It is
tllcrcfore possible that Lhis paillljng also represents another legendary
battle between Romans and Samnites, but there is no general agreemem
as to which historical evelll is depicted.
(3) Silius ltalicus, IX, 420; late Republic
The poel Silius ltalicus records thatlhe cloaks for lictOI'S were red while
lhose used by senior commanders were eitller purple or while.
(4) Africa" War, LVII, 54-6; late Republic
Q.C'ledlius Scipio. commander orlhe Pompeian forces in Africa. al....'ays

",ore a pUl'ple cloak. This apparently angered his African ally,

thc NUlllidian king Juba L because lhis \\~,LS his own norma] aHire;
~ubscquelllly Scipio was obliged to dress in white.
(5) Plutarch (c.ADSQ-120), LilH!$, Crassus, 23
On I.he eve of the Hileful baule of Carrhae (53 BC), Crassus mistakenly
",ore a black cloak. Nol surprisingly this was imcrpreted as a b.,d omen,
and Cmssus had 1.0 revert to wearing a purple cloak. Why he should
choose to wear black instead of the general's colour ;s unexplained.

(6) Mosaic, Palestrina, Italy; date lmcertain

The next lWO subjects, the I>alcslrina mosaic and the ~udgcment of
Solomon' fresco, are perhaps the 1110st controversial as lOgetJlcr tJ1CY
fonncd the cornerslOne of Fuentes' hypOlhesis on Roman lunic colours,
in panicula.r for tJ10SC of tJ1C PmelOrian Guard and centllrions.
The Palestrina mosaic (sec lhis page) shows an Egyplian landscape
";lh the Nile in flood and numerous exolic African animals. At the lower
centre a grollp of soldiers are galhered around what is clearly a temple.
Fucntcs refen-ed to the datc of this mosaic as ranging from anywhere
bttwccn the lSI cell1ury BC and the 3rd Centllll' AD, but believed that
the mosaic documented a \;s;t to Egypl by OClav;an (later the Emperor
AugusfllS) aflcr the b.'mle of ACliulll in 31 BC. OCla\"ian's visit coincided
\\;tl1 an abundanl flooding of the Nile which tJ1C mosaic seems 10 record.
~e\'ertJlelcss, other scholars are of the opinion tJlat the mosaic is in fact
A copy of a much carlier work by an Alexandrian school, tJmt no longcr
~ur\'ives (this theory W'<ts most recently st.,led by Meyboom in his detailcd
"llUdy of the mosaic).
Il should be noted that tJu:re are slight discrepancies in tJle details
of uniform colours between the obser\'ations of Fuentes and Mc)"boom,
in particular regarding the figures Fuenles belie\"ed to be tJle
admiral Agrippa and his centurion bodrguard. For example, Fuentcs
described one man (Agrippa) as weal'ing a very pale blue tunic and
moulded cuir-dss. and accompanied b)' another figlll'c dressed in a red,
<t.lee\"eless, knolled wnic
(3 centurion). Me}'boom,
hOWCVCl', desCI"ibing the
same twO figures, refers to a
,,()Idier who wears a white
leather or linen muscle
cuirass over a white rhilQn (a
Greek-style tunic), while
the otllel' is dl'essed ;n a
brownish)'ello\ll c/litOI/.
lllere is some evidence
that lhe colour bluc was,
perhaps not surprisingl)',
.olSS()Ciated Ix)\h ....;th the sea
~nd tJle na\y. llllls Fuentes
bclie\'ed thal the figurc
_eal'ing what he Ihought
",as a blue tunic could well
have been the admiral

From the Palestrina Mosaic,

ttaty (Item 8; In the te...t
CIItalogue) - clOH-UP of the
group of flgu,., frequently
IdenUfled ., PnIelorl.n
Gu.rd,men, ,tanding Delore
.n Egyptl.n temple. A reddl'h
tunic tied .t the b.ck of the
neck Is vl,ible on the central
figure. (Copyright Fnlltelll Alln'rll

Itam 7 In the taxt catalogue:

pygmies dressed es military
figure', from the late Republican
'.Judgemant 01 Solomon' fresco,
Pompeii. 'The armour and
equipment reflect deslgna
commonl)' used throughout
the eastem MCtdltamanean
world In the Hellenistic eta. The
right-hand figure la dressed
entirely In ted clothing, with a
ted helmet crest.

Agrippa. According to
Suelonius (AlIgtl.l"t1IJ". 25)
AgJippa was in fuCI honoured
by Octa"ian. who presented
him witll a blue nag.
The historian Appian
(V, 100) records that in
somewhat similar circumstances Sextus Pompey took
to wearing a blue cloak
because of his victories at
sea. What appears to be tlle
only other reference in
Roman litCf""dturc to blue
military tunics is when Lhe
historian Vegetius (Epi/., IV,
37), writing in the 4th
celllury AD, tells us thallhe
sailors in the British neet
dyed nOI only their tunics
but also the sails of their

ships to match the colour of

the sea.
Onc of the figures in the Patesuina mosaic tllat has been obviously
restored is the charactcr Fuentes idelllilied as Octavian. He appears to
wear a grecn mouldcd linen or leather cuirass - possibly with attached
groin Haps, pteruges - over a white chiton. While moulded lincn or leaLher
cuirasses are not alwd)'S thought of as being Roman, these details do
conespond to othcr Maccdonian tomb paintings from Egypt. Taking this
imo consideration with the other evidcnce, including details of Lhe
soldiers' dress (e.g. Creek-style bool~, unhades, rather Lhan Roman ca1igae,
plus the Ebryptian setting), Me)'boorn concluded Lhat the figures on the
mosaic in fuet represent members ohhe elite troops oftJle Ptolemaic anny,
tJle Macedoncs.
(7) 'The Judgcmcnt of Solomon', frcsco, Pompeii, Italy; latc
Republican/early Imperial
Fuente... argued lhal this court scene enacted by pygmies (see this page)
echoed contemporary dress. Howe\'er. once more il is believed thaI this
fresco is a copy of an earlier Alexandrian work and Lhat tJlerefore tJle
eOlllemporary dress is Ptolemaic rather than Roman. Fuentes asserted
thaI twO of the soldiers wear white lunics, although al first glancc lhe
tunic of Lhe soldier about La cleave Lhe baby in half appears to be light
blue; Sekunda in r.'lCl states thal this lunic is blue. Howevcr, Lhe fresco is
badly damaged in places, including areas around this figure in particular.
Whal appears as bluc may in fact be an attempt to render shading on a
while ganuenl, a lechnique that is evident elsewhere on the fresco.
The second soldier has a salmon-pink cloak. A tllird soldier wears a
dark. red tunic and cloak, which Fuentes believed marked him Olll as an
officer, most likely a centurion. Nevertheless, all three soldiers have
similar rcd helmet crests, which would seem to weaken this tlleory.
Vegetius, writing in the 4th century AD, records Lhat centurions wore

transverse silvered creslS to distinguish thcm in baLLle, but he does nm

melllion them wealing a differelll tunic colour as well. The crest of the
third soldier is certainly not transverse, nor does he show any of the usual
attribUles associated with cellluriollS such as the vine stick or greaves.
However, this soldier's armour and helmet are dilTerelll in colour from
those of his companions, so Fuentes lllay be COITCCI in identifying this
man as all officer. But of which army?
(8) Tavern sign. Pompeii, Italy; late Republican/early Imperial
TIlis wall painting (see this page) shows a soldier being offered wine by
the landlord. Apparently off duty, he wears only a tUllic and cloak but
appears to cany a spear which could well be a weighted fribllTL If so, this
soldier would certainly be a legionary or a Pl'aetolian rather than a local
policeman; he is in fact similar in ovcrall appeanll1Ce to the contemporary
soldiers, probably PraetOlians, on the Ca.nccllelia relief in Rome.
The soldier's cloak is clearly of the pae1l11la style and is a yellow-brown
colour, apparently worn avera grer-green scarf. An alternative colour for
the pOeJwlo may be suggested both by a tombstone \\;th some painted
details surviving, found at Castleford in Yorkshile. and the remains of a
cloak found with a bog body also discovered in Yorkshire in tJ1C 19th
celltul]'. In bolh cases the cloak was green. although il must be stJ'essed
that the identification of eitJler subject as a soldier is doubuul. A more
positive suggeslion is supplied by one of the Vindolanda writing tablclS,
which refers to a white /Jae11111a. In ,'iew of some of the later evidence
from Dura Europos discussed below, this conceivably belonged to ule
commanding officer of the cohort.
(9) Wall painting, House of Valerius Rufus (or of the Trojan Shrine);

Region t, lns.6.4., Pompeii, Italy; late Republican/early Imperial

In the upper registers of a painting are two figures. a male and a female.

