Archaeology From The Earth

GOVERNMENT OF INDIA
ARCH/EOLOGIOAL survey of INDIA
CENTRAL
ARCHiEOLOGICAL
LIBRARY
accession no.__
CALL N». _ 5 7I.QI /
I5SX15-
1
r
t- .
r !
• * . . *
i' »'
? ■ ?» 'i.
\
_
^»jJc St«n^l
archaeology
FROM THE EARTH
ARCHAEOLOGY
FROM THE EARTH
BY
SIR MORTIMER WHEELER
CL I *1 U
sniiE-
uJê-
OXFORD
AT THE CLARENDON PRESS
1954
0)(fard Umxmify ijmkm E.C*4^
MW itiu ratwflo tmA- fsnaa wtujww *
mmtA.1t tMjmtA «AiK4j WAun rnm*
tmr Ji» iht Unitt
li^--
|«IMt1|£k IH CftAAT BtlTAtK
n
a
t ^
f
-L,\ J ,v vi_ AKCHAEOLOCIGAL
1
r preface
r it sonic'hui^ Ulw a quarter of a cenmty siace 1 first undmook lo
write lSis hook. I now Jr ‘>fl lew than I did ihcn. and vtiJl pmbabJj
in the foLowi.T* iw&“ .often recnnameiMl what noi to do than
what to do. That is ^ should be. It caoooT be affimicd tm
often that bad scholar p in the field ecnenlly involves the lîtlew and
final obtitciadon of evi.jciMx, and bad scholanbip is still all too prevalent
there. On the positive side, I have deacribeJ certain methods and pon-
dples which, on the basis of trial and much e^r, I have found lew
hatmfiiJ than oihen lat have been employed, îany of the «lc««l
- * f- prificiplB are derived from those of the gieniesr of lul
î^^thons if, pj,^ ,’hhcrt I have learned
from 'ind fmto the workmen whom I have employed in
various parts ol ibe world. A few may be of my own devising, They are
offcied, not a* laws, but as the tintca and reminisoences of a tcnginy and
varied aichawlogical cjqwrienoc. For the moM port J ha« Tcftained
from disoissing aspetU of field-archaeology of wMch I mysdf have no
considerable first-hand knowledge. The repeated use of ihc first per-
sonol pconmm is a reminder to the render that some «leasi of the luniH-
liuns of this essay ait appredaicd by the author.
'V
If there be a ooniwrcing theme in the following pages, it is this; an
insistence that tlie archaeologist is digging up, not rAinfi, but
r Unless the bits and pieces with which he deals be ali« to him. unless be
have himself the common touch, be had better «ek out other disnplmes
for his oenisc. Of this more wiU be said in the first and last chapters,
but I wmild make it dear at once ihii here is an earthy book, iiupi to
clerkly hands. Nor for an instant, of course, » it pretended that ^
»radc is mightier than the pen; they are twin tnsinimeots; bur, m this
mann of digging, the cooltolling mind must have in a developed degree
that robust ihree-dimensioiial quality which is Ie» immedutcly osw’
rial to same Other tnc|ukies. In » simple direct sense, arehaw ogy i* a
science that mtisi be Uved, mtBt be ‘seasoned with hoauiuty . Dead
archaeology ti the dual dust ihar hlaw*i.
The substance of ibis book constituted the Rhmd Lectures for tps i.
In its preparation I mml isolate two acknowledgements: m Miss Kath¬
leen Kenyon, my collontuc and moeikss critic for many years, ud w
Miss Theodora Ncwbould who has rdcntlcsaly urged me from chapter
PREFACE
to dupicr ind caoo&t dhawn aU respousibiliiy for chc resuJt. For per-
mbsion M nprodUCO illustntiDns ihiut)u are due m the Sodety of
Antiquaries of London, the Louvre Mtisoum, the Prchisioric Society,
me editor of /tntiquiiy, the British School of Egyptian Arehaeology, the
Ofiental Inatinite of the Unirereiiy of Chkago, and the Ameriran
Schools of Oncnrol Research.
R. E. M. W.
INSTITUTE OF ARCHAEOLOOr
UNIvajkSiXy OP LONDON
I9f*
«.
PLATES
T. Excavamm flf m biutvwj 1^44, Mixâzinci
1B52 i
1L A. An Early t rrm Agt pit ii Scarbomugiii Yorkslvfrei eui ts
thnw tliicfl intcrmplcd plltis^ of the fillSuff. Fhotopraph
by F. G, Siinpfnn
a. Section thnwing sJstecn successive tiyctf of plostcring on
the wall Df the icmpic of the moort-god Sin he
Iraq. PhoioBtvpb by The OrienEJil Emtitutc^ University
nf Chiciigo 3^
ijl. StralilkHlioo at BrfthmoffSrl, Mysore StatCi IimJia 48
JV. A. Chacft; csEOavarinii in the Eist^ Jyjy

Discipline" cxcavuTioci it Arziajnediij South liKliai 1945 64
y. Layout of an arei-cKwstioii (Ta3tilii-Bhir Mniind, 1944) *5

VI, Section peRfied tor thrcc-diincnsiotuii rccnitlSnif (Maiden
Castle^ 193S). From R. E. M. Wheeler, Mtudm Cmtk^
(Sodety of Ahtlqtiarifi* of Londoni 1943) m
VI r. A. Section pegged for thrce-fibaicnsional repjniii^

Slrkap» 1945)
a. Similar seciinn linkiEiK two cemetefy-^refla (Harappl^
Pun^b, 1944) <£9
vni- A- *GhE>sr^ wall at Vcnjlamiiioii HcttfofdshiMj 1931
B. VemlanuLiicii^ north gate; port of the plan recovoed by
clearinB trenches made by wail-rubbers. From IL E.
and T. V. WTieeler, Vznflomnim fSodeiy of Aniiquirica
of LondoDj, 1936)
1%^ A- Rnnuin timber-builditig ut RicbbcuouRh, Kenir traced by
cotoratkin of the soil, Frt™ L F. Uuibc-FoAj Fmn-tk
Rcpvrt oa rAe lixcmtadom cgf JJomaa Firrt ai Ei£h-
bifritufih (Sodciy of Antiquariw of Utndon, 1949)
R, Burial of about 5.000 fl.C. at Itarappir Punjab^ showing
ouUmc of wuoden etrlfui So
X, Tracrci of huidle-comtmainn in the nKlithic Iwng-buTow

ai Skendkbyi Lincolnshire. From a photograph by O* G. S.
Crowford reprttduced in ArchatclepUt vob bontv (iSockry
of Antiqumieft of London, 19361) Si
!tj. Trace! of the Sutton Hs>;> *hipr Sulfoik- Phoioiraph* British

Mtnetuti H
IE
PLATES
*11. Section through the mud-bfick defence* "f g<

dtAdel, Puhtitb. Iwilt bcTof* moo »jC. /»"'« f‘
*IIL A. Air-photogtaph of part nf the Ponhian dty ol Tixilo.
Punjab, in the first century A.P-
88
u. A typical uU fTepe Sialk, central Inio)
11V. Ji. EitcevatSon of n burial pit (firtt«
BtShmagitl, My*« State. India. t*owing the ‘qiudtant
method oT cxcivittloii
n. Silver coin erf Tiberim, dated A.». *6-37* fbom Chjwdro-
98
™11J, Myaoite State
XV Map of the tarthvrorta of Wheaduuopswd and Vero^
with intervening djkei. From R E. M, oijd T. V.
Wheekr. FaroV™ CSode.y oT Antiquaries
195®)
*vi, Rouiened ware of the first century AP. imm Ariham^,^
South India /amwp. i
*ifi i. Selection of tooU bom mi fiaatem eacavntina ^5^
XVTiT. Pattety-grid, for use in dry diuHites
IT*. Part of secrion through the ‘Fosse’ «rth^« Ve^-
mium, ahtwfing turf wall imbedded in earth and nmvel but
i3npha^«d by careful
176
% Xa Towex for hlgb-anglc pboio^^phy
[80
**i. K. ’Sir Flindeti Petrie in the ouiirtyard of the School In
Jentsalem.* From Th* fftWiVaP Arehaalegiitt vol- v, no. 3
(1942)
B. ’Copper Jar no. *77-* Ftom M. S. Vata, Hitrapp^ fDelhi,
19,
*940)
ijii. Ovwlcêd phatDKmpbA of unprepBred tub]«^. From
T7u Antiquaries Jcvfttil, voL %m (Sodery of Anaqumet of 1
Ijondcniy 1937)
j%UT- Rccoiiwxtiictbri of m Early lion Age hmnt^leiLd at
Woodfcuryp WUtihire. Phorograph by the Ministry of Itt-
fehrmarirtn
CONTENTS
list of pla«s **
LIST OP FIGURES “
1. IKTRODUOTORY '
II. historical *
III. CHRONOLOGY “3
IV. STRATlORAFrtY ^
V, THE LAYOUT OF AN EXCAVATION ««
YI. THE EXCAVATION OP A STRUCTURE T*
VII. ON DIRGINC TOWN-SITES
VIII, BURIALS ^3
IX. WATCn-MARSRS’ JOBS
X. TACTICS AND STRATEGY IM
XL STAFF
XIL TOOLS
'XIIL THE POTTERY-SHED 'J7
XIV. THE FIELD-LABORATORY
XV, PaOTOGRAPHY , _ *74
IVI. PUBLICATION AND PUBLICITY ■*
XVII. WHAT ARE WE DIGGING UP, AND WHY? 200
SELECT BIBLIOGRAPHY »*®
INDEX 319
TEXT FIGURES
Section through i gjave in ITydenbad Stole,. Indii.
By MiftaiAKyi Tt^k/rt 1851
Sccdtm by Pill Rivens 6Gi™ ihe ditcii of Wor Barrow, Dorset,
iKg^, relicM proicetcd from athreeûnenaio^ ftcurd
- _ . . ^
3- of a section through Tdi cl Afjuli VultsanCj 1938
iB
4- Scciimi tliJtmeh pan of the site of Bethel, Polratmej 1939
1^
5- ScodiDn thrcrtJgh part of the mound of SinUCj Inn, 1938
IQ
6. Section of part of Harappa^ Paktstan, 1940
Scctiofi through fiorr of a ccraetcry at lliEippi, Pakistan, 194® Zl
7-
Section lUustjrating (left) an appartn: encceaaJmof itw^fthiwra
by Bfl eartcnsimi of the enttifkg CtîsÔ ^ have been depoKted
^6
îfnuIiniiDcnuly
Section ftojTi Btahitwgiri, Myiorc Stiiw. India, sthowing three
9^ 50
cultuial phases with overlaps
10* TabuMion ofsherds reptwnting the three culture* at Brahma-
gui
51
11. Diagrams nluatratiog the stratificatiotl of 0 dry-mound and the
fallacy of recording by mcdmaical levels 54
1:2. TedmiEiiic of section-dfawing 5*
Symbols used in secfioml drawings by the Arcbaeologlcd
13- «o
Sim ey of India, 1944-^
"Pictorial* method of Mcdan-drawing. G, Bfrtu fii
14.
Layont of a trench for threc-dimeaslonal reoorduig 70
15-
Tfi. Section* IHustratmg the relatiomhip of strata to a wall 1 and their
removal by siunmary cicaivadon 74
Stratidcation in rdatioji to the eacicmal wall of the Bomoft
t7* 7fi
amphitheatre ai Cacrla^n, Monmouthshire, 1926
18. 'nic Yucei&fls bamw. Flinahire, showing the laywn for
atcantioti. A/ter Cyril Fmc 9fi
Sekewd sections wmwi the Ysedfiog baitow tooi. 5 «kJ &
19- 97
(Kmlited), After Cyri} FtDC
20.
SuecBsaiw site* of pre^Roman and VeniLunjiUK. (Fm»
tIT
Vtruhpttiimt tUporh 193®! P- %■
SI. Map of India ibowing siiei which hav* penduced 'fouleneil
III
ware' of the first ccnniiy *4).
^Quattd le sqL aura £i£ d
[,'AEnB COOIKT
I
Introductory
T hire ]£ no right vniy of diggmg but Uiertarc irumy wrong ways.
i\inungst the kticr ouf &acce»Or$ will no doubt include ways
which we reguid today os relatively right, in accordance with
the narurul prindple wJiereby eixry giencrawn is liable to bdiLile llie
achicvciDczit of its predecessors. This ardtude is often enough tmiusl.
Before hcuping scorn too promiscuouily upon our untuioted forebears,
it is at least only fair to dassi)' thetr sbnncnmings and to diifcrcfiriate
between culpabiliiy and ututtariiri^* It is unprofitable lo blame Xerxes
for omitting to deploy torpedo-boaTs at Salmnis^ or Napoleon for flitadc-
ing the Bririsb squares with cavalry instead of madiine-guns. Nor, by
the âtne token, can we honourably blame an Early Victorian barrow-
digger Jbr ouiiniug to rtcord by the tiirce-^dimeEisioneJ tncrhnd. Bur
there Is much, far roo much, in more recem archseologicat cxcsvaticyri
that falls short of rhe highest avadahle standards and ibeteforc dam^cs
the iflSh. At ihc besr, esc^s-ation is desiructionii and destruciion unmiti¬
gated by all the rewnrcca of comcmporaiy kBOWiedge and accumulned
cxperictKC cannot be too ziguruusly impugned. In the ffillnwing pages
artcniiou will be tirawn &oni time to diac m the crimes oa Less thitii to
the virrues of oonicmporarici and forebears, in the ftdl awareness that,
m a fdlow-pTacotiuncTj the author b himsdf a vulnerable mrget.
And when in these pages cCruUn methods are suggtslcd as
prderahlE to others in certain couieiCts^ I would ai once make it Clear
that I am indulging in temiiibceuee, not laying down laws. For thirty

yean it has been my occupadoP tO dig up antiquity in a variety of coun¬
tries and eLnmmsuineeSi and I huve sought to profit from experience and
criticism. But ibcrt is tu> (erru to dthcr, and 1 have no doubt that, u* 1
write, further experience and criticisTii arc buildini; up or climinaiing
w'hat ! have writidL WliO would have it otherwise? wCn-a ^’e can¬
not twice descend into the same stream, wd the phUoiopber.
Today it is indeed no easy nmiwr to ptiam'c a just balance berween
oJm and meihud Lu this bUiinws of digging up the pift. l>e%'cloping
technique is liable to Obscum the ohjeetive; at the best t wn of leap¬
frog progress ensues. Aneiupts of a sam-phibsopbic kind to adjust the
priorities are not dwnys very helpful. What in Cict is Archaeology? T do
l
INTRODUCTORY
not myself really know. Theses have been witten to demonstrate that it
is This or That or not the Other Thing; for example, I may refer the
reader, if he will, to a laborious analysis with a bibliography of 612 works
which was produced in 1948 by an American exponent.* It may be
doubted whether all this matters very much to us in our daily application.
We may be content to see that the grass is green without understanding
the mysteries of chlorophyl or attempting to distinguish too cleverly
between botany and chemistry; we may appreciate the Unfinished Sym¬
phony without a profound kno\êdge of the physics of the vibratory
disturbance which we call sound. I do not even know whether Archaeo¬
logy is to be described as an art or as a science; more will be said of this
I in the concluding chapter. But it is at least abundandy clear that
[Archaeology is increasingly dependent on a multitude of sciences and
iis itself increasingly adopting the methodology of a natural science. It
draws today upon physics, chemistry, geology, biology, economics,
jpoliucal science, sociology, climatology, botany, and I know not what
else. As a science, it is pre-eminendy a synthedc process; and if we
Eprefer to regard it as an art, or even as a philosophy, we must still afi&rm

at it is an integration of scientifically observed and disseaed pheno-
ena relating to man; it is still a synthesis/^These are not definitions,
ey are merely descriptions, and incomplete at that. It will not help us
forward to our main objective, the study of human cultures, to spend
time and ingenuity up)on the academic niceties of definition.
But there is one guiding condition of our work which I would empha¬
size at the outset. We belong, some of us, to a generation which has been
involved actively in two wars. Military similes are therefore not entirely
alien to us. And I would accordingly urge that in one \ital respect at
least there is an analogy between archaeological and military field-work
that is recurrent and illuminating. The analogy rests—strangely enough
as between the dead and the deadly—in the underlying humanity of both
disciplines. The soldier, for his part, is fighting not against a block of
coloured squares on a war-map; he is fighting against a fellow-being,
with different but discoverable idiosyncrades which must be under¬
stood and allowed for in every reaction and manoeuvre. Equally, as has
been urged in the preface, the archaeological excavator is not digging up
things^ he is digging up people^ however much he may analyse and tabu¬
late and desiccate his discoveries in the laboratory, the ultimate appeal
> Walter W, Taylor, *A Study of Archaeology’, American Anthropologist, 1,
Memoir no. 69 (Indiana University, 1948).
2
INTRODUCTOflV
across itic ages* whether the time-inLetvui be 500 or 500^000 ymn, h

from miiid 10 tnwlMgem mind* trom mail 10 semicnt man. Otir graphs
and schedules mean nodimg if they do not uJdmalely mean LbaL Of our
scraps and pieces we imy say, with Mork Antony in the market-place,
'You are noi wood* you are not stones, but It is a cruism of whicJi
I constandy find it necessary to remind the student and indeed m}^lf:
that the life of ihc petsr and the present are diverse bin mdltisIbJe; that
Ardiacology, in so far as it is a sdeticc^ !S a sdence which musi be cac-
tenJed ijiLo the li^'ing and must indeed ixself be lived if k is to panakc of
a. proper vitality.
Lei me for a moment amplify this maner. I have said that we cannot
properly unJeisiond the past unJess wt have a living aympathy with the
Jiufiuin itufT whJeh its relies repment. Wc eannot undentand, fur ex¬
ample^ ihe stnucluml meeliaabm of an andent buiiHl''inniiiid unless wc
cun bring to bear upon its details a rational imu|dtuiEion capable of com-
prehendlug md vitalmAg tliem. If wc fail 10 do that, w e are nuL huntan-
ists but metecoHcciors of chsjeaed minutiae^ signifying atmusi noihing.
We would be belter employed coUcctmg bus-tkkets, an occtipadon
which at least involves no damage to sdcntific evidence. Afld since I
have mentioned buriol-nioimdî Ice me carry my example a $tagc farther.
No one hns done more to vivify the mumbo-jumbo of our Bionae Age
burials than has Sir Cyril Fox dujfng a long series of cxcavatioiis os
Dirccior of ibe National Museum of Wales. One example Ln pnrticukr
recurs 10 my mind, Many years ago, lôx and [ were trudging across a
desolate W'dsh moorland and came upon a mail barrow iter wirhin an
carrhen circle. OfFa^s Dyke came steadfaerly up xo the lip of the drdc
and I hen on the orher side Rtarred olT again with equal dcxmninaiion on
n neni' alignmenr. The whole scene sdrred Fox's ready cuthusiasm, and
a week or two larer he had dug himself wdl into the lancbicapc. The
Enoiind had by now vôiahedT and Fox smcNi, in the spint, azoongiÊ in
makcn». He was nlmost physically present at the living ritual^ the actual
procedure of huriaL I quote hit own words:
Under the centre of the mound was a deep and Uige grave-pii on the
floor of which lay the skdetun of a fidl-grown man^ To enable this grave
10 he enrered with caije and dignity ■ iloping passage from gxuunddevcl
liad been cut on the rmfih side. Surrounding the grave-area was 1 circuiar
trench^ which also hud a sjopiiog entroncen and on the tome ilde. Zlj^r fi î«/
tm cjRt: the area round the grave was iwJaEed. The CDtiduakini dnwn from
Uiesr facts w*ete thax the dead manî home was on the north tide of the litc
3
introductory
chosen for his bmialj that he had b«n cerîrifinially borac by ijicuds Of
kiniiûlk up lo, and inm, the noich: that ihosc who carded him were rot
allowed to enter the conficctatfrtl ^tea. round thegnive> but tlml the perstttw
chsir^cd wirh the pcrformaiaci: uf tbc burial liie$ were awaiting the bcarens
beyood the trench. The body was handed over;^ and these persona deaccn-
ded with it klo die grave.'
Wbat iFiarim m im here ajid now is of conrscj ihc paitiinilar
episode which 1 have cited but the creative act of reasoned irmginatioii
that has gone to ihc making ot tenaiJdiig Of iL Foic^s inicrprctation may
TbOt be correcT in all its details; id any event the objecrivc facta upon
which ii is based arc Mly recorded^ and the interpretation of rhem ean
be reshaped in the light of fuller knowledge. The great Thing U that
those fsOT are infuacd with a nuiomil inteUigence; they ernerge from
Fos^s brain os thrcc'-dimcasioELuI entities. Contrast the ordinal)' eicava-
tion-TcpoTt. Year after yeax;! intlividiial aAer tudividtialj learned i»icicty
afier learned sudet^v we are ptosaicaUy revealing and caologuiiig our
discoveri^. Too often we dig up mere thingSj luirepentantly forgetful
Thar nur proper jim h to dig up people.
So mo With andent fortiBcadons. It iS BO accident that leadens in thdr
inicrprctation have so oficii been soldiers: General Roy, for TOmpIct in
the eighteenth ccniurs ‘ Goaend Pitt RlveiS, Mapoleon Ill's colleagues,
and the disiingui^lied ofijcers who manned the German Commis-
sinn in the ninetccuih. Our hid-fortS, as Lelaad long ago remarked, are
the woria of'men of warre^j and thdr study demands t he virile spark
of the mind militaut. It is no mere by-product of the study of culture-
creeps and cccmiic crosswords: both, be it added, adtnirabk and indeed
csscatial pitoccupdiions, and lovable after thdr fashion.
Huough, now, of tJicse general fuauers of approach, Some pains have
been taken to emphasize mitiaily the vital priBCiple because bo much of
the substance of the foUowkig chaptens is conceded with the dry bones,
with procedures, with the cOFactiou of evidence rather than with im
interpretfltiotL It is to be hoped that, even so, thelf contents may be of
some slight interest beyond the Hmiied range of archaeological pracd-
lioncis. Today* we can soirccly touch hisioiy without touching archaeo¬
logy^ itnd for the greater put of human existence we caonoi touch
histut)- at all Prehistory Is a huQilred times as lotu; as history, in the
uioiationist usage of tlie terms. The proper study of mankind involves
^ C. Pot imd Brwre DiekuUk Th4 Cvitwc cf N^rih-tpeit Mtnvpt (Com-
bridg*, 19JO), p. 54.
4
INTRODUCTORY
therefore a disproportionate amount of archaeology; and, unless the

reader has more faith in the professors than I have, he will demand, and
very properly demand, to know something of their credentials. How docs
archaeology work} We know that the historian goes to his documents
and his epigraphs. On what does the prehistorian depend? An attempt
will be made in these chapters to indicate, mainly from personal ex¬
perience, some part of the rather complicated machinery with which he
wrings his evidence from the earth.
II
Historical
IRST it may be useful to see something of the upgrowth of the tech¬
nique of archaeological excavation during the past century, and so
JL to establish, however sununarily, a perspective for our subjea. It
would be of little profit to carry the matter further back behind the
Victorian era, and to follow others in a general discussion of the evolu¬
tion of conscious antiquarian thought from the time of the Renaissance
or earlier. We are concerned here with methodical digging for systematic
information, not with the upturning of earth in a hunt for the bones of
saints and giants or the armoury of heroes, or just plainly for treasure.
It is of no importance to us that, as long ago as the twelfth century, the
monks of St. Albans were digging up, recording, and sanctifying the
remains of some poor Saxon in Hertfordshire; nor even are the early
probings of the curious into sites such as Pompeii or Herculaneum
strictly relevant to the theme, although they no doubt played their part
as a stimulus to subterranean research in partibus. True—a. truth
stranger than fiction—as early as 1784 a future President of the United
States of America, of all people, was already carr^gût an excavation
on surprisingly modem lines in Virginia—:^he first scientific excavation
in the history of archaeology. This astomstnfig episode will be described
in a future chapter (p. 41). It was unique not only in its age but for long
afterwards, and it were bener to begin here with an incident of more
normal kind.
The chosen incident occurred in those forties of the nineteenth cen¬
tury which were more fruitful than any other decade, whether in Britain
or abroad, in the foundation of new institutions for the furtherance of
our science. On all hands new antiquarian societies were being estab¬
lished in England and Wales. The close of the decade was marked by the
transfer of the collections of the Society of Antiquaries of Scotland to the
Crown, and by the establishment of a separate section of British and
Medieval Antiquities in the British Museum. ‘Within no very distant
period’, wrote a contemporary observer, ‘the study of antiquities has
passed, in popular esteem, from contempt to comparative honour.** And
» ‘Introduaory Address* by E. Oldfield to the (Royal) Archaeological In¬
stitute (itself founded in 1844), Arch.Joum. ix (1852), i.
6
HISTORICAL
it was in accordance with the new spirit of the times that on an August
day of 1844 on an obscure hill in Kent above the vale of Maidstone,
there assembled a top-hatted gathering of the local nobility and gentry,
reinforced by some twelve or fourteen labourers who proceeded to hew
a great gash through a tall Romano-British barrow.
It was the labour of four long days [the contemporary account informs
us] to cut entirely through the barrow, but we who were not absolutely
diggers contrived to pass our time to the full satisfaction of all the party....
A plentiful supply of provisions had been procured for pic-nidng on the
hill, and we remained by the barrow all day, watching and directing the
operations.... We contrived to pass our time, at intervals between digging
and pic-nicing, in games of various descriptions . . . and in other amuse¬
ments. The season was fortunately exquisitely fine, and it was only once
or twice that we were visited with a heavy shower from the south-west,
when the only shelter was afforded by the hole we had ourselves dug ...
in which we managed to interlace parasols and umbrellas—much as the
Roman soldiers are said to have joined together their shields when advan¬
cing to the attack of a fortress—so as to form a tolerably impenetrable roof
over our heads. . . .*
The woodcuts (PI. i) speak for themselves.

Such was the sense of light-hearted adventure that stirred our great-
grandparents in the dawn of a popular interest in archaeological field¬
work. That spirit is not to be scorned. These gentry were, of course, the
t veriest amateurs (blessed word!) or merely curious spectators. But their
curiosity was the chrysalis of cultivated opinion from which, in the full¬
ness of time, modem archaeology was to emerge and take v-ing. Of the
same ilk were the ladies and gendemen who a litde later flocked in their
carriages to watch Canon Greenwell dig a barrow on the Yorkshire
wolds. A very notable man, the Canon; at the age of 97 he could still
land his salmon, and those of us who have the good sense to be fisher¬
men are careless of our best interests if we cannot find a ‘GrcenwcU’s
Glory’ somewhere in our fly-books. But it is as the author of British
Barrows that for the moment he recurs to us. The apocryphal history has
it that on one occasion an elegant and admiring assembly clustered round
him whilst, with the intuidon of the inspired connoisseur, he chose his
barrow from a series within range and turned his labourers on to it. In
due course, a handsome burial-ura began to emerge from the excoria¬
tion, amidst the manly gasps and ladylike cries of the speaators. As
» Gentleman's Magazine, Dec. 1852, p. 569.
7
HISTORICAL
alwajrs, the dear and wonderful Canon was infallible! The workmen
gingerly lifted the um—exposing beneath it a copy of the previous day’s
Timesl
Such were the methods of a less finicky age. Nevertheless, a procedure
which was still justifiable in Kent in 1844 or on the wolds a few years
later was in faa rendered obsolete by a young army officer working in
his spare moments in central India as long ago as 1851. These dates are
worth emphasizing because the work of Captain Meadows Taylor,
though limited in scope and scarcely noticed at the time, marked or
should have marked the beginning of a new epoch in technical method
and scientific observation. It did not, of course, stop or even check
picnic-party excavation. But the fact remains that, after 1851, digging
such as that described above, or such as normally indeed persisted imtil
far more recent times, was for ever obsolete and culpable. Let us pause
for a moment to examine this innovation more nearly.
The basis of scientific excavation is the accurately observed and
adequately recorded section. More will be said of this matter in a later
chapter. Here it will suffice to premise that the successive accumulations
of construction and debris on a buried occupation-site have much the
same validity as the successive pages of a book, and, to be understood,
must be comprehended in their proper sequence, like the pages of a
book. In a haphazard excavation such as that of the Kentish barrow it is
scarcely necessary to observe that such accumulations or strata as may
have been present are not merely tom ruthlessly from their context but
are not even px)stulated by the carefree excavator. The whole mech¬
anism of the ‘book’ is ignored or misunderstood. If we set aside the
eighteenth-century American statesman of whom mention has been
made, MeadoT^^ Taylor was the first man, so far as I know, to hint im¬
plicitly at the true function of the excavator and recorder in this vital
matter. He was an officer and highly successful administrator in the
relatively obscure service of the Nizam of Hyderabad State. He is known
to fame, if at all, as the author of The Confessions of a Thug; but of his
work as an'amateur archaeologist the Dictionary of National Biography
says nothing. For that we have to turn to three papers published in the
Journal of the Bombay Branch of the Royal Asiatic Society^ iii (1851),
i79-93> and iv (1852), 380-429; and, unexpeaedly, in the Transactions
of the Royal Irish Academy^ xxiv, pt. iii. Antiquities (1862), 329-62.
These three papers show an acuteness of perception and a technical
competence far in advance of the time. Meadows Taylor dug into a
8
l^LATE t
Ev»viikN:i of 1 bvrnw, 1K44

Mdiairiiv,
C^« f- ^,5
I
HISTOKICAL
number of The megiiliThic tombs charafterisiic of central and soutbetn

Indiaj and drew and described scctioTis which prejer\'^c an infetmative
and on^î^dTlg rccoTd of what he fatiTidj with diiTercntinied strula
(Fig, f). ^'hen one iecall$ that cfoly so™ or e^ht years divide the irork
of The Kenrish parry from that of the lone fever-stridten Englishman
in naiivc India^ the latter's achimêmenr stands out as i kodmarje in
the annals of archaeology* Uofortunaidy, Hritish aithaeology in India
was not destined to mointaiD this prc-cfntnence.
During the following deoidcs much pick-and'shovd work of a spec¬
tacular kind dreiiv iDCteitsbig attention to the possibilities of the mft.
I'toy and Alyoenae set Homer and SchJiemannhrmly upon their feet. In
France^ under the eagle of Napoleon III, the admirable Colonel SiofFel
ran Julius Caesar to earth with n cnnsiderable measure of sutxas, a^m-
bming military manoeuvre with a ocnain nidimemary icchuial sMll.
He reoJured, for example, the signiheanr fact that soil, ouce dkturbed,
imely quite resumes its original compactOKS, and that in particular the
ditdies of Caesarea enmps^ now urtcrly levelled, COuld still be detected
in section. His method of March, reasonably cnoughj was to attack his
buried fortiiications with massed fortmtioii of trial-trenches.
I placed the workmen [he wTote] with pidcs and shovels In se^êral
htirs, in a direction perpendieukr m one of the presumed sides of the
I buried] camp, the workmen in each hic lo ur 30 metres fram one another.
Each of than wraa ordered to remove the Uyerof Aumut to a width ofs frtT^
If, after having removed this layer to a depth of 70 ceutiinetfes, they fdt
that rhcLr picks were vtiikiug a ledstont soil, the Jufercnce was that this
soil had never been removed and that there was m Rirunan dlmh_ The
wTnkmcn then continued to advance so losig as nothing di^creat was en-
countxmed. But when they arrived, beyond doubt, at the ditehi the dUfci-
cncc was at once apporeur
IL may be added that the Colonel's rejênred enthusiasm was shared

by his Emperor who, at Gergovk, 'was so wonder-smick nn iiedng she
profiles [of the Caesarian ditches thus revealed tnsecTion] that he thought
of buying the hill to preserve them. He abaadened this idea when he
Itamcd that tlic inhabitauis did not desire 10 be dispossetwd, and he
ordered me to fill in my trenchoi and to restore evtrythhig to iii former
The path nf f^derK^ nm'crdid nio smooLhl
But ii wuH left to another soldier, a Bninn, to make the first substantive
advance in the technique of ^locavari^ and^peurding. In the yeirjS|dJ
* T. Rice UoEnwi, CjiMr** tidnquat cj Gnu! (Difordi f&ir)j pp. fL
9
rT fn ^ in
10
HrSTDiiJCAt.
Gcncral^L^c^F^ by & ^kurpming ôrici of diuocts sooceeilcd Lo flit