The fonncr is almoured and wears a MOlllefortino-st)'le helmet with a

red crest, a ,,,,hite linen cuirass and bronze greaves. He carries a large
0"0'31 shield and a spear in his left hand, and a<ljuslS his cloak with
the other. His tunic is a pinkish-red and tJle cloak is while. AlulOugh
the ligure is probably meall\
10 represent a deity or a
hero he appears lO be
depicted in a contemporary
panoply, which in fact looks
\'el')' similar lO the figure
reg-arded as Mars on the
_\liar of Domitius Allenebarbus. Thi.s is supported by
the inclusion of a Montefortino helmel rather than a
sr:lised Attic type.
(10) Historical fresco,

Pompeii, Italy; late

Republican/early Imperial
One more scene from
Pompeii, which shows a

Details of frelIcos from Pompell

(}tems 8 and 10 In the text
catalogue). Lett, soldlet" In a
'.llowbnI...... pHnllla ..loak
with dar1o: l"IIdbrown clavi and
grey-green scal11s drinking .t
tavem. Although the fTesc:o Is
d.m~ It Is Just poulte thlt
trices of. while tunic .... Y1sltNe
bene.th the hem of the cloak
and st the soldler'1lllett s l _ .
HellenIstic al11.ts tTequenUy did
not paint In white .re.s, the,
simply drew an outlIne on the
unpainted plaster and I.tt It
blink. The right-hand flgu... I'
usually described .s a trlllellef",
but his 'ppe8f1nc:e I, lIery
military. He wears II ,eUowbrown s89Vm cloak over II
whit. tunic.

co Domus Aurea Rome Italy; early

d pic d in a [r 0 rom
p 'ibl repr ents th

in th
reen.

ir naJu "
01 ur n

A wax encaustic portrait from

Egypt (Item 23 In the text
catalogue), which may provide
further evidence that the
wearing of red cloaks was much
more widespread than previously
believed. The man wears a sword
on his right side - the white,
perhaps IVOry pommel with a
bronze terminal Is Just visible;
and this position Is usually an
indication of the common soldier
of the early Imperial period. The
Antonlne hairstyle would appear
to Indicate a date In the latter
part of the 2nd century AD, a
date which supports the beUef
that his sword position is not
that of a centurion. His tunic Is
whl1e, with red clavI, and over
his left shoulder he wears a red
sagum cloak which partially
obscures his sword belt wIth Its
alternate bronze and sliver studs.
(The Manchester Museum, The
University of Manchester)

ugg

Lh

R man oldi

, ore r d
olour r

. r Ii bl
or hi

Declamati01les ill D clamalio 'Pr

ain (
a \ivid
r
n
Idi r'

(15) Textiles, Vindolanda, Britain; late Ist century AD

OUlofa ampl ofnearl 10 l xliI [ragmen from ind landa, 50 weI'
anal d fi r d
and vid ne\
fi und n nl nin f Lh . Man
lh
th r le til w r [unpigm filed
I and Lh re r 'uirabl fi r
d in , bUl had b n t h aviJ r Lain d rr m burial l h VI an o-a e f
d . Of Lh nin l til
anal '" d. i hl w r d ed I' d' rn
th r

Two more wax encaustic

portraits from ElMlt dating
between the reigns of ~an
(AD 98-117) and Antonlnus Pius
(AD 138-1611. Both men are
Identffled 8$ soldiers and wear
white tunics with dark blue cloaks
on th.elr left shoulders. /AI seems
to be a common soldier, since a
sword pommel Is Just visible on his
right side. In contrast, (8) has 8
sword belt crossing his right
shoulde,., so at this date he could
be a centurion. His tunic has black
clavi; he also w ars a gilt wreathsee Plate G2.

appeared to have been a purple suipc. Thc red d)'c used was almoSl
certninly madder, rubia /hlct()l'U/lt, which in the case of Britain would have
had to be imponed. While a repon on the Vindolanda textiles concluded
thaI the picture of Roman soldiers in red tunics may well be correct, we
should remember thaI the textile fragmcnLS are extremely small and may
not be exclusively from tunics at all. A recently translated document from
Vindolanda actually refers to purple and red curtains...
(16) New Testament Gospels, Matthew, 28, 28; 1st century AD
It is perhaps of more than passing interest thaI the cloak the auxiliary
soldiers of theJerusalem ganison used to mock Christ, as documented by
Mauhew, was red. TIle reed stick and the crown of thoms immediately
recall the vine .stick that centurions carried, and the crowns orJeaves that
could be awarded to these officers. When the soldiers nagged Cluisl
could they have been \~cariously getting their own back on someone else?
The otller Gospels all refer to a purple doak. Under the circumstances a
red cloak is far more likely to have been readily available, but il is
ob\ious in what connotation the other Gospel writers regarded the
significance of a pUl'ple cloak.
(17) Martial, Epigrams, XIV, 129; 1st century AD
'When Manial described red canllsian cloaks he said that Rome wore
brown, the Gauls red, but that this was a colour which bo)'s and soldiers
also Iikcd, although it is not dear whether Martial meant all soldiers.
(18) Pliny the Elder (AD 23/4-79), Natllral History, XX, 3;
1st century AD
Pliny records that scarlet dye from the kennes was used for dyeing the
paludimleTllwn, the distinctive officer's cloak. He mentions thaI the most
commonly used red dye. madder, was grown near Rome; but that the
most prestigious red dye was obtained from the kermes (kmllococC1lS
venniliQ), a parasitic insect which infests the kennes oak throughout
southem Europe. He wrongly believed thaI the dre came from the berries
of the tree: it was in fact obtained from the female insects' egg sacks.
While red cloaks are ill\'aliably thought of as belonging to generals e.g. Caesar (Gallic War, VII, 5) - even here we must exercise caution.
When Sextus Pompeius (died c.36 Be) changed the colour of his cloak to
blue as mentioncd above, Appian (died cAD 165) sa}'S lhat he changed it
from the purple thal commandcrs usually wore
100). (h'e"tl!, howe\'cr.
there seems to be some confusion in texts both ancient and modem as to
the poi1ll. on the spectrum at which 'red' becomes purple.

ev,

(19) Tombstone (elL xiii 6277), Mainz, Germany; 1st century AD

A rare example of a tombstone that sUlvived with some of its original
colouring mar possibly indicate that the practice of wearing red cloaks
was marc widespread among ordinary soldiers. Although the paint did
nOl sur...ive for long it was possible to delermine that either Silius, a
ca...alry trooper in the Ala Picentiana, or his attendant (calo), had a red
cloak. Meanwhile the remains of paint all another tombstone, that of
Gnaeus Musius. the aquiliJer standard-bearer with Lcgio XliII Gemina
Martia Victrlx, led the Gennan archaeologists who were creating a
reconstruction of the monument to restore the tunic as white.

amlmrwl on pngr JJ

(20) Pay receipt and textiles, Masada, Isrdcl; 1st century AD

A recclllly excavated receipt sheds Iiglll on military lll11ics, and seems to
con finn lh:u even ordinary soldiers owned more than one. The receipt
belonged to C.~lessius, son of Gaius of the Fabian uibe, recnLitcd from
Beirut. In spile of his local origin il is believed lh;l1 he sen'cd in Lc~rjo X
Fn:lcnsis, possibly as a cavalryman, rmher limn in an auxiliary unit.
Among the lisl of compulsory deductions from his pa}' were 7 clellani for
a linen lllllie, and an unspecified alllount for a while tllnic. Cotton and
Geiger suggested that as linen would have been an expensive item to
purchase. this garment \,'ould !l,lve been kept for special occasions.
However, I'>'C could expecl that in E..-.stcm climates soldiers may have
preferred linen to ,,'oollen garments. For inst;mce, during t.he Persian
wars of the 4th century. Ammianus Marcellinus (XIX, 8.8) describes how
he and a companion tore up their linen clothes to make: a line so that
they could lower a cdp into a welt to obtain water.
A lIulllber of linen and wool fragments were also uncovered during
the excavations al A'!asada: these llluSt, however, be treated "ith great
caUlion as possible e\idence for Roman ann)' tunics. Man}' of the textiles
would undoubtedly have belonged to lheJe\\;sh defenders and civilians
Part of a tombatone from VIenna,
present at the siege. and some of lhe fragmenu; were dearly not from
",,-tria, eommemorating Tltu'
clothing at all. Nearly all of tJle linen textiles were left undyed but it was
R.Iu, Drac:eus. He we
-..lrymen, ..rvln, with Ala I
110t dear how white they would have been originally. It is t.herefore fairly
FWrle Domltla'" Britannlea
safe to suggest that the linen Ilmic mentioned in the Masada pay receipt
"'l~ri. Clvfum Aomeoorum was white. A tme while had probably only been achieved by one textile
thousandstrong British
fragment, which had been made from lamb's wool.
..,.Iment, Mnoured lor outBy wa}' of contrast, more than half of the 105 wool textiles that were
stM1dlng ..rvlc:. during the
German Of' Daelsn eampalgn, 01
analysed had been dyed. Of these 14 were red. which ranged in shade
the Emperor Domlt~n (AD 81-H1
from a salmon-pink to dark maroon, whilc six other examples were
with bloc:k IInlnt 01 Roman
eiulci' plajn blue or blue-green. ContraSting colour bands in dark blue,
dtkenshlp. n.. tombatone h.d
purple or mauve were observed on several fmgmenLS. It \l'"dS noted that
tnc:e, of origInal blue c:olouring
most of the coloured textiles would originally have been wom by
still survlvlnll on Drac:cus' upper
pnnent.
women. and this is supponcd by Jewish literaLUre and the Egyptian
funerary ponraits from U1C Roman period, where men's lunics are
.-nc:r:Z;:.:"!;1;l'!'''''~
almost wilholll exception white. Ilowe\'cr, one textile
frdgmclll. consistlllg of tWO tluck sel\'edges which had
been sewn tOgelilCr and dyed red. \..~.:tS believed to come
from a man's tunic, and with reference to Fuentes'
anic1e il ......AS thought possibly a mililal)' onc. But at
least one Egyptian funeral poru-ait depicts a
woman in il scarlet-red tunic. so once more lhis
evidence is inconclusive. In addition, while
Jewish law forbade men from \\'caring coloured
lunics there is evidence that, like the Romans,
Jewish people did not always follow the rules.
The Babylonian Talmud refers to a Jewish man
who wore a Ilew Roman red tunic while wiulin a
pcriod of mourning.
"~J.~"
We owe .nllld~ of our kno\\:ledg~ Ol~ tJle Roman
'<:.~~: .~.
:.~ifft:..~~'
anny ll"?~n thiS I~nod to the JewIsh hlstonan Josephus.
':'~'7t'~~"~~~~.;f"
who paruclpated Ill. t.he wal~ was captul'ed, and later
'~~~=~;;>~J'J"";'~~:4.t!~~""
collabomted with U1C Romans. In desclibing Ule full of
Jerusalem he relates how one of lhe main Je\\ish leadcl"S. Simon

""',.,

",,,'

.' "

_:..:.

. .....

Bar Cioras, tlied to tunnel out of the cit)' with

somc of his followcrs, but was unable to get past
the Roman lines. It appears that Simon tried unsuccessfully-to bluff his way pa.'it Roman guards
by disguising himself as eit.her a Rornan soldier
or an onicer. for he wore a number of white
tllnic.'i and a reddish purple cloak (8j, 7.26-36).
Interestingly.lhe combination ofa ,I'hite tunic ,\;th
a red or purple cloak will appearscvcral times later.
(2l) Papyri, Egyptj 2nd century AD
Among tJ1C huge collection of ancient documents
discovered in Egypt in a remarkable state of
preservation are at least two pap)'Ii which refer to
the delivery of clothing to provillcial armies.
Together thcy suppOrt the belief that local
communities supplied tJle army: nevertheless. it is
interesting to note that villages in Egypt were
supplying troops as far awa) as .Iudaea and
C'\ppadocia (now part of modern Turkey), There
is, howe-ver, other evidence that the sending or
collecting of military supplies over great distances
was not llnusual. A strength report of c.AD JO,iJ
relating to COhOIOS I Vcterana Hispanorum
Equilata, found in Eb')'pt but probably dating 10 a
period when the unit was based in Macedonia,
acruall)' records tJU\I soldiers had been SClll allLhe
way LO Gaul (France) to obt.'\in clothing.
Fuentes found it surplising that one of these
Eg)'Ptian documents recording the delivery of 19 tunics for the 'guards
and soldiers' scning in Judaea also referred to fi\'e white cloaks. 111is
seemed to contradict the otJler pictorial evidence, which largely
suggested that military cloaks in genernl should be )'ellow-brown.
Howevcr, the papyT1ls documelll BCD 1564 also refers to four white
cloaks, providing the addit.ional term Syrian'. To Ihese can now be added
the white cloak from the Vindolanda document, and IWO while cloaks on
the Fresco from Dura Emopos (to be discllssed in the forthcoming second
\'olume of this study); while anmher papynLs from a slightl)' later daLe in
fact mentions lhe 'comrollers of lunics and white cloaks'.
There is anothl'r point of interest. bearing in mind that Roman
milit'H)' documents arc quite minute in their demil. Thesl' docllments
relate to the suppl), of clothing firsuy to the soldiers in Judaea, and
secondly to Lhe armies in Cappadocia. It is kI101\'I1 that botJl provincial
garrisons at this period comprised both lcgionall' and auxiliary troops.
So il is surprising that the documents do nOI specif}' whether the
clothing iii for legionaries or auxiliaries. It is debatable. IhereFore.
whether an)' attempt was made to differentiate between citizen and noncitizen soldiers b)' tllC wearing of dineremly coloured llmics. Although
Tacitus (/-list., I, 38: III, 47) refelOS on a number of occasions to distinctive
weapons ;uld equipment used by auxiliaries, in contrast he also alludes
lO a policy of Romanising the allied armies by gi\'ing them Roman
armour and Slllndards. If both types of soldier ret:ei\'ed their clothing

a>POSITE Detail of the mld2nd

c.rtury Bridgeness dlstsnce sl.b
....... the Antonln. W.II In
~tland litem 25 In the t.1tt
cNlog....l. The scene shows
.-tary IJIICrifice carried out by
of leglo II Augutltlll, lilt

_be..,

_lcat.d by the lIeJI1//um flag In

... bltCkground. Behind the
-.tnI1 figure, dressed In
. . . - possibly the legion"
_mander Cl.udlus Ch.rall_ four other ...ldl..... On the