Riven cmie sitiiatod in Wilt&likt uod easttm DorseE^ and under the
temu oftbe wiU osûmed the JiameFitlHehnd already varied a
military coreex by i study of arlifa^ along evolutionary Uoe$i and bad
been admitted to die Royal Society for (intbcopolQglcal work. Tbe
principle of die evoludon ofliuman ui$tititdon$ was no new concept, t i
hud indeed found eipression 05 to back os 17S6, when Sir Willkm
Jones in Calcutta bad enunciated die evolutionary telariaasbipofccrfain
languages—Sanskrit, Greek* Lfltio> Persian, Cclric, and ^Gnihicfc^—
onotber example of British pioneer scholarship m parijbm. Bur ihc
dramatic developmefit of that principle in relation to numral spedes
during the fifth and sixth decades of the ninciccnth century bud given
d new unpCTUS to its applicaTion in ihe humuiisijc field. And now the
General* In rhe course of bU musketry, bud found the same basic prla-
ciple Ln rhe d^’eôpm^t nf fec-ar™, and buid extended it to other
human insTruments^ HU w^holc jppmsch tn oichaeology was thu^ from
a modem angki and for twenty years he exploied the ancient sites on or
adjoining his estate with a sdenec and sdiolatshlp that half a century of
subsequent work has supplcmenied rutlier than superseded.
Above aUj the Gcncral^s constant plea was for "gteatet precision and
derail in cxcB^dpn^.'
It will, perhaps, be thought by nunc [be rcmarl^] that I have recorded

the acuunlious ^ . * with unnecessary fuUneUp and 1 am ¥ure that 1 have
done iiin greater detai! than has been customary, bm my experience ai in
excuvuror has led me to think that investigalJaaa of this nature are not
generally stiffidently searchingT and that imich valuable evidence is lose
by amittiug to tecoid them carefilllyi *. * Eaccavaion, asm rule, record only
those things which appear to ihnn iffiportant at the tfmei hut freik prob¬
lems in Archtwology and AjubropolDgy are cvnstantly arialngi, and it cm
hardly fail tu have eicopcd the twice of iniduopolcguFts ... that, on turn¬
ing back HJ old accounis m search of evidence, ihc poLnia wliich wûid
have been mosr valuable have been passed over from hdtig thought un-
intrrrfting at the time. Every draiJ ihmild, therefore, be recorded in the
manner moii condiicuve to facility of reference, and it ought at all tiroca
to be the chief object of an cxeavalur to reduce his own personal equadEsn
to a irnnimum.^
In practice, Pitt Rivm's method was to record every objM bi such

a manner that it ootlid bc replaced accuraicly in ita on the
* txatTixtHm irt CfCftojrf, 1 tiil7)p PP- xri-xm.
n
OltEH J^BSUT 4«amT m
13
KVWOI
13
HTSTOniCAL
recorded pLm dad sectioa. Thut is the eucace of thicc-dliutnsloiial^
recording, and thiec-diiDcnîonkl rccordiag is the essency^ of modcrii
□i^'ution. Since the time afPiii Rivafs we liavc in some respects ehbor-
Hied his tcchoioi] ptocc&i»es^ und there is no doubt tfuit the be^t records
of the present day stupusii die Genend's. It issaliitaiy to reflect^ however^
whflt he might himself have done with another fifty ycftr$ of experience
behind himf
Let ns gldiioe for a fuither monient it the master's working prlndpks.
From the outsetj he grasped the vital need of an adequate stifT], a need
wltkh has since his time been loo often neglected^ widi deplorable con¬
sequences. He at once
detennined to orgoinizc a regolor staff of assbiints^ and to Train Them to
thdr respective functions after establishing a proper divi»ou of worlu «« /
'^rhe work of mpcriatgndJnfl the dj^png—rhough I never ailowed ii to be
carried on in my absence!^ always visiting the eiccDvation^ at least three
times a day imd iimuÊng to be sent for whenever any dung of importunce
vfos found—more than I eould iindcrrake single-handed ... and I had
by ample experience been taught that no excavation ought ercr to be
permitted except under the immediate eye of a responsible and trust-
worthy lupcrintcxidcnc.^
Or agata:
The citcavazlonjs in Winkclbury Ja Wllrshiro*camp^Jhavfrig been carried
on before my assistants were sulfidently trained^ I never left ihe itround
during any great port of them. One or more af the asais-taiiti 'wib always
engaged in niperiiiTcriding the workmen upon the groiuid or in drnwfng
the objects, in repairs to the skulls and the pots^ and in forming the itdic
[ttbEcij by whieh means the records have been kept up lo ditCj and it has
been found unpcTtanE that, oa far pnoifble, everything should be rccot-
ded whihi It Wa5 fresh in memory.*
The fruits of this imrpmachablc vysrem are shown in the classic sections
aenm the Dorset dykes or the ditches of Wor Barrow (Fig. 2)1 wheiC
every obicct found L& proicoicd carefully on to ^tradhed acetiom M a
fiuhioD which has stood the test of con.^tant hade-refeeJace by Cworf
generations of arehanologislx,
1 pass on to 1904, four yearx after the Geoerai's deaih^ In that year,
Flindm Prtijci whose genius will long outlive the occasional gibes of
his suoccssDTs, doorihed hti aims and methods in o dassic cnonugraph
' Excotatwni trt i, pp+ xviiir
* Tbid., ii (j8^)| pp. 3dv*
M
HISTORICAL
which is itself a curiosity. Typical, for example, both of the man and of
the age in which he worked is his advocacy of the payment of labourers
on a piece-work basis rather than by the day. ‘Working by the piece
saves all this trouble [i.e. of constant supervision], and, if the men are
well-trained, and the work is simple, it goes on automatically and takes
the smallest amount of attention. In detached small sites men may even
be left unvisited for two or three days, merely reporting each evening
how they have worked.** My pen melts as I transcribe those words. The
almost complete absence of measured sections from Petrie’s reports is
the inevitable corollary of his ‘method*. His great problem was to keep
his labourers at work. An air of vigilant surprises had to be sustained. He
devised sunken approaches to the scene of operations, so that he could
come upon his diggers imawares and catch them out; and he supple¬
mented this device by long-range snooping through a telescope, with
results which he triumphandy retails.* The same spirit, with an alterna¬
tive remedy, survived a quarter of a century later in a manual on field¬
work issued in 1929 by no less exalted a body than the French Pre¬
historic Society. There also it is the congenital ‘dishonesty* of the work¬
man rather than the scientific need for constant skilled supervision that
constitutes the major problem; but the remedy, printed in italics, and
praiseworthy enough in itself, is: The best way to ensure the honesty of
your workmen is not to leave them a minute? That is by the way; to return
to Petrie, it is only fair to add that I knew him well, and, like all who
knew him, profoundly admired his untiring search for truth by such
means as he understood. But it is abundantly apparent that, between the
technical standards of Petrie and those of his older contemp)orary Pitt
Rivers, there ya>;vTied a gulf into which two generations of Near Eastern
archaeologists have in faa plunged to destruction. Petrie worked for
more than the normal span and with more than the normal energy in a
particularly spcaacular field. His pupils were legion and his hold upon
them was manifested in an unquestioning fidelity that was sufficiently
intelligible and creditable but was itself a bar to progress. When I last
visited him, on his deathbed in Jerusalem at the beginning of 1942, his
restless brain was still hovering over a multitude of problems and possi¬
bilities which extended the smaller minds of his listeners, and I left
* W. M. Flinders Petrie, Methods and Aims in Archaeology^ p. 29.

* Ibid., p. 28.
> Manuel de recherches prihistoriques, published by the Sod6t6 Prihistorique
Franûse (Paris, 1929), p. 23.
15
HISTORICAL
him for ibc l^t liinc vi-itb a itnwcd scnsfi; of thai dftvodon which he
inspired in the hEans of his pupils aiiid frieods. It is [ilmosi wiih a fee ling
of guik ihax I now, after ponsiticniblc cxpeticnc^ of his work and of ihc
iradirion which he established widely over the Eastji lind myself com¬
pelled to deplore an inducnei: which in much of its technique so long
outlawed iia sciaidftc usefulness r
Thirty ycant after E^ctrie^s ingenuous self-KVelatioiij another archaeo¬
logist, arguing frum expeiiciict In Fhlestlftc (where taore sins have
probably bccai committed in the Ckojne of oichflCOlogy thou on any com-
mcnsuiatc pnrtiun of the edith^s surface)^ could still write as follows:
[the foreman] receives general instrucriofts from the direcior for each day'i
vnhTkf pidts men tor spccml tanksi,«* ices to jt thal regulations are carried
ouit . . . [and] usually sianrLs on some high ppuic from which he can
uvcrsce the cstcavatwns,, ^ - O/ i> trmtd be wmiufsr a dirccofr
JO Itiivi him ar cimetri cf i0 iftnV hfign una ihmr
unikrsJandifig fimfuidi it michatticai*^ | italics mine].
jlliC ghmmerings of con-^dcnce are visible in thm scdtcn^^j but the old
sin shows through. There iic Bdll no rcail understanding of Lhc primary
principle of at! ntcavahuni that no shovdfid of curlh sliall be cut save
tinder direct and skilled supKivision. And those who have witnessed
Falcstinian eTCCîvation with a ciideal eye know all too well how wide¬
spread and enduring has been the leclmical Irresponsibility of much
of its direcdDn throughout an active half-century.
hex us transfer uuT dudooajy tale from ptecepL to practiis i not to stir
more tnud but to poiiu the Itsson^ however uegalive it be. The sections
(from I'eJI el Aj jtll in Palestine) nrproductdiiiFig. ^ represent a long and
hard-worn tradition which dies hard and stilJ awâits a emip dc
lliey were drawn in 193S and published in 1915^ ^ they may be
regarded as of compaiadvely nuHiem date. NeverihelesSj ih<^ betoi^
icchnically to the infancy of archaeoloigy and wetc, in foetj obsolete mon:
than a centuty ago+ Regard the absence ofassodaiedstraiai the omission
even of symbolicnl lev'cLs^ so that the walk arc suspended in section as in
a vacuurn. And that this suit uf tcclmiai] atavism is stiil intemutiudal in
ii^ vogue is thooiti by a seedEta from another Palestinian site (fiethd),
reptoditccd here from an authoritative American cummentary publishcrd
in 1939 (Fig^ 4)- Better far is a section through one of the mounds of
Ji Sialkj in central lon^ published in 193^ (Fig. 5): yet] evTn here, obscr^
'' W. ¥. fiadCj. #4! iManuaJ ^ Exc£Tvatii>p[ rn the Nt^n' Faijr (Uoiv, of CaliTOnul
Prcuh iwip aa-
16
Fig. 3. Part of a section through Tell el Ajjiil> Palestine, 1938
HlSTOiltCAL
vijtion and tecord arc smiW, strata are uiilabcllcd ^ are left mcDm-
pjtre and in mid'Sir, deiaiJed reconatniction of the picture is unpoisible
tM The accurate praietTioft of finds un to ifie seema would be jmpiac-
Ucahlccvea were ific necessary data tabulated, Rw for dowjuiet’t
nical incDinpettDce, the laigest archaeological dqiarnncni in tie world
old ArtJiJ«tHo«ical Survey of India-was unbearable. Here Hic
nriinaty bkune rcsia not with the IndLaas, who are quick and ready to
leam, but with rhe sutcessivc Uriiisb stait rwpoasiblc foi tlw establish-
nicnt and initial ninninjt of the dcpaitmeiit. It i* abnust beyohd belie
FtC. 4- JkoicKi ihivy^ part of the site Bethel j PsiJcitiiic* 133^
I that B3 feechtly as 1940 the Survey cmii publish ia itKsniimcntfll furm

1 suCh as those here iUustrfttied (Fip- 6 aetd 7): the one showing
walls suspended, Jihe those of Bethel* in a featuielcss pro&le of ihc site,
with neither btiilding-Liiics cor wcupadon smta, Mitied only by indita-
tions of the completely immeanlng piles of earth on which the excavator
left scpdie of his walls itamhcgi the other showing the burials of two
^'ariimi CUlluftt Boating like a niLhei disorderly halinon-bamge^ without
him of the striTa and the giave-Hiics Which would have indicated their
idemilic inLcr-rclafiDnjhip. It is sad to compitre thcac caricaluies of
sdeoce with the admirable sktich^recotds of Meadows Taylor^ ne^y
d otJXiMJT^ etirlJcr.
This dreumstance in the mure remarkable in that the average sitandani
cf Bcld-archiM!ology in Great Bximin itself during the past balf-ceniury
has beco unsurpassed, if approached, by that of any Other country. T*hat
statement Is made wirh no irvoiiar prejudice. In HoUand* Dr. vnn Ciif|
fen and others have es'olvcd methods cf cacavaiiou which mark a new
KLandoid of finesse^ arid cerratn of the Genrum caiavaiDrs* notably Dr.
Gerhard Berstij need no commeevdarion from me. ^cc its inauguraaon
iB
19
Fig. 5. Section through part of the mound of Sialk, Iran, 1938
HISTORICAL
in tSw, ihc ricnHsm Lifiut Conomisswn, thougb oft™ tcduuQiIIy bdow

the Geoenl Pin Rivm »iaad«4 tas produced adminbly coôi^naicd
wirlc, pinicukrly voliuhlc in itt oiienijon to siOflll f^ds. Bat ifl the 'vc^
limited ana of the Bririih Isles, rich in reiMias of dêtenwted nil-
tmes, with the Getwral’s otamplc behind ihejn, fidd-archfl«lopsis
have worked withia doac migc of one jinolhei and undcf the fire of
coastani and even fierce mutual criticism . Bad work has been done, m
fflielv with impunity. RapcriciicC has heen leoday shared and a steady
ptogress assured, ftom the days of j. P. Buslie-Fo* in the first quarter
ri[T la I {I 'p .
li lim-r'i"--“ X. ^
Fig. 6. ScctiMi part of HaroppA, Fakistmij

of the century lo those of Ian Richmond and Grahamc Clark in the
second.
In the face of this adutn aaent at home, what has gone wnu^ snth
Eastem or NearEasiem Bcld'afchaeology ? The question is worth asking
if a remedy is to be found. The awwer is not in fact ditficult, and the
remedy follows. The error hos lain not in quantity but in quality. In the
first place, few cMavators who have gone East liave in the past received
adequate pieliminary Lmlniog under dosdy critical obstTvatiun iUCb
as has been noted above fiom Britain. They have generally epm East as
dass-room or museum orientalists, with tbc bare tnowlcdgc of an
indent language or of naoie-or-less classified exhibits and diêcia but
widiout practiml cipetiencc of the field-problcni. And, once there, they
a« generally lost. An excavator may work for years on 8 remote Asiatic
site withoui'bcing able to discuss his methods and results cti tht ipot with
competent ciirics. In the second place, Eastem excavation has in the
past tended to atrnict relatively llbexol endowment, dthei thiotigh the
natural lure of a Biblical context, or of associatioii with famous and im¬
pressive dviliaiions, nr of the geneml ‘romance’ nf the Orient, or even
of the East as a vrinter-tnurist objective to wealthy Western benefactors.
20
HISTD1MCAL
TM$ libcnit cndnwTDcnt, cDuplol with the rektivdy cb^p cost of native
(jabour^ has encouraged wholcsdc mass-ôavutiOO^ rewarded by emcn-
Sive building'plaiw and ample hods whidi gratify the patton bur are far
beyond the capadry of anything approaching CKOCt record. Indeed,
Fig, 7. Seetton ihraugh part of a ttni««r n Huappa> FaXiiton, t«o

chop Diiaiml labour has in the pait been tt consrant snaie (essmi^
PI. IVA). ‘The moidiniiin nurnher of labourers employed at soy om time
f was somethiMR over tbinecn humbtxl', states the report of the oicava-
. tion (in 1912-13) of aa outstandingly itaportoni Rastem site whete only
lone supervisor was present. In the tUrd plaiX, the commcin use of mud'
bricl; for building and the impact of exircne wcathcr-oonditioiis have
often (though not always) combined to deepen the stmts nn an Eastern
site, 90 that tfadr ovei-aJl depth may be at least five tuno as great as
on a closely IniErlea^-ed British site. It is understandable that these deep
21
HISTORICA I-
d«.. ..1
with a thick tiKUMlft of uadilfcimtiated ^dl Wrttnnal ^