IMt 1$ m.n .P98rently we.ring

a -.gum cloak; the ..cond from
. . right w...... a paenula. Red
,.,mt w.s discovered on the
lIocly of the figure at the elttreme
dght, who Is al... possibly
_ring a p.enula.
Apart fTom paenula .nd sagum,
411ther lerms such abolle end
-.:ema .,.. u.ed In ancient
_.relure, bYt the may simply
hrte been e.t.m.tI...e n.m.s for
. . . Arne cloaks rath.r then
_pletely different d.slgns.
a.lc.1 write.... "".re fer more
CDnCemed with style th.n with
....utla. of physical desc:riptlon,
..a
trained to .'lOkI
rwpeatlng the ..m. words.

w.,..

from the same supply sources one could conclude that it was likely that
bOlh wore the same colour tunics.
On the other hand, Fucntes suggested thallocally recruited auxiliaries
could have \\'0111 lheil' own u-aditiomlily coloul'ed regional clothing. For
instance, in thc earl)' 1st cenlUI1' lhe Creek histOlian Su-abo (GfoIK., 3. 3.
7) described hO\\' the men in a number of Spanish lribeS dressed in black,
which may have been reneclcd in the colour of the tunics of lhe
tribesmen rccnliled illlO the A'Itlllian infanu1' and cavah1' units raised
from lhese regions. Almost cenainl)', before the reJorms of Augustus
which established the attxi!ill on a professional footing, new recruits
would have provided lheir own clolhing, a practice which mayor Ill:!}'
nO! have conlinucd for some Lime aflcrwards. If there was a
designatcd colour differcnce between dLil.cn and non-citizen soldiers it is
not known how this applied to those dlizens who are known LO havc
fonned some units of the auxllill. or if Lhe cohorts of citi7CIIS were
distingui~hed in an}' war
The best cvidence for the colour or auxiliary tunics are lhc rrescoes
from Dura Europos, Luxor and C,stcllum Dimidi (to be discussed in lhe
forthcoming second pan of !.his study). Frustratingl)'. lhese all dal,e from
after lhl: Constillltio AnLoniniana b)' which the Emperor Car::.calla (AD
211-217) granted Roman citizenship to all freeborn inhabitants of the
empire. wiping Oul at a sLroke the u'::'lditional differences between citil.en
legionaries and non-citizcn auxiliaries.
(22) Terlullian (born c.AD 160/AD 170). De CQrolla Militi.s, 1.3.
Tertullian describes a soldier being accused of being a Chrislian because
he rerl.l5eS lO wear a military crown. Before uial call commence the
accused is slripped of his milit;u1' insignia including his cloak, sword and
sandals, until he is lefl clad 'on1)' in rcd'. Of furthcr interest is
Termllian's descripliol1 of the milit<lry boots as mliga s!Jef'lIla/lJI'ia, which
suggests lhal lhis unfonunale soldier \\~dS a member of the emperor's
elite ca\'llry bodrgu;u'd, the SpeCUkltlJll!S Ill/gus/ii or 'Aubrllstan Scouts'.
(23) Encaustic (wax) portraits, Egypt; 2nd/3rd century AD
it is beSI LO inlroduce here the fUllcrdl1' portJ'rtits from
Ihe Fayum district of Eg)'pt. as lhe finest examples d,ue from lhe
1-ladrianic-SC\'el'rtn peliod. Apion. an Eg)'plian who cnlisled in !.he
Lmpcrial neet and ....'as u'rtnsferred to Misenml1 ill haly, wrOle to his falher
tdling him lhal he had sent by way ofa friclld a little picture of himself
- doubtless a portrait similar to those discovered in Eb'Jpt iLSClr. This
reminds us that cven II r.,irl)' humble soldier could anord lhese works of
art and th,11 lhey do not represent soleI)' the onicer class.
Hundred.. of llwse paintin&"S have no\\' been uncovered. They arc
painted in wa.'\: on either linen or wooden panels, a technique known as
cncaustic. DUling the person's liretime these portraits were probabl)'
displa)'ed around the house, but when !.hey died the)' were placed over
the race or the deceased and incorporated into the funernl bandages
wrapped around lhe bod),. In vicw or the al>o\'e letter it would not be
surprising if some or lhe slIl"viving ponrnil'i represented lllililary Illen; in
lact at lea.,>t six paintings ha\'c no\\' been identified as soldiers, and not as
members of .he PtOlemaic dynasty as previollsly believed. This identifi
calion resl'i chieny on the studded red leather s.....ord baldrics that lhese
Chronolo~,'ically

figures wear. A comparison with tombstones from the early Prindpate

...."Quld sU'ongly suggest thaI those figures with sword belts crossing from
their light shoulder to their left side would be centurions, who al this
time wore their swords on the left. Most of these men, whether their belts
cross to the ten or the lighl side, wear white tunics and blue cloaks; if
some of them are centurions then the}' are nOl distinguished in any
ob\ious way. One (see page 22) wears a red cloak.
Although the Fa)'um portraits genera.lly depict ant)' lhc subjcct's head
and shoulders another ponrait. from Dcir-eJ..Mcdinah neou' Luxor and
dating to the l1lid-~rd cenlury AD. shows the soldier dO\\11 to his waist and
reveals much of his ,sv.'ord. Like the men in the otht'r portf<lits he tou
,,'cars a \,'hiIC tunic. although it can be seen thaI l.he sleevcs arc long; but
this Tllan 's cloak is red. His sword is worn on his left side. but unfortunau:l)'
b)' this date tl1is feature is no longer unique to centurions.
(24) Arrian (c.AD 129-AD 130), At'S Tactica, 34. I.
DcsCI;bing thc spcctacular training displa)'s carlied OUI b)' I..he Roman
camlr)', Arrian mentioned that instead of armour the troopers \\'ore
coloured tunics. Arrian calls these tunics 'Cimmerian', and adds that
l.hey were shaped like the body armour. I-Ie goes on to say that they were
either sc;u'!el. blue or a variely of colours. Together with the decorated
sLandards and lhe }'eUow horse-hair crests displa)'ed on these occasions,
the. cavalry would have prcsented a coloUl'ful spectacle. which was
obviollsly whal was intended.
What may in fact be one of these cavalry sports tunics may be illuJi..
tmted on the tombstone ofT.Fla\~lIs DraccllS, who servcd .....ith the Ala I
Fla\;a Domitiana BriL.l.nnica to.'ICR. This tombstone rel.ained c\;dcnce of
its original colouring, but unfortunately the decea.~d is on I)' reprcsclll.cd
in a head-and-shouldcrs format, unlike the full figures depicted elsewhere
on cavalry gr.we stelae, Draccus appears to wear a bluc tunic ovcr a scarf.
but this could cqually be a blue cloak. \\'hich would at least be in keeping
\\;th man}' of the portraits from Egypt (sec page 33).
(25) Antonioe Wall distance slab, Bridgcness, Scotland; c.AD 142
Wild (1968 & 1985) notes thaI when painted colour sun;ves on
tombstones of civilians ill nonJlwest Europe uleir coats, capes, cloaks and
scarves are invdl;ably yellow or )'ellow-orange outlined in red. The I.races
of red painton the cloak ofa soldi('r all the Bridgeness distance slab from
Scotland dating La tJ1C Amonine period (see page 34) could therefore be
from a similar outline. The cloak consequelllly might originally have
beell yellow 01' yellowbrown: bUl in the light of some of the evidence
cited above a red cloak ma)' nol. be out of lhe question eilller.
The scene on the slab represents a lllililal1' sacrifice, slIUfJelOlirilill, .....ith
a SCllior officer dressed in a toga and ulree other figures including a
vexiliarills behind. One of these soldiers has a paenllla and scarf while the
other wears a SflgWII. It was the sa!Jum whicb reL.l.ined traces of paim, and
Fuentes belicved that this man ....~dS ulerefore a centurion.
(26) Tombstone, Haghia Triada, Crete; lale 2nd/early 3rd century AD
A lOmbstone of a marine named Sabinianus detached from the
J\lisenum neel.. found in Crete. He wears a !Jrmlllla over a tunic; trdCes Qf
rcd paitH were found 011 both the cloak and tunic.

OTHER CLOTHING

On Trajan's Column all the

legionary soldiers have their
scarves tucked beneath their
armour; but the aweJlla. both
foot and horse. wear the rs
outside the armour. tied at the
front - as Illustrated by this
dramatic example, an auxiliary
with his hands full While taking
trophy heads from Daclan dead.
There seems no practical reason
for this convention, which Is
repeated on the Column of
arcus Aurelius; pam ps the
only explanation Is that this was
one of the artists' conv ntlonal
ways of dlstlnguishlng the noncitizen auxiliaries from the
citizen legionaries.

fi

.it

again

e fr m

ci

The tombstone of Apinosus, "

wortlman from Nievre in France,
shows" rare Instance of a scarf
worn olltslde the other clothing.
This particular example is a long
strip of material - est mated at
about 1.70m - with fringed ends.
It Is very similar to two scarves
found in an Iron Age burial from
Denmartl; and, of course, to
countless scarves worn today.

BELOW Barbarian auxiliary from

Tl"aJan's Column, wearing baggy
trousers; there appears to be a
fabric belt around his waist, but
It could also be the top of the
trousers rolled down. Finds of
similar trousers from Germanic
sites have added
belt-loops
around the
waist.

Children" soc:b fTom elUler end

of the Roman EmpIre: top, from
Egypt, end ebove, from
Vlndolanda, England, They
llluatrate the varying quality In
'loman tertllea aa welt .a two
dlfferenl soc:k dealgna. TtMo
ElWPtlan verslOf'! would be worn
with open, Hndal.type boola .nd
h.a. se~rately knitted big toe
10 allow e boot strap to pa..
between the toea. The
Vlndolanda eumple la cNde by
com~ria.on, end Is made from an
upper and Hie roughly tacked
tog.eth.... thl. ty~ could be wom
wtth the enclosed type of boola.
(The Manchealar Ml/Mum, The
Unlventlty of ManchHter, and
Vlndolanda Museum)

A ca~u. - en enclo.ed boot of

the type used by orncera - found
at the IHlatecl deMlrt fOI1 al
Our, lbrim, Egypt. (Copyright
The Britlah MUlle1Jm)

in Pompeii shows two fares wearing purple-coloured sashes over I.heir

white tunics, A similar colour scheme could be envisaged Ivorn by ofTdUly military men.
Experiments by Peter de Haas with a reconstnlcted \\'aistband used a
piece of material 350cm long x 30cm widc. A llumber of obscwduons
were made when the waistband was I\'orn with replica armour. II. was
found that the I"'listband supported the back, making il easier 10 endure
the \\'eight of the armour, A waistband would also have prc\'ellled any
rough edges from the rivets on the back of the bells from ripping !..he
tunic; and would certainly cover an)' LUCks around the ''''list if thcy wcrc
deemed unsighuy. If ule last turn of ule waistband \....."'s doubled a pOLich
abOLlt IScm deep \\~dS created, providing a useful poekCL On military
tombstones a rectangular object often appears to be tLicked illlo U1C
waistband. and in Ult~ case of me aI/xi/ill ulis has been referred to as uleir
discharge diplomas. However. a." le{,riomuics c."lrry them 100, alternative
suggestions are thattJ1CY are wriling tablets or Icaliler purses,
If Roman soldiers did indeed adopl the practice of wearing
waistbands, Ihen two other sculptures may suggest where they
oligimued. An El.ruscan I""'rrior on an urn from Volterra has a band of
material wound around his waist instead of:t belt: and the sculplllre of
a Celt now in Avignon r.,'IUSCUIll, France, also reveals a similar band of
materia!' but in Illis case JUSt visible beneath his sword belt. In bolll
instances ule warriors wear theil' sashes over a mail shirt. perhaps adding
a splash of colour to an armour iliat was othen\'ise quite dull.
Military boots
In spite of being perhaps Ule bcst-knOl\'n item of mililary c!ouling. there
is pr-dcucally no evidence - archaeological, sculptural or oLherwise - for
Republic'lIl military footweal', lnvariabl)' sculptures from ulis period
depict Roman soldiers barefoot, bm it is generally accepted among
modern scholars that Ille details of footwear \\'ould originally ha\'e been
added in paint which ha" long since WOn1 ofT.
One notable exccpuon is a tombstOne from Padua in haly
representing Minucius, a centlll'ion, which appears to show an enclosed
boot probably of the type known in the ltupeJial period as a cakeus. The
Cflkew c:.lme in three variants which reflected ule status of the .....earereither patrician, senalorial or equestrian. nlerefore in the anny uley
would only ever be WOI11 by senior oOkers, including IllC emperor if he
led his troops in Ihe field. There are references in Roman literature to the
booLS of SenalOl1i being
either scarlel or black,
which coLild be interpreted
as meaning they were made
of leather dyed scarlet tied
up with black thonged
laces. ealai were almost
certainly made with soft
leallter uppers thai do not
survive very well, I\'hile
sculptures also sholv Illcm
as having a clearly defined
scparalc sole. Conscquenu)'

0'

n,. sole
blIdl)' worn hobn.lIed mlllt.ry boot 'rom
Vlndol.nd England, pemaps
evidence of the .pparent
dlnleulty 01 .upj)l)'lng thl.
partleular fort .Ituated In u..
eefll,. of the northem frontier
zone. While providIng a firm grip
on grass the nail. on the IOle.
CO<.lld produc:e fatal reautt. on .n
eKe_lvel)' smooth .urtaee:
JOMphu. re<:onI. the death 01
c:enturlon during the .I_Ish
w..... who slipped and feU while
leading a eharge aero.. the
.tone-fl8!iJ'!iIed floor of the
Antonia fortress In Jerusalem
lBJ. e.1.8}. Modem I1Hlnacto...
....::ount.....Imllar problem. with
recan.trueted boola on many of
today" smooth road and
pedestrian .urf.ee
(Vlndoland. Mu.eum)

a number of sewn bootsolcs round on military sites of the 1st cenUlt)' AD

have been idclllificd as the remains of (alui.
Due to lilck or evidence it is impossible to S<"1Y when the 'dassic'
military bOOLS. caliga~, first appeal'ed, but they were certainl)' in cOlllmon
lISC b)' the German campaigns of Augustus and his successor Tiberius.
The stOry or Gaius. the future emperor. is well known but illulllr::lleS how
closely mlig(U'were associated with lhe military attbis date, As a small boy
Gaius accompanied his flit her Germanicus, the adoplcd son ofTibcrius,
on campaign in Germany; the child commonly \\'ore miniature versions
of the soldier's boOLS, and was given the nickname 'Little Boots' by which
he is inr."1ll1ously known to posterity - Caligula.
This as.'iociation is further supported by an episode in 'The Sat}'ricon'
by Petronius. His hem Encolpills tries to pass himself ofT as a soldier
when he is sl.Opped in the street by a soldier for wearing a sword; the real
soldier immediately sees through Encolpius' deception because of the
whit.e slippers he is weating.
Despite their appearance caligtlf are in fact boots r.uher than S<"\l\dals.
Gillfgo!' were made from cow or ox leat.her prepared by a vegetable
tanning process which took at least two years to complete, E.."\ch boot
consisted of three layers: an insole. an outer sole. and an upper, which
with its distinctive latticework Clil-ollts gave the boots their familiar
sandal-like look, The layers of the sole were clenched togelher by iron
nails which, "'hen hammered illlo a boot placed on an anvil or last,
recur...ed illlO tile leather. It has been noted that the sound ofthouS3nds
of men in nailed boots marching in step on metalled roolds would have
pro\'ided an audible reminder of tile power of Rome. Juvenal warns
pedestrians about encoLlntering a soldier in the street and gelling their
toes trodden on by his hob-nailed boots; and advises abrainst provoking
soldiers, who might kick tlleir shins in retaliation (Sat. iii. 232 & xvi, 25),
The open latticework upper provided cxcel1em ventilation, reducing
the possibility of sweaty leel and blisters. as \\'cl1 as being frce-dr,tining,
In addition the flexible straps could be at\.justed to adapt to tile wearer's
1'001. PrnctiC.dl experiments with reconSlructions silo\\' that abrasions to
the feet are minimal because of the absence of pressure 011 tOejoil1ts, the
ankle and big toe, but socks could also be worn for additional comfort
or in cold wealher, The caligrlf' were laced together with thongs tied
through Ihe openwork SUotp ends: indeed, the tidge efTecl causcd by t.his
method is often the onl}' fealure of the b<xlt \\'hich appears on soldiers'
tombstones, the other details ob\~ollsly being added in paint.
Apart from minor V"J.riations surviving {(IUgal' are extreme I)'
sllllldardiscd, suggesting tbat palLer-n models were issued fOl' the
soldiers thetllseh'es to copy. perhaps under thc SUI>CI... ~sion of a unit
shoemaker, Except in extreme cases there is liltle evidence to suggest
that boots were constantly repaired, which implies lhat it was quickel'
and easier to replace them enlirel)' than to attempt repail'S. A
dOCUlUctll from Egypt records how a soldier was issued with three pairs
of boots per year. which gives a rough lifespan for (oligot: that modern
cxperimenu with reconstructed cX<llt1ples seem to corroborate,
However, the lell.cr to his fatller frOI1l Claudius Teretllianlls, who was
aUilched to the Alcxanddan fleet carly in thc 2nd century AD, reveals
that some hoots "'cre clearly unsuitablc, He describes a particular style
of boot as wOl'lhless, states that he has to provide bimsdfwith fOOtwear