fer pwncrialm built and nny wiifieially mtcnaiX and level it- _
stated th»t tbereu often nn.pptedahlc .t^nfiesuonon a
sun-h1ead.ed, mor««>n^wepf EnsterU she. Tl,e ' iJ.ii
true, There if Hfl ^«etkod h> the iMoeatiCf ef a^luh Û v.hi
i, mt WirnW.'-oî t>e appUe<!-t^ n n/f A^c A^
TherLedy ihetefotc is thar the oriental fidd-wnitef shall, as
m of roulinc, have a prehminary and thonnjj;U E«>uti^g m the W ttl.
whete eriiicd control is at hand, and ^here straw ate liable to he more
condsc and so to provide a more cmcentnitcd ”r“«ice thanjm
liastera siw can be expected to fumish. *l here .s scar^ty anything, be «
reraicd, in the work of Pitt Rivets or his accredited Bnusb succirswrs
that ia not relevant to the encavaiion of a ult in Turkestan or a tom ill
Svriii Ii is the more asronishbs that those ilandards have so rarely
pîed to the further shores of the Meditmancanj cast of Suez, they
vfcre until very iceent yean almoit tinkiiowii.
li will thus, 1 hope, be appredated that these remarks th^y veil an
appeal—an appeal above all id the youageT gmerationof archMoIogisw-
1 h^fiomcspcrience,r«oiitmimdedd5«wlTCro'tlicuSCOfRn^
as a suitable basic training-gruund for excavators of all smd «•
nous, by reason ofits stTuctuiulvoriety.ini ahunOuiil stradjjruphy, ns pro-
durtiveness, and its ready avaikbiUtv. There are many aUcmauv« but
the principle is the same: always that of precise and detailed iraining.
We muCT give our reemits their Aldershot disdpline before sendhig them
out on to the baitlcbetds nf the world. In ibe past we have been unwisely
casual iJi This mattcr-
In the prcseal diflpicr oji attempt has been iruidc by way of iniToduc-
don TO do t«Tj things: fint to sketch, vety iûmmarily, the main phases of
technical devdoptnent dunng the past cetinifTy and scoioslly TO hint at a
sustained and danffMOUS dcavagc in technicaJ smndards between Eait
smd WesT^ This dciivagc will be reasasidcied at a bter stage; ttieaawhik,
Ln the next two chapters something of u mote posiri™ and constructive
kind must be ^d of ihc ordering of our ficld’eiHdcna,and of its iutcr-
pKtatiDn Ln lems of on absotee or [dative chronology.
' ' The ArclucDki^y of the Roman FcorincCit fttid Beyxmd/ Fifih Anms!
pflPt cf tM Ltfndim Utav^rnty [mtiiuU p/ Ardidîoe}p\ 194^.
12
Ill
Chronology
I T is a iniiiin ihat tlie b[»Cfcboiic of biiWty a an agreed chranology.
Yet it soiaetiines scans tbai loday chionoln^ of ihe old iHpd ty^ a
a trifle out of feuhlon in the routine of historical studies. ' I'o an older
BMeraiioD ibe matheimiics of the business were iniporliiJit;rlic
hotise could only be unlodced by anabinJtions of integer paiîiHy
remerobcied often long after the treasure itself had passed into lunbo.
Now that is In some raeusuie ebangp], doubtless ou the whole for
better. Aichaeolofy. mii ss a nite dogged with ovcûch antbacUC,
may be in part responsible ftir the shifiing btw. It has begun to petcolaw
through our universities to our schools, ond sometimes fon^ôcuous
pools of somewhat oolourlcss knnwledge—mostly D rehned Dawinism
-in which OUT tdndcfgarteret me encouraged to paddte. Ships and
sooUng-wax are bcgbuiing re tank almost with king^tbrelcss everyday
things with calendared State rectm. The recentress oftlm«netg^ce nf
mchaeological as distinct ftoin historical edueauon Is hard to icahie on-
tm we recall, for ™mpk, that only 0 generauon ago ti was îbte re
take 3 First ta Greats at Oitfonl uilhoui wen glancmg at a raithenon
„ m A»ic v,«, Toi.,. «!>»> .h= J"f;r "
history has been supplemented by broadly bssed culttir^ s^cs. it
rf,taok IX. ™ the n.M that ia«

and ultmiate and unrdeniing imponanec. And by da«» *
simply those nebulous phases arid setjuoices, ih™ date-substîis,
with which archaeologists oftoi enough tiy to bluff us. I mean time m
hjard ficurw. I mam Bradshaw. , . t ^
l-hc^S for re-establishing the rflaiivc sequence of ancient cultures
or culretal episodes, if wt are re begin to undetsi^d ihcir
und values, is sdf-evideni, and strangiaphical excavation,
mote will be said in the next diapter, is a pnmacy met^ "
The old comparison between the succaasavc strare lo ^ ^ ™
recessiveloivesmaboolt holds good: and is
the faa that displacement in either case mvoh^s
confuse and fmsrrate and, in mccavarion, c^t be
mailer I shaU recur. But een the most careful maiDtenitnce of mrteci
^3
CHRONOLOGY
sucotssion flnd ibc poiiiiliddn of a caavIndflE OJlnire-sequmCC arc dot

in tlicmsthYS enough. And for iwn itawni. First, without an absolute
chronology culruna of different tEgious cantJOt accurately be compared*
their intenelarioiwhip cannot be assessed: in other words, the vital causa¬
tive factors of human ‘ptofiicss* cannot be authoritadvely PcconslruclOd,
and may be widely misiindcisiood. SDcoiully]i the ducttuiiing tempo of
human flchievtmcnt—iiidf on integral quality of that achicvemtaal—
c-aunot be csrimatcd: tlic lighmirg flash, for ciiunple, of Pcnelaiu
Athens, or rhe glow of the slow-moving riverine civiliaaiions. It is im-
ponant but not ennugh to know that in the tweiitielb ceocuty a.d. an
aeroplaoe flew from London to Singapore. It ia ulmost equally impot-
lant* in Our Miimatc of human achievement, to know that in 1950 the
aeroplane took ja hours for the joumey, and in 1999 only SO minutes. Do
not 1« us fot^ the sigmlGoince of tempo; and that implies a dme-wble
in the literal sen^ ncfLIuCkg less.
This qiac-TOon of tempo is worth a fiinhcr mnmcnt'ii ihougbl. Il is
least ds dwenang of the attendots Of the archaeotng:ist who is concetoed
with the cvalutinn of hiimaa mstituMoas as it is of ihe hiologisi who
Concerned with the Rirmdl evolmion cf naniral apedes and senm, Ifl
the tariguage of th^^ biolo^psl, phases of ‘incfe3sc% phases—HispnrsSive
term—ofÊsplcKsivc^ c^’ulnuoiij ^staiiociatryôi'decliiie* phasesj the w hole
rhythm of human as of âinoAl life is itself ^ absorbbg pbenomenonj
full of mcomi^, oficii fyU of I happen recently to have beest
making Mime slight Study cf one OT the great civilizatioiii of the aiu-ka:
Easti the Indus VitUey OvOmiiciti of the third and second millaiciia
n.c, in what used to be called India but is now Potkistanj andj though it is
true that at present wc kacw ad too little about the begtuniiigs: of that
dvilizadoDi the indicadons sjc that it flowered wiUi m aknost dramanc
cclcriiyi the sudden oHspriug of opportunity and genius. It Was preceded
by a misodlany of upland vilLoges and oiltutes, set in ihe inmulnious
castciTi borderland of \1k great Iraniau platrau. Such grouped
and coniined b^' piedpitous ranges, ptutide the optimuin conditions for
the earlier essa>'s in cummuiiai Life within the boundaries of a parochial
sclf-sulfideney. But for wider politicai horizons wi der geographical hod'^
zuus are a prior coudidou. Teutadve movesIkim the highland zone down
on to the great riveHne plain must have drawn an immediaie and com-
peliing ehalletige. On the plalu, the dangûs annual flood can only be
constrained or niihzcd by combined clfon on a large scale. On the
pliiin^ fertile soil resuitiog from that iâme flood is a$ abundaot as mineral
CHRONOLOGY
and other resources are scarce. With the added urgency of one cause and
another, the river itself and its flanking lowlands facilitate and stimulate
traffic, commercial or military, and at once enlarge human relations far
beyond the precedent of the upland valley. The opportunities and diffi¬
culties implicit in civilization, in the full sense of the term, are at once
present and insistent. The remaining postulate is that of a creative
imagination sufficient to grasp the occasion. And without that crea¬
tive imagination, no stretch of time could provide a substitute. A
phase of increase, amounting even to ‘explosive’ evolution, may be
assumed. , .
Many other significant examples of rapid evolution might readily be
dted: an obvious one is the invention of the great windows, barred with
a simpUfied perpe^fficular grid of tracery, wherewith the medieval
builders illuminat^ w dim interiors of English churches in the latter
part of th^fourttenth Antury—sudden and triumph^t answer to an
insistent problem of enêering towards which earlier builders had
merely groped. And on thk other hand the patient ‘stttionary’ evolunon
of Byzantine art is sufficiently familiar and significant to need no elabora¬
tion in the present context.Kpeed is a mighty faaor in our evaluation of
human achievement, and it te a mere truism to affirm that our appreaa-
tion of speed is contingent mion a nice chronology.
How then are we archaeologists, we fiimblers in the earth, to attain
this difficult degree of predsiJffi? For something ê one-hundredth of
the vast period with which mMcm archaeology is concerned, the his¬
torians have given us a frarnewmk in the Old World. Let us be grated
to them and take them into the fkest partnership. In a region which has
a history, the archaeologist must know the framework of that history
from A to Z (not merely the miserable fraction of it in which he happens
to be interested) before ever he enters the field. The day is gone by
when the Egyptologist, intent upon the Pharaonic period, nwy cast ffie
Greek and the Roman, the Arab and the Chinaman, carelessly upon his
tip-heap. But that is by the way. The important thing is that the archaeo¬
logist must know his dates and how to use them: rerarded dates where
they are valid, and imwritten dates where geological or physical or
chemical or botanical science can win them from the earth. And year
by year the objective sciences are coming more and more to the rescue
of that subjective science, the study of man.
Chronology was once a simple enough matter. In a notorious pro¬
nouncement, Archbishop Ussher, properly styled ad mraculum doctus,
25
CHEONOLOGV
afiinncd dial ihf world WM maicd m dio year b.C. A mo^^^ sceptical
kjji beta coniem^ unril quite recenUfj to pmtukie 4241 5X. as die
earliest oileadsT-date. It vi-sr bcUcved^sicid die Textbooks sliil have it—
that the EgyplJaiis^ having ob&m cd the approximate coLuddenoe of the
roippciirance Quit before dawn^ of Sirios or Sothis, die Dog Star> after
a period of iavisibiliTy, wirh the begjamng of the Kile floods choso the
date oa which this phenDmcaon occurred in ihnt year fjuly 19lh Of the
Iiilim Cêodar) as ihcir ralcndncal Kew Year's Day- The initial date
was infened backwards from Ji-O. when the synLhiDnkalion again
occurred, by o logical ctiaiputauoa which wâs generally accepted. The
Eg>T>tiaii calKidar^ rctxigniziiijf the inoompanbiliry of the lunar months
w^th the $oIar year, divided the latter into artificial calcjiihii-monihs each
of thirty doys^ and added five fffiLSl-dai-S in an anonpt to make up the
inquired tODl. 'fh^c inxerailary periods^ howevc^:^ s(dl fell short of the
Sodtic ycar (5ippTf>xirnatdy the sitme lengdi as the solar year) by one day
in four yean Whence our Leap Year}; with the result that the sîiclironi-
zation was oraa only once in 1,460 years. Bur it seefned likely that the
calendar was already in used in the time of the Pyramid-builderi of the
Lgyptisn I Yth dynasty, who were computed on ihe basis nf native annuk
to have lived before 1775 B.c,; which took the calendar back to an iniiiid
Snibtc synchronism not later than 27S1 B.C find more probably not Later
than 4241 K.c. So much fur the lojig-«tabli$hed view% In recent ycjits
this view has been modified^ on cidtuia] grounds (the absence of writing)
so Esuly a date ns 4241 is now regarded as impoBsible. It iR now siigucd
that, although the Egyptkns at an early period were able hy their ob¬
servations of the hclujcul rising of SothI$ to check the position of their
365'ddy cslciuiarlti the Solar year, they pwêt had a Soihic calendar. But
on the gcneniliy accepted assumption that the ^ti^-day calendar operated
cominuoualy Iknn early Dynastic times, cotmbined with rhe onnrem-
porary iccords which have survived of hehficaJ risings of Sothis in terms
of that calendar and the icngihs of reigns given in ancient King Lists,
Egyptian hiitor)^—the oldest history in the wofld—caii $till be carried
back ooatinudusly, if not to 4241 B.C., at least to the Ia$t ccfitutics of the
foiutb millcnmuni B.tL
It would be irrclcnmt here to develop the far-reachmg Loiplications of
thh Egyptian calendar. It iindcfEes the whole of our prexlasslcfil chron¬
ology in HO far as that ^tand^ on a hhtoiieal banb. Slightly modified, we
still use it. But it is mure to our purpose to la^mider parallel and supple^
mentary cnethods of jadculation, and, by wsy of tranHitioOf refoieuce mBy
26
fir^t bn luad-c la an in^cruc^m theory proptFiinikd in by Dr. Ckiude
Schaeffer.*
Ax Ras ^hamra^ on the coast of Syrinr Dr. Schaefler has for many
years rtcavaicd the BmrLÊ Age metropulls uf Ugarit;^ and lias there
detenmucd five main micassivc layers. Of these the lastj joiajrking
the end of the Late Bronze Age^ shows evideoL'e of destruction by a
violent emthquiikc which dijd-ocatcd the buildings at gjoiind-lcvel. Chn-
teoiponuy site in SyTia> FaleBtinc, and Asia Minor have yielded isifnitar
evidence of violence; on dating now superseded^ the taUen wnllx of
Jericho were witness tn it; and Dr. Schiieircr ascribes the widespread
disaster to c, 1365 BiC, whens aeconling to a Tell el Amsma letter,
'Ugaiit has been dcsrrcjyed by firc^ half the dty has been bumi, the mher
half has ^âsed to be\ If this basis be acccptcdâd it is not perhaps
quite as clear a$ SJebaeffer maintainii'—a definite phase in all ihe dries in
quesiioo is dated with an indgEufitaini oiargiia of error. Nor need this
line of research end there* Another camstrophic canhqriahe seiana to
have centred npoti Asia Minor at the beginiung of the Middle Bronze
AgCj between ZlOO and 1000 a-c., and may be thought to tic together the
whole or partial dfrsmiciion of Troy and Tonus* Alaja HnyOk un the
Anatolian plateau* Chagar Bazar and Td BiOk, Tepe Gawra north-east
of Mosul, and a seri« of Syro-FalerLiac dti«, once mnn: induditig
UgaHi iiself. And yet other earthquakesj between 2400 and 2300 B.C.
and about 1730 B.C*, provide fiirthcr dttoiiologicil bondfi. It all scmii
too good to be true* but is not on that gfotind false. SuhadTcr was not
indeed the first to recognize something of the archacalogical poicntialiiy
of carthqualtes. In 1916 Sir Arthur Evans wrote a description of a
violent eaithqnake shock whidi he had just c^penented al Knossos In
Crcie* and added:
The archaenloglcaJ inquifur of this is very impartiint- When in the gnai
Palace of Kn0550S» we find evidence of a series of overthrows* some of
them on a aoilc that could h^nUy be the work of man, thm accnH real
reason for Tracing tlie caurc to the same seismic Agencies that we have
certainly to deal with in ihe case described alwve. It may be possible even
to fir approximately tite dnlc of seven earthquakes four of them of
severity, between the bat century of the third millennium, and the bepn-
ning of the 14th century DuCl^
^ <1 F. A. Schaeffer, Scrai^pAtV o>mparU rr eAriJunlcieiV dp rAn* uiîdnntak