Enclosed marching boot from

Vindolanda, England. These were
among the early successors to
the classic caligae. (Vlndolanda
Museum)

Late 1st century AD nailing

patterns from Vindolanda, after
C.van DrlelMurray. Nalls were
not hammered into the sole In a
random manner, but In regular
patterns which demonstrate that
the Romans understood the
principles of the distribution of
weight on the foot during
marching. Weight is placed first
on the heel and then transferred
forwards diagonally towards the
big toe. Accordingly, the
placement of the nails generally
follows a 'D'-shaped pattern.
Modem scholars have noted that
the Romans anticipated the
principles of computer-designed
sport shoes almost 2,000 years
earlier.

MISCELLANEOUS ITEMS

Detail of a relief In Rome

showing a sacrificial scene from
the time of Marcus Aurelius. The
attendant Is dressed eJl8ctly as
similar mill1ary attendants are
shown on Trajan's Column.

BIBLIOGRAPHY
monuc i,

., L'Eserrito di' Cesare 54-44 a.c.,

(Milan, 1996)
B alti, ., Tlw A.rt of nrimt 'reere and Rome
(L ndon, 196 )
Bi 'h p, M. '., Thl' Devl'lopmpnt oj Roman Alilitar)'

eqnipment in tilt> Fint Rlltwy AD and its reLellanrl!

to tlU' rm and, oriety, ( h meld 19 1)
Bishop,
oulsLOQ, j. . ., Roman lilitary
Equipment, (London 19.3)
ou n, .L.,
.eig r, j., lHasada fl: Filial Report,
The Latin and rl'ek Documl'ntJ~ U r aI 111 19 9)
iadi, , The My tmous Fa wn P01traif : Fare
from Rorrut1l l~'gypt, (L nd n, 199 )
Fink, R. " RfJlnnn l'vlililaT)l Records on Pap),rus,
(Cas We t rn Re rv
niv rsi ,1971)
Fu me',
,,'Th Roman LiliLa, Tuni' in
Daw n ( d.), Roman Military Equipment: Thl'

rroutml/Pllt

oj W(I/; Proc/' ding. oj lhe Third

1ilitary EquijJmPIII Re earrh ,eminnr Brili..\h
rrhaeological RRport, 1m r 336, 1 7
x.ti rd)
\Ie b om, P. .P., The ile Mo.wir oj Pale tlina,
( work. 1 95)
be ta j., c B nfant , L. (ds.) Thf' World oj
Roman 'ostwlle ( i n in, 19 4
kunda ., he Ptolemaic .tinny, ( Lo kp n, 19 5)
haw, ,,'R man 1 a' " Exerritl , 1 4 and 5
J9 2 ( lou e l r)
heffi r,
ran r-Ta 'lor, H., 1asada I : Fi1wl
Report, The Textiftt-s, (Jem alem, 1 4)
tunnel'
"Roman nn)': War: oj the Empire,
(L ndon, 1 7)
Wild, J.p., ' h
B 19i a and
p.362-423, 19

alIi
RW Teil II,

Principal ancient ources (avaiJabJ as ither

Lo b Classical Library or Penguin tran lations):
.-\.ppian,
ivil War.. The A Illfl tan Hi tone.
criptor Hi I ri
ugusla - I ); a al~
The Civil 1 aI' and Til.!' Gallic War: Dio
Hi 101)1; Fr n linu., trategem '; Horae, atiT:;
Hdoru , Origin .. J llvenal The Sixteen ali res; Li
The Hist(1)l of Roml!' Martial Epigrams; vid Ar
.-\matoria; Peu' niu, at'rieon: PIau l l . Mile
G/ol'io 'us; Plio til Eld r,
atural Hi to ':
Plutarch, Live.; u t niu
Li e oj the TweltlP
.fIe (IT:;
a illl, , nna!.s ;md IE torie " arr, On
tgricultltTl', nd Leain Language;
tius, c-)itome
of MilUm cience: irgil, The ellcid.

Tombstone of a cavalryman, M.Aemllius Durlses from

Cologne, Gennany, The deceased Is shown as if at his
funerary banquet dressed In civilian clothing, along with his
servant.

..
A

,.

Examples of recycled clothing - not to scale. (A) Both sides

of a tunic made up from an old cloak, found at Mons
Claudlanus, Egypt (after Mannering). (BI Textile fragment
with a 'gamma' pattern found at Vlndolenda, England (after
Wildl; the gamma was a darf(er colour, probably purple, and
was almost certainly from a cloak. Like the eumple from
Egypt, this fragment appears to have come from a garment
made up from re-used cloth.

THE PATES
A: THE FALL OF THE REPUBLIC
A 1: Gaius Julius Caesar, c.57 BC
Almost everything Caesar did was calcuLated to get himself
noticed, and this naturally extended to his taste In dress.
Apart from his famous red cloak there is little mention of
distinctive military equipment; but Suetonius tells us that on
at least one occasion during the Gallic war, Caesar dressed
as a Gaul to pass through enemy territory (Suet.. LVIII). In the
battle against the Nervii (57 BC) Caesar grabbed a legionary
shield and fought in the front rank to encourage his men.
Here the shield is based on an example shown on stone
reliefs from Nart:Jonne. France, probably from a monument
commemorating the conquest of Gaul. It Is decorated with
gold- and silver-plated fittings as described by Suetonius
(Suet., LXVII). Caesar's muscle cuirass is also based on the
Narbonne reliefs. Although it is decorated with a lozenge or
rhomboid motif it is rather plain by comparison with other
statues showing armoured figures, and the reliefs may
therefore accurately record Caesar's battle armour.
A2: Marcus Llcinius Crass us, c.53 BC
Although Crassus earned a reputation for amassing a
vast fortune by highly unscrupulous methods, his career
was not without military success,
his most notable exploit being
the defeat of Spartacus in
71 BC. It was h s desire
to emulate the achievements of Caesar and
Pompey the Great that
led to his death at
Carrhae
dUring
a
disastrous campaign
against the Parthians.
Crassus Is depicted
here as described by
Plutarch, 'replacing
his black cloak with a
purple one' (Plu.,
XXIII). His portrait is
copied from a bust in
the Louvre, Paris, but
amended to match
Cicero's description
of Crassus as the
'bald heir of the
Nannei'.
A3: Sextus
Pompeius (died
c.36 BC)
Although represented by
Augustan propaganda
as little more than a
pirate, Pompeius was In
fact one of Rome's few
successful admirals. He
established an ndependent
power base in Sicily, and
remained a thom in the side of
the second Triumvirate until his
defeat in 36 BC. As a result of

Remains of lhe slatue of

a soldier from Cassaco,
northern Italy,
apparently wearing a
sash-like waistband
beneath his
weapons belts - cf
Plate C3.
A writing tablet from
Vi.ndolanda refers to
soldiers repairing
clothing. Including a
waistband. ventra/em.
to be repaired and
restltched. The
rectangular object
(centre) Is probably a
purse tucked Into the
waist band. It Is not evident
whether this soldier Is a legIonary
or an aultlllary.

his tong run of victories at sea Pompelus styled himself 'Son

of Neptune' and wore a blue cloak, according to both Appian
and Dio Cassius (XLVIII. 48). It was unusual for Romans of this
period to wear beards but Pompeius clearly wears one on his
coins, perhaps in mourning for his father Pompey the Great.
All three men were both patricians and senators and thus
entitled to a paludamentum cloak and a tunic with broad
purple stripes - the tunica laticlavia. All three wear the
distinctive calcel boots worn by Rome's elite; various
versions of these are Indicated in Roman literature and art.
Caesar wore red-coloured boots (Dio Cassius, XLIII), while
others could be tied with black thongs or decorated with
silver crescents (JuvenaJ VII, 192).

B: GUARDIANS OF THE NORTH

B1: Legionary, Germany. AD 14
At the beginning of the reign of nberius the armies
on the Rhine and Danube frontiers mutinied over
service conditions. The physical appearance of this
Legionary matohes the account by Tacitus, while his
armour and equipment are based as closely as
possible on the latest discoveries from Kalkrlese,
acknowledged as one of the sites of the running
battle known as the 'Varus disaster' in AD 9. He Is
wearing a red military tunic and has replaced the
standard issue boots With a type called carbatinae.
which are frequently found on milrtary sites in
Britain and Germany, but are not normally
considered military boots. He has adopted a native
fashion - around his legs are wool wrappings. which
sculptural evidence from this period would suggest was
certainly not an official practice. The shield, based on the
Arch of Orange, France, shows a design associated with
Tombstone of a soldIer - possibly attached to the governor's
staff - found at camomile Street, London. He wears a simple
tunic beneath a psenula cloak; the latter appears to have two
different styles of fastener, classed as the 'button and loop'
type. See Plate 83. (Courtesy of the Museum of London).

Leglo II Augusta, wllo were Involved In the mutinies and

.....ere later based in Britain.
82: Auxiliary, Germany, AD 20-50
Reconstruction of a dress unifOlTTl for an auxiliary Infantryman
based on Rhineland tombstones, in particular that of Annaius
Daverzus from Cohors Ill! Delmatarum (see page 3). WlthotJt
evidence from the inscriptions on these tombstones there is
often 00 obvious way of telling whether the soldier Is a
legionary or an auxiliary - in fact the auxiliary tombstones are
sometimes more elaborate than their legionary counterparts.
As neither tunics nor cloaks had pockets, valuables could be
tucked into the waistband, and many tombstones appear to
shoW a purse protruding from its folds.
B3: Benefic;arius, Britain, AD 70
This figure )s based on the tombstone found at Camomile
Street, London, which almost certainly represents one of
the provincial governor's staff. The soldier was possibly
a beneficiarius, a man selected to carry out wide-ranging
specialist administrative duties, Indicated by the wax-coated
Wliting tablets wIlich he carries. These duties included
inteUlgence, internal security and supervision of taxcollecting. Both legionaries and auxiliaries IleId this rank. and
again there Is no known distinction, perhaps other than
quality of equipment. Benefic/arii were Identified by a special
lancehead: this example Is from Germany - later vernlons
became very elaborate. He wears a whlle (unbleached)
paenuia, as described In a document from Vindolanda on
Hadrian's Wall, alld enclosed boots with socks based on
finds from the same site.

c:

OUTPOSTS OF EMPIRE
C1: Auxiliary, Caesarea, Algeria, AD 40
Unlike the tombstones of similar date from the Rhineland,

those from modern-day Algeria reveal a simple style of tunic

and cloak. Black clavi are seen In several North African
sources but their use at this pefiod is conjectural. Uke many
cosmopolitan centres In the empire Caesarea had a volatile
population and riots were common. This tigure Is probably in
walking-out-dress, and a cudgel known as a tusUs was a
necessary addition. They were used in crowd oontrol- and to
beat to death soldiers who fell asleep on sentry duty, a
punishment known as fustuafium.
C2: Auxiliary, Judaea c.AD 30
This soldier is dressed in local garb, which follows Greek
fashions, while carrying out surveillance duties. A truncheon
Is concealed beneath his cloak (known In the Greek-speaking
East as a h/malion): although not visible from this angle, the
corners woold probably be decorated with 'gamma',shaped
designs in contrasting coloor - see Plate H3. Jewish sandals
appear to have been worn without characteristic Roman
hobnails (Shabbat 6.2). One modern ekplanation tor this was
that nailed boots enabled Jews to hear when Roman soldiers
were approaching...
C3: Auxiliary centurion, Judaea, c.AD 30
This centunon is dressed in a tunic, paludamentum and
waistband. He holds a vitls vine stick, the centurion's
notorious badge of rank; It is held in a manner quite different
from the way ordinary soldlern hold the fustis. According to
the Gospels, one 01 the auxiliary units stationed in Judaea
was an 'lIaJian' cohort: details of costume and equipmeot are
therefore based on finds from Italy, including the waistband
from the Gassaco sculpture. However, the sword scabbard is
from Pula, Croatia. The Roman soldiers at the crucifixk>n of
Christ divided his clothes amongst themselves by casting
lots (Matthew, 28, 35-36): if nothing else, this tells us that
they had a use for civilian garments.

LEFT Marching soldlar from the AdamkU..l

monument, Romania, earty 2nd century ADsee Plat. D1. Unllk. his cOl"It.mporarl.s
depicted on Trajan's Column In Rome he
wears no a""our and doee not carry a
marching pack; thie may therelore Indlc:at.
eom. to"" of parade or drill dress. Also unlike
hie lellow leglonariee on the Column, he Ie
shown wearing caltlength breeches.

ABOVE Two una""ourvd $Oldl....

from TraJan's Column - . sllnger
and a stone-thrower. Both men
use their cloakS to carry
missiles, and wear similar
tunica 10 the leglOl"larles shown
on tha Column - cf Plata E3.

D: EXPANDING THE EMPIRE

01: Legionary, Dacian Wars, late 1st century AD
Based on the Adamklissl monument in Romania, this
marching legionary wears no armour and has no marching
pack - unlike his contemporaries depleted on Tra]an's
Column; this could represent a light marching order or even
drill dress. Another difference from TraJan's Column IS the use
of wool bracae, breeches worn to iust below the knee. This
manner of carrying the pl/um, like the modern 'Slope arms'
position, IS seen on both the Adamklissi monument and the
Cancelleria relief in Rome.
02 & D3: legionaries, fatigue dress, Dacian wars
Both men wear the light 'working order' as depicted on
Trajan's Column. The rear view D2 shows the tunic gathered
and tied into a knot at the back of the neck. Both men carry
the standard army pickaxe or do/abra. Trajan's Column still
shows all soldiers weanng caligae. Figure 03 Is a German
recruited into Imperial service. As well as the basic Roman
tunic, wom untied and off the right shoulder to allow free
movement of the arm, he has retained Items of native ongln
ncluding the knee-length bracae and leggings; the design of
the latter is based on finds from Denmark. but in thiS case
they have been made from re-cycled tunic cloth.