(lir CE mjUiruiifes).
> CUctl by Join Tiftt iusi CAdince {tiOitiwfij *W3Jf P* 3®*.
*7
CHRONOLOGY
Therein lies a germ of the Schaeffer scheme, though not the scheme
itself. The general suggestion is indeed a serious one, and archaeology
may take a cautious cognizance of it, with the neutralizing proviso that
the earthquake synthesis is used only where other evidence has already
established contemporaneity. In other words, we must in faa regret¬
fully admit that earthquakes, in view of our limited knowledge of their
incidence and frequency in ancient times, are a shaky basis on which to
build a precise chronology. But the theory is of interest.
In another context, geology has given us a surer and now very familiar
footing. The recognition of the clearly stratified varved or laminated
clays of Sweden (and elsewhere) as the annual deposits of the retreating
ice, and as the time-table therefore of a related human phase, is now
notorious. By counting these annual deposits in a series of sections from
the south of Sweden to recent deposits in north-central Sweden, the
Swedish geologist de Geer and some of his colleagues calculated 6839
B.c. as the beginning of the p>ost-glacial period in their country and
as the beginning, therefore, of potential human life there.* Others prefer
a slightly earher date, but the principle is the same. On this basis, a
series of changes in the Scandinavian coast-line, climate and vegeta¬
tion have been given approximate dates which can now claim a value
approaching the absolute and are fundamental for the mesolithic of
northern and north-western Europe.* Incidentally, it follows that a pro¬
portionate chronological value is thereby attached to the identification
of pollen and the reconstruction of vegetation on mesolithic sites.*
Geological stratification further plays an important part in the chrono¬
logy of the palaeolithic cultures. Artifacts belonging to the Old Stone
Age arc frequently found in the gravels of river terraces, in brick-earths,
in cave sediments and so forth, under conditions which enable the
Pleistocene geologist to determine the age of these artifaas relative to the
sequence of climate-fluctuations that make up what is commonly called
the Quaternary Ice Age. In particular the \\ind-bome loess deposits and
solifluction layers formed in the cold phases and the fossil soils resulting
from the weathering of ancient land-surfaces have contributed to the
establishment of a relative chronology in an area ranging from northern
France through Germany, Austria, Bohemia, and Hungary to the
Ukraine. But there is more in it than that; for the attempt has repeatedly
* Summary by F. E. Zeuner, Dating the Past (2nd ed., London, 1950), pp. 20 ff.
* Ibid., pp. 46 ff.
^ Ibid., pp. 56 ff. and bibliography.
28
CHRONOLOGY
been nude lo crnivcrt iHs relalrte into an absoluie lime-iwalc. *TiniC'

gauges' are prmidicd by the rJies of WEatlieniig) of denudatiDii nod
of sedimentiidon) caJtulaicd upua such evidence as ihai pruii'idcJ by
vurved days and, more rcDcmly, for restiicwd periods by C14 (see be¬
low); so thatj if ihe gcoIogiLail icadlUr obtained by Penefe and otbers ore
applied to the sequence of paJacDliihic Industries, the Abbcsillian would
appear to b< abooi Haifa miUion years old, the middle AchEuliaa about
a quarter ofa nrillion,and the Moustetian about a h undred t housand years*
^Tien we leave geological stiaiiliciitioii and mm 10 atiatilicaiion de¬
rived mainly from human occupatioii, inference is teas certain, li is tm-
possible to lay down any Uw for the equalioti of man-made suata with
an absolute time-scale. For cianiplc, at a site (Chandravalli) which 1
excavated La India, coins that wen: not, apparently, earlier chan SO B.c)
or much later Uum A.n, soq ranged ihiou^ a vertical aucumulatiOfi of
5 feet; the period thus represented sras probably in fact not more ihiti
two ccntuiies. In whai is now Pakiitan, at the fanraua site of Taiila
1 (Sirkap), excavations in 1944-5 indicated that ft-9 floors and
debris wrae deposited during some two centuries of very intensive ocoi-
paiioD. In an earlier phase of the same diy, 'I'rmla (Biai Mound), an
untidy shc^ 14-IS feet were ascribed to three centuries or a little ronre;
hut the masonry and building-methods were here of so unstable 0 char-
Bcicr that the aocumularion may well have been exceptionality rapid- In
every instance a multitude of unfenown and variable factors is ittvolvcd,
and abjective calculation nn tbe bails of depth b> sirtuflUy impossible,
nevertheless, fortified by the !;easoiuLl regu^ty of the Nile ^cy,
Petrie was greatly daring. ^Cnwrolly', he maintained, if i* possible to
date the latest date of a town fay the potsherds lying on the surface; and
to allow a niie of growth of ao inches a ccnmty down to the visible level;
if that pves a long period we may further carry down the certainly arti¬
ficial level by 4 inches in a ocniuiy for the Mile deposits when in culti¬
vated ground’—and so on.' Such calculaiious have, if ony, a piirdy
aaidemic or abstract iniereat. They make no allowance for the intermtl-
tendca and vagaries whidi, alike in human and in geolopcsl history,
defy the uonfincs of mathsnatieal formulae.
Only DC tote occasions can a cbTOnolngioiI coonoiarion in terms of
calendar-years be ttttribuicd witlt plautihility to man-made strata. 1
will give two examplK. My first rames from Scarborough in Yorfcthire
where, during the excavation of ptefaisioric pits of the fifth ceniuiy B.c.
' Jtuu ami AUihodf in TP- lo-ii.
29
CRROKOLOCY
lUidcdyLng die Roeimii sigool-iiiatiuii, Mr. F. G, SiiBp&oiij mostlj

scrupulou$ of excai'atjors^ obeyed thut die hufflua dcbm—sherds^
bOELCS^ Ûm &c.—In Lhe pits was interJeaved widi hiytts of eleun SijiJ
[FI. llA), He furdiet observed^ on retiiming to the siie after a w-inter^s
absence^ diat m Identical dean layer had acciuniilaied in iJie rc-cixeu-
^^ted pits 0$ a result of die rains und of the pm.i±di£i|t winter
months. From this observatioiit infaxnees were dedudblc: firsts
that each of the pits had only been used by the prchiiiLene mhabitunts
for some threo or four years (represented by ilirec suDcessive ancient
rainwash layers with interposed occupsition^biis); aDd, scx^DDdly^ that
lhe liiic^ OD n stormy headlaiid oycriooldpg ihc North Sea, had been
abandoned during itie W'lnter months, when the deposition of occupa¬
tion-debris hod ceascd> and that die place ivas used only as a mimncr
Btadon.^ Both these infeneoces are of importance in the chronological
and sodolngical evalo&don of the setdement.
Aly second example comes from Iraq. At Kliafhjah, in rhe DiyaJa
region noirh of Baghdad^ a temple dedicated to the moGn-god Sin wok
built and rebuilt ten tkees on the some site in the foiuth and Third mil¬
lennia, and an ekbomte attempt wai made to înstruct an absolute
dimnolni^' for the building^ and hence for its dynastic or culninl baefc-
gruund, upon ecnain stmctnnd evidences. It h out necessary here lo re¬
produce the argumcot in aU its more theoteuLÎ ramiftcanonsj." but its
root is of iniercsip Tlic fie^'cnth Sfo temple showed two phases, with two
door-Jcvels for the second (laiet) phase. Between these two secundary
fioor-lcvds, the tcmplowan had been mud-pIastcred sixiccD times, each
plaster coat being applied to the surface of its predcenisor (PL I In).
Nowj today tt is the cusinm in the Near East to plaster the surface of
adobe (or mud) walls, sndi as those of the temple, cver^' summer in
prepaTation for the wmicr rains; and^ since dteinnstanccs hav^ nor
maieriaUy changed, it is a fair assumption that the sixteen ptasrerings
represent in fact a lapse of ^xccen years. But d uring that inferred sixteen
years ihe ground-le^'cl zt the entrance to the nmiplc nysc la cm.—^the
vertical distance between the two iloors. Applying this rario to the accU'*
□Tuiarion on the floor of the firm phase, namely 75 cm., w^c may calculate
• A, Rewnin!^, Huiffiy Hcmifcrougk (Londoiif 1911J, pp, ao and aij
<S.
* They wiU be found U% Pinlui^ Ddongaz uiil Seton Lltiyd, l^n-Sargiyiud
Tmplsi IH Jfcff Diytila /^qpoii (UcUT. of QUcqkO Oriem, Imt, FublkiluHw, Iviiis
PP. I3S ff. I mil mdebud in Profe-^jr , E, L. .Malldwwi for drmwme mv
atitfiuion to thii.
JO
CHRONOLOGY
ihc limt represented by thui acaunnlaiiuii ul ifi.- ]l = lOO yens. Ta

This we must add the 16 years already calculmcd for the inierrai bciwccn
the two Hoots of the later phase, and wc jjet a total of 116 ycara for tlw
time-inturva] between die fim and ta&t floors Of oiades-les'ds.of Sin
icniptE VII. There we will laaTC the (X9inputiill0i)» at a point beyond
which infeteace becomes inocasingly coDjectura]. Bui as far aa wc have
carried it, the argutncQi has considerable weight, and well rqireacnis the
type of evidence for which the cita.s'aiot in search of ptccUion may use*
fhlly IcEcp lui open eye.
From vertical sttVitiliraTion wc may luin for a moment tn what is in
effect hnrlMJttiiil stratification: a method of computation, simple in
theory but full of pitftlJs in practice, whidi has been disguised under the
term DcndrochfonOlogyr ^J'he printaplcs oB LJendtochronnliigy ho.S'obccn
widely advertised, and a convenient summary will be found in Ptefessnt
Zeuncr’s handbook.' It is a suffidcnily amiliar fiia that a section across
a tree grown In a climate with sciLsonfll variations reveals more or less
concentric growth-rinss, usually repiescntine annual aceretions, which
will differ witli ilie age of the tro: and the climate of the particular year.
In ycara of drought the grow-th will natutally be Ices than in wet years,
buT 3 Tendency for the rings to group in ii-year cydes, in conforouty
with the sun-spor cycle, has suggested iBiai solar radiation is a
further and Lndepcndcni factor. On the basis of the ring-sequence of
long-lived nees in Califcmia, Dr. A. E. Douglass and hi» associaiw were
able to work out dinutic curves Jbr that part of North America for the
last j.ooo years j but it is more imponanr to ns that variant fcatuns of
mdividual rings or their grouping can be plotted and compared from
long-lived trees to titnbos cut in the Hire region as the type^pcciinecs
in luuient dnies, Thus in Azunna ttee-ring analysis has been brou^t to
the rescue of American prehistoty with astonishing results. Here ê
correlation of the iree-rings of timbera from prcllistoric [i.e. prc-sLi-
leenih ccntuiy) Indian houses with daicd sequences enabled Douj^ss
ici dcteimiae, with a minute margin of error, dates at which the andent
limbers had been cut as far back as hi* maieml could take him, in fact
10 the eighth century *.□. and eariicr- Thete always remains, of course,
the impoudcrible factor of the date of oonsiructioo relative to the date at
wliich the component timbers were actually cut. Timbers were liable W
be reused from building to successive building, and a particuLir stric¬
ture may thus be as much as some centuries later than the tree-ring
• Daiiiyi rkr Pan, pi*. 4 d*-
CHRONOtOGY
dotting of out or jitore of irs Timbers wnulti uloac suggtÛ In orher Ti'ords^
Lrce-ring wuilysis can only pmvidc a /rnmnuj pcs£ qucfft,
Attompti Mve been made to apply the itee-iing method to cnaierial
froKi EurDpe* but so far with limired aldaough $onie value has
been daiitked oa iasecure evidence for trcc-naf^ graphs prepared in this
country ftom Romiin and tncdJcval timbers.' The absence of very
oacicat trees^ houneverj and the mueh giciitcr rtmoteness Of European
prehistory combine with a (irsimlly) less distnimiDUtuig dimate to mili¬
tate against success. Nor h-ive anempts to project Ainericoa data across
the Atiaadc (^telccoiinncjoii^) mci wirh vîdc aeueptancc. On the other
hand Africa aad Asia have nor yet been adequately exploited by deudro-
chronologists and may in future add new laurels to a technique which
has ceminly been dramadoilly jtueociifdul in the hind of its orLgui.
llie mendon of solar radiadon opens up nidc vistas in the computa¬
tion of geological rime and the daring of rcmorc men Ln thdr geological
semng. It is not for a laymin like mpdf to pretend to more thaji the
OKwr general understanding of the methods cmplayctL It is, however^
comforting to be lold that ancicni climatic ducniatioiis, deduced f^m
the examinadon of loess-deposits and the analysis of zivrr-gnivelsj apee
closely with cenain duenmnons of the intensiiy of solar ladiadon cal^
culered on an asironomicil ha-sis^ vîfh the corollary that therse fluctua-
tionsj with thdr related human phenomena^ arc approximately daieii
objectively. Thus it is calculated that in the open spaces of north¬
western Europe palaccilithic man began hia Sisyphean task some
yenrs ogo, without an>^ pedando liinrDition within }0,0DO or 40^000 years
on either side: a date which is fuiiud to be in excdleni agreenieni with
that which 1 have already meuduned as obtained Itom the rates of
wcatheting and denudadou.^ It inay therefore be regarded for some
time to come as siilDcicaily absolute in rekdon to an q>oeh when fashions
Id crdbmaDsIdp may have lasted for anything up tOj or even beyond, a
quartet of a milliofi years.
I meutiofT these cakulations^ merely to huny like the mc3snit>
ment of geologica] time on the premise that the time-T3tr at which the
disintepadou of a radioactive mmend proceeds is constant and detcr-
miuabki^ 1 accept them calmly with other man’cte of ihc age. But it is
* A. W. -G. Lowthcr, *I>codiTChiTjnotoffy‘| TArf Ardi. Nttm L£ntr (L4m4Dn)p
March 1^.
’ F. E- in JVdtf. IJwr Sm. t62 (a)^ p. 125.
* I rtfet the rcB^lcr to Zeuner^ Dating fAj Fan (and cd., l$SO), with
bibUcgnpby^
PLATE TI
A. All Farly In)n Arp pit at ScnitjurnuBl', YoikiturcT cut I* »Tk*w three
iriterltipicd pbaM'A of I bta fUlinE.
_ ^'.■•■>14; : J irfM-
B, Seettef Ihowt^g tajfer* cf the ^l\ tifthc
icnxptc af fiHKm-ijiDd Spci fft Iraq,
(Sh p. )
CHRONOLOGY
important to know what the gcochronologists want for their alchemy,

and to see that they get it. It is the duty of ±e field-archaeologist to sup¬
ply them, for example, with classified gravel and loess and with related
artifaas. They are gradually bringing order into our remoter prehistory,
transforming the jellyfish of its chronology into something venebrate.
Their work is integral with ours, both in the laboratory and in the field.
The marvels of science as applied to archaeology would not be fully
catalogued, however, without a reference to two developments of an¬
other kind. The first is the so-called Fluorine Test. Since 1844 it has
been recognized that buried bone absorbs an element known as fluorine
from the moisture of the sand or gravel in which it lies. The process is a
very slow one, and it varies quantitatively with the amount of fluorine
which happens to be present in the ground-water of a particular site.
This variability obviously robs fluorine-content of any value as an
absolute time-indicator. On the other hand, as Dr. Kenneth Oakley and
his colleagues have demonstrated, it may be of use within very wide
limits as an indicator of the approximate contemporaneity or otherwise
of bones tirom the same locality, and one notable instance of this utility
is already on record. A brief glance at it will suffice to indicate the poten¬
tialities of the method.*
The example relates to the much-discussed skeleton found in 1888 in
the Galley Hill (north) gravel-pit at Swanscombe in Kent, 8 feet below
the surface. The gravel in which the skeleton was found is of middle
pleistocene or early palaeolithic age, and controversy has raged as to
whether the skeleton, unmineralized and modem in character as it is,
was (say) a quarter of a million years old or of relatively modem date. If
the former it is obviously a remarkable proof of the antiquity of a modem
skeletal typ)e. The general problem has now been set at rest. A number of
fossil bones, including the famous Swanscombe skull, has been col¬
lected from the same early palaeolithic gravel, and others from later
palaeolithic deposits in neighbouring gravel-pits, together with some
from local recent deposits, including part of an Anglo-Saxon skeleton.
Analysis showed that the Swanscombe skull and other early palaeolithic
* See Kenneth Oakley, ‘Some AppUcations of the Fluorine Test’, The Archaeo-
logical News Lettery Nov.-Dee. I949> PP* *0* ff.; K. P. Oakley and M. F.
Montagu, ‘A Reconsideration of the Galley Hill Skeleton, Bulletin of the British
Museum (Nat. Hist.), i, no. 2 (i949); K. P. Oakley, ‘ReUtive DatingÂe PUt-
down Skull’, The Advancement of Science, vi, no. 24 (195°); K. P. Cêy and
C. Randall Hoskins, ‘New Evidence on the Antiquity of Piltdown Man’, Nature,
clxv (II Mar. 1950), 379.
D
6579 33
CHRONOÔljy
bonoi ExtRtaincd about 2 per oentf Aiionnc^ the laicr palaeolithic about
t per cent., and the recenr hnnoi o j m 0 05 per ocor^ The Galley IliH
skeleton was found to contain only n-j per cent., and its reladvely
modem dare is tbuJi apparent, li paaiies finally and inescapably out of the
textbooks into Umbo.
Similar fluorine tests have been applied to the much-discussed human
or simtan bones from the of PilidowTi in Sussex. The results^
confimircd whiLsi this book was in the press, are dramatic in the extreme k
The jaw and canine roorh, found subsequently to the skuU-cap^ are now
dcmonAitahly of mndein date, altered and coJouicd by a forger and
camfijlly 'pbmred^ for discovety by distinguished visitors. The cranium
found nearby by gravel-diggers, on the other hand, b gmuktc and of
tTpper Plehrocene daic. In itself it prcseois no special pmblem, but tlic
removal of all lemprarion lo associate it with the jaw removes also a
potential anomaly from the sufficiently complex problein of skull-
evolution. See the IMUim ^ rh^ Briiiift AUiSimm (Naiurat H(sit/Ty%
voL 2, no. 5 (tqy^).
At the best, bowes'er, the local and relative value of the fiuorint leti
must not be forgotTcn; ar the best, as Dr, Qakiey emphasize^j
it does not provide a mciru of reldrive dating. A given bone or gruup
of bctficA shows a Certain range in Guorliie culit^t. Unless the diifcrcnce in
age between the boiin which are being compared is considerable (e.g,
lOôoo yeEnp}j there is usually an overlap in the range of rheir fluonne-
contenta. For this reason ic would probably be impossible by this method
10 difTermTinlc clearly between say a Siwtow and it neolithic skdccon;
whereas it should enable one, for example^ to duiinguish bnn^ of neo¬
lithic or kter age from others of Acheulkn age, when botit occur under
almflar condlttons at the same locality.
Potentially more seiiimvc is a new' method of dating organic material
by mcana of its radlncarboa content. This method, a by-pfuducr of
Atomic research, was announced in 1949 from America and, though Rxill
in the cxpenmcnral stage, may wenrualiy eonble sped mens up to
20j000 years old id be dated objectively. The methodj known summarily
as the 'Cartxhn 14’ methcMi, haa been de^ribed by Professor F,
Libby, Dr. E, Cl Anderson^ and Dr. R. J, Arnold, of the Imtitutc of
Nuclear Studim Ln the Oniveniiy of Chicago, who have tested it upon
WDod-sompIca from Egyptian lombs/ I'hc possible ertor is at present
■ Harare, di, no. aSay, 137-I U Mar. 194^)1 Sa^rnt Todgy, v, no. 155
(14 Nki. 1949); Xi6iii (iwigh 113-14 "nd 3A9. See aho HaLkm L,
Mwjuv in iNil. wtiv (1930), 99-toi; repotr of Brit. Auoc. diveuision in Annert,
3^
CHRONOLOGY
apprtdable but will probably be reduced^ and h êcou likely that fur
the mesolithic at .any mte che uevr inethod^ if cmtde ûHdcady a[M;a-
sible^ wm prodi4ce tisefid rtsule-
The principle of the metbt^di if not lu praciJcei h simple and lugioil
citeugb and may be stated categorically oud very summarily:
(i) Cosmic my$j which oirlYC ftom Uic outer space» produce radio¬
active carbon atoms of atomic weight 14 m ±e atmosphere. The C]4
thus formed is an isotope of ordinary carbon of atomic weight tl {Ct2);
and bolJi arc contained in tlie carbon dioxide of the atmosphere in a
proportion wlikh is stable and cotiesponds with the rates of production
and disiniegraLioa of C14.
(ii) Now tlii^ carbon dioxide is taken in by plants^ and» Since odl
derive Lheir boJy-imterial uJtimaiely from plants^ it b univer¬
sally ineorpurated in living urg^c matter. TherefocCj the proportion of
Cj4 to Cia ID ail living uigamc maitct is the same as tn the atmosphere.
{iii) But onec on organistn is dead (e.g. when a tree iS Cut down)^ it
oeas^ to take up carbon frmn the atiCBOSpherc. On the contrary^ the C14
coQtEait riowdy diminishes^ Tcvcrting to nitrogen at Such a rate ih^r after
about 5]6ao yesjn {termed the ‘half-lLfd') only half the Original amount
of Cl 4 is left. After twice that pcriodi only half the residue—i.c. a
quarter of the original quantity—is lefi^ and so forth utuii all the Cl 4
has dUappejred.
(iv) In dead DT]îiic mauer^ thereforej the rario of Ct4 to C12 de¬
creases with time at a known rale. The proponion of C14 tn Cta in a
given oTgame spedmen on be deienmned in the laboratory, and firan
it the time dapsed since the 'death* of this organic maner can be calcu¬
lated.
Thai k the principle. The chief practical dlfhculry k that, at the bestj
the initial raiio of C14 to C12 is exceedingly small and tbemfore difficult
to measiJTE with precis iqr. Some organic siibsiancca arc indeed ida-
tlsxly caaier than others to deal widii such arc wood or charcnali which
contain rmieh carbon. Boncsj ou the other hondi enntain only small
amounts of carbon;! and it settus doubtful whether c^th very large
dK%i (4 Nev. 1S3^), 756 a. V. R Look, R E. Zeencr, and K- P. Omkh^y), F
ûncf in Sdma fVtJjfriif, M. 154 (fjonsloni Afinl ISH) PP- 0*5-^? C^th
biblJf^j5hy)sO. H. S. Buihnell ln^JtfiÔ'i ^ rissrii 14S-&; ami eipeei^y
'Radiecorbon Danug' wscmblnJ by Fr^ricfc JohJiHU in iHjwiti«h4
avil case for Awth., Salt Lafcc Cdy, Ucahi July 1951J, and F. Libby,
(UaI v. of diiict^r
35
CHRONOLOGY
quintitie£ of tlicm, virlth special ‘boostin^^' supetadded^ will m fact yield

tbe Ttecessafy coiiccnxration.' As a whole, die technique is still (1^52) in
the- experunenta] and a Dumber of iiutesolved dilEc'uliies remuid.
For exompk^ a spedmen of juidcnt chiirood may have been eoniamiii-
ated by the absorption of substances cutnamlng charccml uf younger
agcj a porticukrly doogciOLis source of coDtaminaLioD is fungal and
bactedld groTAth whlcb^ even when removed fmin the upedmeiij muy
have £ilsi£ed (reduced) its upparoii age. But, in spite of theoc and other
complicatlonii there is reason to hope that it may be possible to aehic^^c
an average accuracy of 4,200 years in the dodiig of chareoal and certain
other c^teôrie^ of orgimic material ranging in date from the late paJaeo-^
iithlc to the begkmiug of the piescnt era. ConMstent aecniacy vdthin
those limits nûld mark a fonnldable advance in the pursuit of pre-
hisioiic Chronology.
A number of examples of die application of CJ4 analysis have rvnw
been released- Thus it has been announeed thar the (iMankam)
advance of the Wisconsin gladatinn of North Amsnsa. pawd o ver rrec-
irunks the awrage age of which by the Carbon 14 metbe^ isabout 11,400
years from the present daj% Jess than half the age expected by gcologistar
Ii is shortly after that lime that iitun is bdie^cd 10 have hrsi appeared
in North America. The oldest ariifacis determined in Amcnen are
severaJ pairs of rope sandals coveted by voLi^nic deposits in Oit^gaiiT
the agp Lrudicamd was about %050 years, i,c, about 7000 b.c. In the
eacftcm United States, the earhe$t cultures are pcovlug so for to be
later than in the wesr. Giibon 14 dat^ Suggest lhat man spread there
undCT 5,000 years ago, bur much verilcaijon is needed. Other C14 dc-
tennmadons arc rcporfcd to confirm the btgihmng of the First Dynasty
of Egypt at about 3000 s.c, and that primitive agriculture was oheady
being practised in the Khurdish hills at about 4700 B.C—chaicod and
sheUs of Lmd-moiluscn from a vUbge^te excflvtitcd by Mr. R. J. Braid-
wood at Jiirtno in northern Iraq have given the remarkably cousisicnt
C14 dates of 4757 B.Ci 320 yean, 4654 bc ± 330 years, and 4743 B.C. ±
360 years, Nearer home. Dr. Grahame Chilli's already-famous village
of mesohihlc food-gsthcrcrs at Scamer in east Yorkshire secrus^ by Ltda
^ In order to obtain the qmintity cif pure curbon required for a saTiiriajctary
defcrmiruiimi hwd on acverul tcit nmitihc iniiurnuiii ctîanlitin arc
rvciimmctukd; S5 Kraminci of charcafi], mo ftrvnmn of other typci of vefzrahle
rcmami giaiiip ha^c-wnrki nod TOO gramixKi of ULolluâii ibciJi.
JJ. L. Movim in /tnli^Vii^, icriv 99~]DI.
3*
CHRONOLOGY
meihod of to Iwvc fknimhcd at aboui the a(xq>Uible dflte of

70DO iH.c But It is Mt to repeat the emphadc afBxnmiictu of Dr. liblw
and his assodates lhal oU such rwults are ai: prracni lentatlve.
In this chapter nothiog has been swid of typoio^- anabasis for chrono¬
logy, nor do I ptop«c [0 enter generaUy upon ihia thorny subject, llic
study of (he ifyiiicmallc dti clopmenr of types or fottus Vi'fls long normal
to bioiogicaj studicii before it beenme a roongnbed method in archaeo¬
logy j indeed it smedy entered into mix diKaphiM: undJ Fctne based the
relative chmiwsiogj^ of his famous Diospalia Pmn cemetori^ upon it in
1^1/ Its virtues its vices®^ now constitute u standard question in
orehaeologiCT] cummaiion-papersj and it is not necessary here to at-
teropt u model answer. But there are rwo points upon which a word may
be said. Aluch work—and much able work—hi^ been done La ihe Euro¬
pean and in the Asiade fidds upon the oDuipairadve typology of indus-
triesj particularly of pottery. In ihii valuable cndnivoLir may beobsm’cd
the recurrence of two sources of doubL The first is a tendency m give a
diagnostic significance m types of panem which are either msufftedendy
spedallzed or arc ai any rate repeated, without likelihood of contact,
in widely dtsparate dmEi und places. The second tendcnc}', not imic-
kled to Ihe first, is to devolve the comparative study of poncry into a
pursuic EEiore akin to philately than to archaeology: 10 compare linc-
illuatiutions of pauem with insuffidcni reference to other factori such as
paste, method of EmnuTacture, and firing. For a ad ahsurdum of
this abstract and theoretical procedure I may refer D^^3c mure to that
unfailing source of auiionory cxamplcsg PaJestinc- In a munuinentaJ
report on Tell en-Nasbeh, the dare of a senes of tombs has been eaJeu-
lated' by on asionbhiog statistical mcthcKL
AJl paraSleli in form to a pot-type tuive b«n tmeed, including Middle

Biouie ones lo a dearly Early lion pot CdiEfcrtnces in tcehiuqtie being
ignored). The mult, as a graph of frequency, 19 plotted for the tomb* in
combination wiih the frequency graphs of oihcr forms (from ihc iMe
tomb). TMs yields a aeries of which niore or Ices coincide, bur
givea the ludicrous rcsuli that the eatremes of the graph irneich Jran the
(Egypt Ejcplonitiijn Fund, 1901)^ pp^ 4 - -

* For ciEiinple, Ke O. ailTey, Caidi to the Cdoe Antiî«3 of dw OrtJiiun
Perwd ^p. 4; *1e ii noc m be suppoîcd thai irefirt nfprepenwx fc™* wwayi
«ttl«pcDd to . nf dll». After lh=
iiEpr jiuj tt-ijjpeiii thr whtJf wria f™n<ne * ^ rnnimatt fram which
the iiiti£n:r could dnrw.'
37
CMBONOLOQT
171I1 tD ilic ind centuries B,c.I On ihc other hnndt no HCwiiiit ii ta]ien of
the appcnrance of uunâppciiTance of diagnostic funns or techniques, .,,'
In other word*, u-e have here a audc csninple of the impact of iwo-
dimensioDfll diiijEncMis upon three^diraenaiDiul nuiKtidl
The other point uîeh may be i tressed is ihc fit* iofaertret in the
broadcast appliiation of typological oiicria which may to st maiority of
i Instances be pcifcctJy sound in themselves within their proper and re-
For caampie, tliose andtjuaries who arc aecusrntncd to
deal with ancient huilding-constniction are familiar enough with the
poTcndal time-values ofevolginestriictiiiail or ornamental fatiifw, ThuSj
studcoQ of medieval moulding;, ofandcntstyic$of masoniy, of ancient
briefeworit, of andem scaitcases, window-frames, roof-consmictions,
have evolved criteria of a rule-of-ihumh character which arc more often
correct than not, if auritnisly used. 'Rusticated’ masuaiy—ashlar with
rough siiffiice and drafted margins-^ a,i>. 1-50 in Rome/ Roman
brtclcs and brit^-joinia vary in absolute and tcliitive thidtness in accord¬
ance with a timescale/ Deeply involuted GoUiie nwuldings of the
thtrreenih ceaiuiy connaai vdth the plump mouldinga of the fourteenth
and the flatter mtiuldrags of the fifteenth ccutuiy. Queen-post roof-
tniascs h«gm in the time of Henry VIH And so on. But in aU these
rules lurk hidden daggeia. Thus outside Rome, in Palcadne for ejtaniplc,
lusticaied masoniy may be at least as eatly as foe ninth century b.c.;'
foe smooth chroDoli^ of Roman bricha has been established only for
Roim Itseff, and ^cie perhaps Only for cenain categories of building;
foe Gothic mouldings of Bricain do not apply to Gothic France; most
British queen-post toof-tiu»e$ are of the sixieenifa ocnraiy, bul they
occur octaaiflnalJy as early as the fourtetmth.' Taken alone, foerefrac,
almost am' one of these and companble miejia may trap the unwary
Them ûca sk liable to be locstl mther than univeraal, a„d must be
esublisbcd afresh and objccdvcly for every fresh locality. With that
proviso, they may be of great use, but the proviso is an important one.
wmmHirfng upon C, C.
I <iw7), with special trr«vi,« » pp. ^6-^^
O. T. Rivuira, lipwAt (ur« [Gltfofd, igivl o TO awrilm ii miw;
Shy JO period, bu, „o.« its tX^'
p. li'ald p'l^! (Han-ard, 19,4).
3B
CHFONOLOGY
So much, tHai, fot some of ihc priadpUiji Or pracrioea undcrlyiiig
aiCbflcolo^calchroDdOgy. Thdr importance bnwoiftsi; so loiichao that
it is perhaps advisable W cod on a note of w-arainff. We moit oat allow
chiooology to numopolkc our dudptinc. It is a means re an end, ™t
an cod in itself. Those of us who know the (tecp-seaicid satis&ctioo
which ficcnie* from the sCKlenacm of the chioaolog; of a site know also
sonKtliine of the danger inuninen; in thai satisfaction. What are we
really out for? I began this chapter by describing chronology as the hack-
bone Of orchflccilogyj and so ic i», but the backbone ts not the whole
skeleton, still lew is it the flesh and blood and spirit of our subject. Oui
objective is a reoonscnictioii of human cultuisl achievement in all those
aspccls of life which ate susceptible to material evidena, A chromlogist
is not an archaeologist. As an American wriict has put it, 'Chronology tS
admittedly on important factor in any archaeological research, and the
ouiicst and surest method of eatablisbhig it Is to be cornmendetL But
after a sequence of periods has been establiahtd, if then the culttue
ofthnseperiods is unknown, we may justifiably ask "so what?” ” Some¬
thing will be said about ‘so what?’ in Chapter XVlt. Mcrnwhile, we
imiat proceed to consider how in practice oux cultiual maieiial may ben
be ctjuaiod with OUT litnc-scale both in the digging and in our records
of it.
’ W, W. Tnjlflr, ‘A Study of Ardweotefy’. Arnttkait Aiahniptifep**^ 1
(lAdiana Univ., 1948)1 p< da.
39
IV
Stratigraphy
I N the lost chapter nfereocc was made lo a tnunbcr of ways, sumo of
them very familiar ones, whereby an appnmimnoa to absolute
chronolap ™MmedjiKS bccimaedEromiimlDaimciited archaeo¬
logical oialeriil. Such absolute chnitHilogy is esscndal alike to llie appre-
darioD of the vai^-ing jempo of humaa achirt-cmeni, and, above aU,
TO the establishment of the cultural lotcrrektinnships which help to
mrioiuiliie human ‘iwogress’. But in much of mir archaeology fticd
nme-pomts arc intcimiiieat ami chancy. More imialJy, the mehaHJlogist
must be cooicnt ro establish the ntsitixv xquact of his evitkoce, to
Mure that, hown-cr iU-focosed, his pcKpective is esscotially oonect.
Therein lies his primary dtiiyi to secure bejtwd doubt the orderly
succession of the vestigts with which he deals, cveu though, in any riven
P e of reseaich, he may be ooinpelled to Ickt finer adjustment and
ictcrprmdon to his succewon. To come at once to tJic core of the
nuttcTi his flrsl cask as an exoivitDT h with stmii/katim*
In picmresqiit fashion, Mrs. Jacquetto Hawkes Im had aomethinE
to «y tff tic Law of Straiukation. It is as simple, she remarks, as gravity.
^ falling do™ st^, and «indeed rather liJte that. There foUuws a iinJe
lantas^T which I absinict iroin her lasr,
If [she soy,] of one apple rdUnson the head of Sir I«oc Newton
a hfiivenlT ordimd had lei ramble a rain of fruit, one of the greatS^
^ “nd ti<«i buried. Aiy ono eaamining
Lie* r**!™ after^ds in a properly saentiiu; spirit, cfcarfi^ the applu
layer ■’y would be able lo deduce certain facts. He would be able to
^vc that the man was ri^re before the apples. Furaicnuore. that the
Hath found immediately over and nnind Sir Isaac fell
Im^ 0^ than the snuU iwanhy russets that lay above iliem. If on too
*““1 obMrver, even If he enme froJn MjuS
rr t!** “"*• “S« aw not enough, (Mrs