E: EXPANDING THE EMPIRE

E1, 2 & 3: Auxiliaries, 1st-2nd centuries AD
These are based on the columns of Trajan and Marcus
Aurelius. E2 Is a barbarian club-man, probably of German
origin, carrying a simple shield made from planks of wood
glued together (later even Roman shields would be made In
this way). He wears baggy wool trousers Similar to those
found at Thorsberg, Germany. The short half IS not necessarily
indicative of service in the Roman army; Tacitus informs us
that it was a custom amongst some Germans to shave off
their beards and cut their hair after they had slain an enemy
(Germania. 31). E1 is a Irregular archer with a long-sleeVed
tunic, trousers and wool cap; leg bindings - like 19th/20th
century puttees - were common in the later Roman penod,

but the existence of bandage-like material at the 1st century

British site of Vindolanda suggests that they were used even
earlier. E3 Is a stone-thrower. perhaps from northern Greece,
as troops of this type known as PSI/Ol had fought in ancIent
Greek armies. He uses a 'bagged' fold of his cloak to hold a
supply of ammunition.
F: THE POWER BEHIND THE THRONE
1, 2 & 3: Praetorian Guards,
1st-2nd centuries AD
We can only guess at how the personal whims of profligate
rulers like Nero (AD 54-68) or Elagabalus (AD 218-222) were
reflected in the dress and equipment of the Imperial guards.
Figure F1 may offer a clue, but the evidence is slight and Is
based on a single wall painting in the Golden House of Nero.
He wears a slivered and gilded bronze 'Iorica segmentata' of
Corbridge type. but with bronze edging. Together with the
Etrusco-Corinthian style of helmet, it gives an overall
Hellenistic effect which would be In keeping With the tastes
of many emperors. inclUding Hadrian. Around his body is
draped a red paludamentum.
Figure F2 is taken from a tavern sign In Pompeii dating
before the eruption of Vesuvius in AD 79. The cloak Is the
common yellow-brown colour but with dark reddish-brown
clavi. The shield is of traditional design and Is based on the
near-contemporary Cancellaria relief in Rome; a figure on this
relief is dressed in similar fashion to the Pompeii figure, but his
cloak is shorter. The men in both sources hold weighted pi/a.
Figure F3 is a soldier from the Hadrianic relief now in
Chatsworth House, England. He wears a loose-frtting tun c
over an under-tunic, gathered and tied behmd the neck. A
studded strap crosses the body from left shoulder to rfght
side at a high angle; it resembles some seen in the later
encaustic funerary portraits from Egypt.

G: POLICING THE GREEK EAST

G1: Marine, Athens, mid2nd century AD
This reconstruction is based on a well preserved tombstone

LEFT Detail of a fresco from the Golden House,

Rome, dating from the time of Nero (AD 5~8).
The figure Is annoured and equipped In Greek
style but seems to wear a Roman '/ories
segmentata'. perhaps Indicating that the figure
was modelled on a Roman soldier or even a
Praetorian Guardsman. His tunic and helmet
crest are green, and the cloak - wom like a
pBludamentum -Is red. See Plate F1.
RIGHT This soldier, probably a Praetorian, is one
of a similar group from a relief dating to the
reign of Hadrian lAD 117-138) now In Chatsworth
House, England - see Plate F3. He wears a
loose-fitting tunic tied behind the neck In a knot,
with the edge of an under-tunic visible at the
hem. In this group severel men have a narrow
strap over the left shoulder and under the right
ann; It Is clearly not a sword baldric, and it has
been suggested that Its purpose was to keep the
voluminous folds of the tunic out of the way of
the sword hilt.

found In Greece. Note that the sagum has a small tassel on

the botlom corner. On the monument the sword appears to

be fastened to a fabric belt similar to those noonally seen
muscle-cuirassed statues. The use of red lor both
doak and tunic Is atlested by the tombstone 01 another
IfUlIine, sablnianus. found In Crete. The lantern is based on
irChaeological finds from Pompeii, and is also seen on ships
depicted on Trajan's Column.
G2: Centurion, Alexandria, c,AO 150
thIS figure is largely based on an encaustic portrait found In
Egypt. He has a white tunic with black clavi, and wears a blue
paJudamenrum brooched on the left shoulder, On his head Is
a gold-plated tin wreath based on an example lound In
Egypt: a number of portraits show both Civilians and military
men wearing these crowns, which may therelore indicate
prowess on the athletics track rather than the battlefield.
G3: Diogmitol, Ephesos, 2nd century
This policeman is taken Irom a relief of a senior police officer
- called In Greek a paraphylaJr - found near Ephesos In
Turkey. this man appears to be greeting his commander,
and on the relief he Is accompanied by two other men
identically equipped. The green cloak - the colour taken
from a contemporary Egyptian encaustic portrait - Is worn
like a pa!udamenfUm, so he could be of centurial
rank. Around the waist is what appears to be an extremely
wide waistband, but this reconstruction follows the Italian
historian AntonuccI. who suggests that It Is actually the
cloak wrapped around the waist.

on

A pollee official
..Iutn his commander In this detail
lrom relief fOllnd
near Ephesos in
TUr1<ey _ see Plate 03.
Note the fabric
wrapped around hi'
w,lst. 'these paramllltlry pollee were
known in the Oreek,peaking eastern part
of the emplra as
diogmitol.

H: SERVANTS OF ROME
H1: Legionary under punishment, 1st-2nd
century AD
A mil'lOf punIShment pmscribed by Augustus was IBCOI'ded
by SuetonIUS: 'He woolcl order men to stand all day In front 01
the general's headquarters. sometimes clad only in their
tunics and without sword belts, or sometimes holding
a len-fool pole or even a clod of
eanh' ~gustus, 24,2). The ten1001 pole would be a surveyor's
decempeda used for measuring,
and IUrfs were the basic material
for camp ramparts, which
might suggest that this
particular punishment was
lor sloppy work during
surveying or construction
duties. The humiliation
seems to have lain in the
removal 01 the weapon
belt - cingula militaris
- which was particularly associated with
military status.

This tombstone
from Athens showa e
me"ne weertng
$~lIm clo.k wtUI
sm.lI tls..1visible
on one edge - see
Plate 01,

Tombstone of T,CI.udllll
Halotl.ls from Cologne, Gennany.
This coml'l'llnder
of. Dalmatian luJtlIlary
cohort evfdentty chose to
be represented in civlllan glm
rather then mllltlry drelS lor his
funeral portrait _ se. Pllte H3.

THE FAll Of THE REPUBUC

I: Gaius JUlius Cae$il~, c.S7 BC

2: Marcus Llcinius Crassus, c.S3 BC

3: Sel(lus Pompelul {died c.3!3 BCI

GUARDIANS OF THE NORTH

1: legionary, Germany, AD 14

2: AUJllliary, G rmany. AD 20-50

3: Benefic/arius, Britain, AD 70

OUTPOSTS OF EMPIRE
Causa,.a,

I: AllJdIlary,

Algeria, AD 4()
2: AUXiliary, JUdaea.

....0'"

3: AuJr11lary C&rlIUlio<J,
JUdaea, cAD 30

,
3

EXPANDING THE EMPIRE

1: legionary, Daelan wars. late 1$1 century AD
2 & 3: legionaries, fatigue dress. Oaelao Wars

THE POWER BEHIND THE THRONE

1,2 & 3, Praetorian Guerds, lSI-2nd eenlur>eS AD

POUCING THE GREEK EAST

1: M.lM, Atn.ls, mId2nd ctI'Itur'Y AD
2: c.nlurion, AIto"andri., mid-2nd century AD
3: Dio{pnItol, ~ 2nd Qentur)' AD

SERVANTS OF ROME
1: Legionary under punishment, 1st-2nd centuries AD
2: Senior tribune, 2nd century AD
3: Commander of au~lIiary cohort, 1st-2nd centuries AD