Hawke* BddiJ the nb*^nter would want on abwlutc date, and thS* wh«c
^ csBunlnaTiw of his clothe*, the toxigsfcirted
toes ’Jf^d*** ^7*^**/“** cut of hi* linen, the long square-
toe* pointing » fMlnraJy up to the sky, would date the man
40
STRATIGRAPHY
tixtith ceoniry. Here wmild be a due tai the aê ihe apples and ilw
snDw.i
In more plOSAjc seose^ this letm scniiification ]ms aineady appared

here and there in the previous chapter* and will icmain a r«xum-nr
theme. It wb* loug understood by geologists before it beanie the enm-
monpbee of archaeology. The proiagouist in the geological use of the
J term vm William (^Strata’) Smith who^ in i8i6i began the tsiuc of a
' work entitled "Strata Idcntihed by Organized Fossils', and thereby
opened up new geological profnndities md sticccssions. Wiih so long
and familmr a history bcliJad it, it i$ utmecessary to enter bene intn any
deuilcd expoiuiian of the stratigrapMcai prindplc^^ but, for complctc-
nesA's sake, I will biieSy reoiU the natuic of its application to aichaeo-
ûgy-
In thisj enniext^ its pruidples arc—in theory^ at any rate—simple -
enoiagh. Yhe human uceupatioti of a site normally t#ults m ibc accumu-
ladon of material of nne kind or another on and about the area occupied.
Objects arc lost or discarded aud become embedded in the eanh. Floors
arc TCDcwcd and old ones buried. Buildings Ciuniblc and new ones are
built upon the mins. A Good may destroy a building or a kswti and dc-
puMit a layer of allirvium upon its debris j and latcr^ when the Good has
subsided^ The levelled site may be leocxupicd-I^Sorncriine!, the process is
in the reverse direction: etidenocs of occupation jnay he removMij as
in the deepening of an iidsurraced Street by fmlhc, or the digid^g of a
pit for the disposal of rubbish ot fOr burin!. At Sabratha in Tripolitama,
a Roman temple was found in r948 to have been rebuilt ai a /ctot level
than its prcdEccasnr on the same sJtCi thtou^ an intcntioniil lowering of
the whole at the lime of rebuilding. In one way or anodicr^ the
surface of an apdent town of viGflgc i$ oonsiamJy altering in res|5on5e to
human eftorf or pcglcct; and it is by initfprcring rightly these ci-’idcnces
of alteration That wc may hope to reOoiBtrucT something of the vJdssi*
tudes of The rite and its oecupatiLS.
As it happcnsi the Gret recorded imiancc, so ftr as 1 am aware^ of the
observ'apnn flfferchaoalogjcal struLiticatinn reiired not to itn occupation-
f site huT TO a bural-moundyAi the lime when the episode ocCurted, the
%tjrd in the sense in which I hav^c just used it, was not yci
in cireularion, but the clcmcius of the matier woe preoodously appfc-
L dated by that ronarhablc DLan, Thojiiaê1lcr?onj whose majot tide to
f " - ■
■ A Land CLotidcnip tî), p. a6.
4i
n
STfSATlGRAPHY
fame I suppose,, that he was third Pttâident oF the UnJied States Mid [
on author of the Dcclaradoii of America ri Independence. I’or iis he laay
rank in the lesser but nor ncgligibk toIc of ihc first scientific digger. HiSf
sociological interests wicrc wide; in pameuJar^ as governor of VirgMaj
he conctmed himnclf with the problems of the Red Miin aiid the NegrOj
and In the courifc of hL>i studio he became interested inddentally in cer^
tain 'barrows nf which many are to be found ah over in this Country'
(i.e. Vxt]idma). One of rhem^ situated "on the low grounds of the Rivanna,
about twu miles above Its principal forkin and opposite to some bUlSj on
which bad been an Indian lo^Ti^ he opened in order to satisfy bknictf
of the ODneoneiis cf npintems and uaditlons relating to these mounds.
His report is a testimony to his careJbl Observation:
Appeanmw cerminly iadJcaic that ii [The mound] has derived both
Origin lind growth from ihc accustommy cuUcction of boncs^ and ihe
depontion of ihcm together; that the first collection had been deposited
on the cofliErtou surfnee of the canh, .1 stones put over itj wid then
a covering uf earth, that the stoind hod been laid on this^ had coverKl more
or less of it in proportioD to the number of hones, and w-as then also
«?vcred with carthj and » on.
These facts caused him in rejEct two notions: firsts that the mound
cuvened only bones of ihusc slain in battle (not a single weapon-wound
was found); md secondly 'that it the COmiiioii sepulchre of a. town,
ID which the bodies were placed upright, and toucbiag eadi other'. He
iduttthed that a few stones (omid in the mmind were ‘brought fniiii a
cliff a quarter of a mile off, and Hum the river, one'dgbth of a mile off'.
He abo nworded that laTaots wen: huried there, since a rib of an mfiint,
part of the iaw of a child wjjich had nut cut Its teeth {the right halTof the
under jaw^, dee., were discovered.'
All that, be it noted, was in 1784. and I have tlicre&ie thought it
worth while to relate the episode at wme length. Let us consider the
contents of Jeffenmn's dear and concise report. He describes the situa¬
tion of tlic mound in relarioQ to natural features and evidences of human
occupaffoD. He delects oomponents of gcDlogica] inieresl iu its marcrials
and traces their soureu. He indicata the stradgtaphical stages in the
commiction of the mound. He icconi} cerrain significunt fatiucs of the
skeletal retnaiia. And he relalcs hia evidence obfectivdy to current
theories. No mean achievement for a busy statesman to 1784!
' T. Ni-roeo >'»>*..uh (*th ed., i60().pp. 143-7” awwint here dc-
£|VE from A. F. Cihimbcrlaiii in T’ftr vfht (ff707), 490 f.
STRATiGEArHIf
^ ^ Q^' scientific skill fell upan LnTtrtlLc ]

soijl For a cenruiy after JeffecsoGj mass-eatcavaticiii remained the itJe
of the doy, In Chaprer II the successive Jayccs in the mU^ in other word:!
stnidflcatioa, vitrc compared to the successive pages of a book. The
analogy is essentially a mic one, and indudes ibc coroHaiy: a prime
eoi3ditioo of InteHipbility i$ that the layers, like the pages, shall be
brought to our cy« reasonably inraa and in proper sequence. W counsel
moss-excavation has not always been devoid of aU mcrif. In a nidimen-
tary stage of research, it may hdp to poini the tiây and stimulate ad¬
vance along It. We may be grateful to Schliemaon for plunging his spade
into Troy* Tlryns, and Myceaac in the sevaitiK of the ceniury, be-|
cause he show^ us what a splendid book had in fact been buried ihcre;
but he lore it to pieces in snatclunglt ftoui the «nh, and ir trmk us up¬
wards of tbree-qtiaiters of a century to stick it more nr leas together
again and to read it aright* with tlie hdp of cribs from other placca. On i
smoUer pbuacr I icnreinber my gratitude in 1544 to ccfraJii ardent but
higldy unskilled French undquories at Fondicheny in Muth India for
scraping up a beterugcnoiis mass Of iDatCfial from an andent siie be¬
cause* although they' knew it not, their spoliarion showed that the site
cemtained imported (and dated) Arretioe ponciy which subsequently, by!
more orthodox methods, enabled uS to dnennine the first aiehaeologicai
datum-line in pre-medievaJ south India, But the accidental benefits con-^
ferred on us by out Fondidieriy fiiencfe, or even by SihlieiiMiin hkiiselF^
are no longer a \iilid excuse for archaeolDgiGail illiieracy. Today, the
digger must leam lu read his sectiensj or he ihould be constmined from
In praeuocjê idcutificaClOO and corr^daiinn of the strata or layers )

which Tcprescnt the successive phases in ihc ardiaeologicaJ ’history" of a
site is one of the piindpal tasks of the aavator and will occupy the
maior portion of his time. So imporranr is this task that^ flt the risk of
w'earying the reader cither with the ncecssi'Siely unfamlhar or, more
likely, with the excessively fiutiiliarj it muld be wrong 10 lorebear from
CertiaJn lathcr arid techulcfll dfrails ^ The task is One which im'olvs
clear and logical thinking remfotced by dperience and infinite paoenec.
Normally, the strata are ddfercntiaied by ■I’ariaiiioiis in colour or marcrifll
Or cootciir.'^ Not infteq neatly, bowTver, these variarions, parricularly
under the bleaching infinences of an Afticau or Asian nm* present
diilicalty c^^cn to the experienced eye: m much so that mom than one
archaeologist who ought to know berrer has -denied the presence of
43
STRATIGRAPHY
Stratigraphy (in the Western sense of the term) on Eastern sites. ‘There
no clear stratification’, writes a well-known American archaeologist
with reference to a Palestiman site, ‘during a good part of the period
covered because there was no complete destruction and rebuilding at
any one time.” This, of course, is nonsense; by ‘stratification’ the I
writer quoted means merely ‘continuous building-levels’, oblivious of ^
the no less important layers which on any site may be expected to sup¬
plement and interrelate phases of actual construction. The fact is that
the observer had simply failed to observe. There are in practice various
wa^ and means of dealing with the reluctant, sun-baked section of the
Orieiit, or indeed with many sections in the West. Damping, and careful
scraping with a knife or turf-cutter, will often provide the remedy by
brin^g out the more subtle variations of colour or material. Observa-
tipn in different lights at different times of the day may help. In a Hiffi-
cult and important section, observation may be continued over a period
of days before certainty is reached. And finaUy an attempt must be
made to ‘rrad’ the section—to discriminate, without prejudice, between
the more significant and the less significant differentiations of strata: for
example, between a mere ‘tip-line’ in a continuous accumulation on the
one hand and a substantive and emphatic occupation-level on the other.
It IS not enough to identify layers, although that is, of course, the
cssennal first step; it is the task of the archaeologist to interpret them, to
understand the sentence as well as to transliterate it.
On this all-important question of interpretation something more must
be said, always with the proviso that the written word is no sore of sub-
sotute for field-experience. The most that the professor can do is to
offer such tots or warnings as shall create in the mind of the student a
healthy wariness and a proper regard for trifles. Let us take once more as
our meme the first and universal question, of which something was said
m the preceding chapter: What is the time-value of archaeological
strara? How long did it take, say, 4 feet of stratified deposit to accumu¬
late. A very searching and important question, weU worthy of the most
^ful consideration: if we could always answer it, half our battle would
be won.
Reference has already been made to a famous lead which geology has
given us by the recognition of the varved clays of Sweden (and else¬
where) as the annual deposits of the retreating ice, and as the time-
* 9 * TeJl en-Nashehy txcavated under the direction of the late
William Frederic Bade (Am. Sch. of Or. Res., 1947), i. 10.
44
STRATIGRAPHY
table, therefore, of a related human phase. But it is rarely, in all con¬

science, that geology deals with us so straight-forwardly; whilst man¬
made strata are capable of every sort of perversity. Some of them, indeed,
are of no chronological import whatsoever, and, as instructive disturbers
of all faith, these nonentities shall be considered first.
It has been my practice, from time to time, to persuade my students
towards the end of the day’s work to cut a section through the dump
which is the outcome of their digging. They normally discover, as is to
be expeaed, that the section thus cut is replete with stratification—tip¬
lines, streaks of variant soil, a miscellany of the materials through which
they have been working in the course of the day. Nothing is more cal¬
culated to disturb their faith in the time-significance of stratification.
Here, in front of them, is the variegated accumulation of a few hours;
how is that to be reconciled with the interpretation of the adjacent strata
below ground in terms of cennuies?
In theory the answer is difficult, in practice it is usually easy. It wUl
often be found that certain of the strata cease towards one end or both, in
such a fashion that the underlying and overlying layers unite to clasp
them and hold them within a uniform mass, as it were in suspension a
sufficient proof of contemporaneity. A succession of layers may join up
laterally, like fingers extended from the palm of the hand, essentially
integral therefore with one another in substance and date. Very rarely do
deposits of this sort consistently ape a prolonged and systematic accumu¬
lation. On the other hand, they present a warning: a seleaed portion of
them may be found to simulate a consistent and logical sequence, and so
serve to point the danger of argument from small sections. Fig. 8 shows
a problem of this kind, solved on the right by an extension of the section.
It should in any event be axiomatic that no chronological sequence can
be regarded as established securely on the basis of a single section.
A good archaeological instance of the insigmficance of stratification is
provided by many of the storage-pits cut by Early Iron Age farmers
round about their dwellings, pamcularly on the chalk-lands. After being
used for a time for the storage of farm-produce, these pits were liable to
‘turn sour* and were summarily filled up with any material which hap¬
pened to be nearby. The result is often a strikmgly variegated filling, of
which the stratification bears no relation to time-sequence. In other
words, the first rule about stratification is that there is no invariable
rule.
The opposite danger to the over-emphasis of what may be called
45
STRATIGRAPHY
‘acodental’ and contemporary features is the under-emphasis of time-
'ûes; the assumption that a succession of strata have accumulated con-
tmuously whereas in fact they represent an intermittent growth, with
one or more time-gaps. Thus the original excavator of Gezer in Pales-
tme m 1902-9 hopelessly telescoped the Iron Age chronology of Pales-
tme through a failure to recognize a gap of some five centuries in the
wcupanon of his jite. The mistake was gradually rectified, years later,
by the evidence of other excavations, but for a time it completely upset
IIITllll||||llllllllllllll|IIIIIIIIMIIITlllll0fTITTT^ THiiiriiiiiiiiiKniiirTmrTTTitm
<=» //A. ^ »//..
.vv§L. A- , h. . . .vvV
IhnJ.' Secti°" illus^ting (A-B) an apparent succession of strata (3-7)

shown by an extension of the cutting (B-C) to have been depoited
simultaneously
centuries in Palestinian archaeology. How is a mis-

(udgement of this sort to be prevented? This question is manifesdy im¬
portant and deserves careful consideration. ^
îble, Md either or both must be considered in the interpretation of a

secnon lUustratmg more than one cultural phase.
The first criterion is the nature of the soU at the point of junction be¬
tween two c^tu^. An mterruption in the occupation may here be
of wind-blown or water-borne sand, or by the de-
Sie I ve^tadon. On a Dorset site, a dark layer separated
*e ^ly Bronze Age from the Early Iron Age, and this dark stratum
was d*aôscd on analysis as ‘brown-earth soil developed on a subsoU
"f u “““ a gap in the occupation and a covering
of the hill with woods'-in fact, a gap of a thousL years or moH
46
STSATIGRAPHV
nicrf 4 or 5 inches af blown soil, compressed far more thnii twenty ctn-
ttiries beneath a lisiag canopy of trampled earth tel] us all lhatj and hint
at changes of dicuiie and water-table which for athouiMulyearsdivencd
funily-life to more homely lowlands. A subtler MOinple comes fima
Hadrian’s Wall, from die filUog found by Mr. Gerald Simpson and pro-
r»»r Ian Richmond in the ditch of The Vallum where it w» lotally
superseded by the Ronuui fort at Birdoswald, in Cumberland. 'J’hc prob¬
lem was to asccnain how long the Valium ditch was open ar ihi* point
before the fon was built over it> and investipation was thetefnre nude
on a seiticnJ secrion of the material now filling the ditch. l*he result, as
reported by Dr. KaiWcen Bbchbum, of the Departmem of Botany in
King’s College, Mewastlc upon Tyne, settled the matter and is worth
quoung. It is as follows:
Above die yellow silt of which tlw botrotn (of the fiUiagl was cnniposcd
were dcpoiits of pent, which varied in ihictaies* in diflcieiit sccricina md
Were obviously composed of lumps of peal put in by hand. ,,, .An in¬
vestigation of the motertol fiom the floor of the ditch showed little ofginue
maner, and no ttaces Of the dork csjlounition which » foe uswal (ûct
of organic decay. Whai little icmaini were present consuted very largely
of weed seeds; foe majotiiy of tliese were of foe Knm-gras*
aptVM^crir), but examples were also found of Chicfcwttd (Sidta rnttaa)
and of a butictoip. which was probably Rumiiiciititi aerif. These mc^
were in perfect pieservotion, pichably due to foe prescrvinit actkiiJ of the
water draining from foe peat immediately above. Such a lloia m UKM
SOoJs SMgg^l is one which would only be fmmd on newly disturbed groim ,
From thus and from ilie absence of orffinic remains, I think we may
assume ihai the dilcb could only have been open il yrar nr iwo bdwt the
le-HUing with The peaL>
The ditch of the Vahum as ihmfurc deliberately bloncd out of

tence here (for foe accommodatioa or tattensim of the fort) almost im-
mcdiately after its creation—fl condusion of rcv'calbg luofon' and of »
notable Unponance in foe hisiorinil diagnosis of the fropti«Hsystcm.
Similar evidence was recovered nearby from foe ditch of the Turf ^ all
where it underlay foe fori, showmg that foe fort was built looss it with
no appreciable inicrvcnmg period—again a 'historical iiifcrenee of the
highest value. . . .,
Even without analysis, careful observation under favouui e c
I Trow, Ciantttlmd aini tTiiawrinnJ Sor. C«4,), MB* {KetwlMl, Ipay),

3o3.
47
STRATIGRAPHY
lions may reduce stratification to terms of a precise chronology, occa¬

sionally comparable with that of the famous varved cla}rs. Two examples,
one from England and one from Iraq, have been cited above (p. 30);
Md time-gaps and evei^ time-spans may thus on occasion be identified
m a number of ways, provided by the close observation of weather¬
worn or overgrown sirfaces and, above all, by soU-analysis. But suitable
phenomena or facilities are not always present—though the latter at
least should be available to a modem excavation. There is, in any event,
mother approach to the problem, and the best way to describe it is by
illustration.
In 1947 I was digging a town-site on the Mysore plateau in south
Inâ, m the vicinity of an undated Iron Age megalithic cemetery, with
a view to the correlation of the culture of the cemetery with one or other
of the phases of the town. In fan, the sections through the town-site re¬
vealed three successive and quite different cultures, of which the middle
one was that of the cemetery. For the understanding of the sequel I must
burden you with names. The topmost culture was known as ‘Andhra’,
from the name of a local kingdom, the middle culture as ‘Megalithic’,
since It was also that of the local megalithic tombs, and the lowest as
btonc _ê, from its characteristic product. Of the three, the only
^table culture was the Andhra, which could be ascribed mainly to the
first cen^ a.d. The problem was to relate the other two cultures to this
and so, for the first time, to obtain some sort of chronology for them
tor this pu^se it was obviously of cardinal importance to ascertain
w ether the three cultures formed a continuous succession or whether
they were iwded from one another by time-gaps. Examination of the
semons failed to reveal any intervening weathered or exposed surfaces;
and mdeed, had there been such surfaces, laboratory facilities for their
^ysis were not available in India. Accordingly, recourse was had to
other types of data. In one instance, chance played happUy into our
hM^: a pot-bunal of the distinctive Stone Axe type was found in¬
sert^ into the lowest overlying stratum of the megalithic culture,
implying a co-occupation by the two cultures at this point. But even
wthout so mcontrovertible a demonstration of continuity, the matter was
set at r«t by what I may call the actuarial analysis of our sections. Such
a^nal an^y^ should, in my view, be made more frequently than
^ are, and the Mysore example may therefore be described in some
detail. Bncfly, the method was this.
In an extensive cutting, chosen as free fr^m complication in the form
48
PLATE III
StTHtilitiitjQn m Mywwc StaUfi Ip^-

iSap.49-^
STKATIGRAPHV
of ininjsive piiB or smurtunaj, a dTicful fic^sfor was fecpti layer by J^cr,
of every por^hord founds and tbe rcmJfa were tabulated (Fipi. 9-10 and
PJ, 111). FcuTurviTtty the fabric and technique used by the three cultui^
were w diranaive fmni one auoihcr that rhdr clasiiificsdou was bc) oud
doubt; the ponery of rhe Stnne Aite pcopk being marsc, band-made
stuff, that of ibc Megatirhic folk being a polished and bcaudful black and
bio'^ and turned ôwly if ar all, ihai of the Anilhia at the top being
lurued more mcchanicalty on a fast wheel and otherwise dabonited wi±
ghuco and patterning. Naw ihe table shows a substantial oveiLipj run¬
ning through three successive layers, betweeo the Stone Axe and tltc
MegalithiC setits, followed by a similar overlup bcmcen the AlcgdithiC
and the Andhra merles. In esiimaring the sigmEonec of these very sub¬
stantial ovej:kp% due altowêe musr be made for the fact liML, from a
varied of causes^ the subsoil is aiwuya in a stale of less or greater move¬
ment. Animal and vegetable life, and climate, are eutîantly at work in
it. Relics of one stratum are always liable to End ihcit way sporadically
mlo anoLher and to confuse our omce cninda. Em the ihrusting'Upw'ards
cespeaiveLy of 235 and 219 sherds from one euliurc into the next above
it in this single SOCdon 1b nor to be thus lighUy dismissed- The only
acceptable ejcplanauou Is that on neither ciccasioii did the attitTil of the
newer culture on the site involv'e the iininE:dkte extimctioo of the older;
io oiher words, that the cultural sequmoe was cmlmueut and* in our
esdniatc of its chronology in teJarion to the fised point at the top* we can
ignore the possibility of iioEoiowti factors arisiug frofu inifemipiians ui
the occupation^ Wc are here in no danger of tepeadn|; the disastious
oversight of the esavaiOT of Gezet Cp* 4*5)- Detail of this spedhe
Mj'sotc example do ttoi ooncEm tii> it wiO sufhee to observe that the
overlap dius deoionsmTcd between the megalirhic culture and the
overlying first century culrujc proved that the megalithicr tombs were
Ibcte in use until the firni centuryj and SO prov-ided the firat llnaa date for
this very abundant caLicgoty of soudi Iiidian smteturot.
WdJ, iheie are examples of various tend* of strarigraphical evidence;
ofkyets tJiai are conicmporary with ooe atKithcrr layers that are repJir-
oted by greater or Iciser dmE-iniervaJSt layer? that have accumulaied in
unbeoken succession. The reading of A section is the reading of a langu¬
age that can only be learned by defbonsrration and experience ■ A word
of advice to the student. Howev er pcaaiwd. do noi tead tew hastily. Be
your own devil's adt-ocate before passing judgcmetit. Atid, whetev'er
possible, discuss your diagnosis W'iih othcraf—with colleagues, with
mu A9 ^
STRATIGRAPHY
pupils, with your foreman. (‘The testimony of one person is no testi¬

mony*, declares Hywel Dda, the wise Welsh law-giver.) Be humble. Do
not ignore the opinion of the uninstruaed. ‘Everyone knows as much as
sov7v-m£jr
a. ANDHRA
CULTURE
MEGALITHIC
CULTURE '
ID. STONL-AXt
MlXtO CHUNAM
^zburial ^
'/a UW
IA STONE-AXE
CULTURE
aAfiU£/0
NATUI2AL SOIL
, - Of Fut 3cal£ Of /HsTnes
Fig. 9* Section from Brahmagin, My^re State, India, showing three

cultural phases with overlaps
the savant. The walls of rude minds are scrawled all over with facts, with
thoughts.’ Emerson said so, and he was right. Even if you do not accept
the views of those you question, the mere act of questioning is at the
same time a restraint and a stimulus.
50
STRATIGRAPHY
We nim now from interpretation to record. But first reference must
be made to a method of recording that not long ago was widespread in
the East and may in fact still survive there. If so, it is the survival of a
LAYER I. STONE AXE II. MEGAUTHIC 111. ANDHRA

X .. .. 52, including i yellow-painted sherd
2 .. .. 384, including lo yellow-painted sherds
3 •• •• 480, including 68 yellow-painted and i
rouletted sherd
3« .. .. 67
4 .. 36 269, including 51 yellow-painted sherds
5 .. 68 219, including 10 yellow-painted sherds*
6 26 115 405, including 7 yellow-painted sherds
7 63 407 ..
8 150 199* ..
8a 36 ., ,,
8b 89 ..
9 76 ..
9* 196 .. ..
10 46 ..
II 33 ..
12 23 ..
13 26 ..
14 48 ..
14a IS .,
15 198 ,.
16 7 ..
17 45 ..
18 25 ..
*9 321* •• ••
‘ In adjacent cuttings, layers equating with 5 and 6 of Br. 21, Le, the lowest Andhra levels,
produced 7 sherds of rouletted ware. ,
' In an adjacent cutting, the layer equating with this contained an um-burial of the Stone Aze
wuiiure.
* Including *Enrly Puintfd* pnd ^ nf thg lA culture^ which this lnycr rcprcscnn^
Fig. 10. Tabulation of sherds representing the three cultures at

Brahmagiri (see Fig. 9)
fantastic and monstrous device evolved in the alluvial plains of the great
river valleys of Egypt and Mesopotamia as a substitute for exaa ob¬
servation in ill-controlled ‘mass excavations*. Its origin is probably to be
found in Petrie’s belief that on an Egyptian town-site it was possible to
equate the accumulation of material with a specific time-scale (p. 29).
The validity of this ‘principle* was doubtful and dangerous enough in its
STItATIGItA?H¥
spcdMiif^d cunt^t; It Ims my pLicc whAtMK:v[:r in the ^ncral
technique of modem hcId-sitdiiculDg)'. Yet in IdJuIt fur cxajnple^ ^ re-^
cendy os 1944^ it was stiU Lhe unly iiicihDd knQ-nTi.
Brieily;^ it consisted of the meehsuiiGil recording of every ohiecr and
structuie Lii reladon to i Bxed bcDcli-Icvd, Thna in the e^tovadons at
the great prehistoric city of Mohenjc-dmi in the Indiu valleji in 1^27-
3 If the records were prcpaiDd hum bcneh-lcveU, in one area *178’7 ft.
above mean scu-leii'cl" anud in another îitQ'9 ft. abo^'e sea-level^ the ^sr*
sumption being tliui all objects andstnicrures at the same level below (or
above] datom line weir in the same ^srratuin'i iê. conremportiiy with
one anotbicrl I have desedbed this system as ‘incredible* and 1 repeat the
■descnption. So incredible is it, and yet sn widespread], tlmt the e^cava-
xof^ own proud accuiint of it may be repeated. Ik says:
In order diet onr deep digging mighr be !i3rC$factdnly carried out, an
extouive system of levelling was necessary. Tlie levels of every building
and of every w^ll were Therefore tHkenj. especiaJ [mention being pitid to
door-siJls and pavemenis as being for purposes of ntrditlhcation the most
unpomnc parts of a building. In additic^n^ both the locuj and level of
every objcei found, whether it was icpudcd st the time as ipnpnfTinJfit or
not. were noted in order not only to correlaie each object witli the build¬
ing in which it wai found, but aJ^ m facilitate the study of the develop¬
ment of an and tedmique. As some Lhousands of objeos were lirLcirrbed
in the sections that wc excavated, it may be thought that this pmcedurc
Wiis unnccessonly tabarious. ThJs» however, w-as not the case. ^Flit level-
ling insmimenu were set up early in the momfng^ and remained in
pofddon all doyi and h wyi quite ■ limple matter to take the level of each
object directly it appeared.
It was, however, admitted that this method was not wholly ftcc ftom
complexity^ that there wm
limimnnns 10 rhe dcductiuns to be draivn from the kvchi at which ob-
iecfs are found. For intumcc, if a ].ir or a seal lies dther below nr ai some
disioocc above a pavement Or door-siJJ, it lb dilfioilr ro decide to what
period it bekmirSi Wc. thereforCj^ adopced the rule that sdl objects found
in or near the foundations of a buildiri^g be assigned to the period of fhar
builduig rather thxn id the previous phaae^ unlcM they acruahy rested on
the rcmoliiB of a pavcmeiic of eorhex datei for ir Is more than probable that
they were dropped or left behind when the foundaôni were being made^
Xhe chapten on the pottery and other finds in the cxcav'utotâ subse-
qucnc report mclude page after page of daboTTite but insignillcaiit tables
based on thix procedurCi
STRATIGRAPHY
III Other be it repeated, the S0H;:aUeil "zitiuiiiiQitiDii^ of the

Indus valley dviluatlcn, one of the major dvilixaduos uf the andrnt
world, was dominated^ not by local cbseii'adon, but by the level of the
sea nearly 500 iii4le$ awayl This meclumlcal dassificaiiou cun only be
categorized as the veiy parody of scientific method. I t bears little more
Telauonsliip to scientific aicbaeology than astrology bears to astronomy.
To appreciate its utter absurdityi we need only recoil tliat, except
perhaps at the earliest level of a site (hardly ever adequately explored^
un ancient di^ in the East is never levd. Very rarely is a city completely
destroyed and completely rebuilt at One moment and at one horkon^
Normally, a house is reconstructed or replaced as it decaiySj or at the
whim of its owner. The town os a whole is constantly in a state of differ-
eniial destruction and construction. Individuai bn ilding-sitcs rise abo^’e
dieir neighbours^ the town-site itself rises and asstmies rhe contour of
a hhl; buildings on its slopes are coiitcmpofary with buildings on its
summit. A doorway or a potsherd may be found at one spor to feci be¬
low a doorway or a potsherd of predsely the Mine date at another spot.
Such diflercnces, of vital importamco to the sdenitfic inicrpretaritin of
tile site, arc ironed out and oblitemted by the bendi-lcvd. If it be
aecessiory to illustrate further the previous fallacy nf xhhi method, two
diagrams (Fig. it) may serve. They arc Ecif-acpl&iuiory.
Yet, for all the obvious absurdity of the daruin-linc system just di^-
cribed, the substitution of so-called 'levels*—wherher abstract building-
levels or purely arbitrary depth^-lints—for facnial srradfiadon dia
hard. It rccuis, for example. In a revised edinon {^95°) Aianuai
Arcfiaioliîail Fisid Matfiwis pieparedby a leading AnKritan uaiveisiLy,
Tlicrc, as sturdily us ever, thtivos the old ourwora system, with its
mechanical ûnit-levels'* governed not by changa of soil but by *thc
length of the shovd-blode to 12 inchcs)\* True the word 'stratifica¬
tion' is not unknown to the 0nthoTS, Ir rqiracnts a phoaomenoni they
admit, which "miiy be vi&ible in the walk of the excat^don\ but, we WC
assured, 'any stradgiaphy of anifaa types and animal bones will appear
after a study has been^madc and nci^ not bother the excavator in the
Geld^ [jie). The noiioa of peeling off the successive strata in conformity
* It may be noted m ppasing that mrdianlciJ *f«Pt-lcvrh" wc« ti«d a% kmg

Afifi OA iftfij by W, renffdly, F,ltS,i in the cKcaranan of Krax"i Cavern at
Torquay. See hti intemiiug MS. Journal, in die pcwieukiti df tbe Sodety.
But it i» Bid lo find the ujne eHUWoni method aJiiJcateil hy ji dtHinatiinhcd
iiniverîry In 1950.
5^
StSAtlGKArilY
with ibcir proper bed-linetj. aud thus ensuing tic accuraw isolation
of stiuctuiai phases and relevaai Brti&as, is not even consideted.
Enough DferitidsTn; ter us turn toa more posJdvc aspect of tlje matter.
The preparatinii for the Kcord of the seoloa bcgiiu with the first spade-
fid dug. Frnni thcouisci, the Sttum are caiefiilly observed, djsdngtiislied.
FtC, it. Diagiams lllusinitutg the Jitfaiificfltioin of adty-mcnttid (below)

and the fiUlacy of recording by mKhanica] Icvcb (above)
and labcHtd as the work proceeds. U is, of course, as ihe vnjrk proceeds
that ‘finds’ are isolated and recorded, and their rooird Is necessariJy
integrtl isiih that of the strata from which they are derived. The supow
visor must iheiefbcc make up his mind dearly froci moment to moment
as to the UmiBand nomenclature of his straw; and his decisions, whether
ulDDiawIy approved or modified, mun be susceptible to occuraic de¬
lineation, if only for the subsequent cofrcladoD of Jus ‘finds’. I n other
words, both he and the spectator or the future reader miisr Itnow exactly
whar iJiinltj ht h dojiig.
I have found in practice that there is only one foolproof method of
54
STRATIORAPHV
ensuring this. Tht successive layers inusT be defined and dearly labelled
as they coniE lo light. By labelling I mean the actual pinning &f a label
by a nail or peg Lnio the side Of the trench in (preferably un tlic top edge
of J each layw. The hibel beats the number of the layer within a cUcle (1
reserve endrded numbers for this purpose, tn avoid all lish of confusion
with numbeis having other osimewtionR), supplcmcnicd by a ujum : for
(Sample ‘lower brown’, 'red Clay', 'porridge*—it matters not what, so
long as a distinctive word or phrase is used to emphasize and cross¬
check the dificrcniiation, Nurobets may nn ocauJon be erroneously
duplicated, but the addition Of a name avoids risk of confusion, And
inddcntally the use of a nuine Wiids to give indiviiiuality to a Liyer and
helps the mind in a piaorUl rcconsmteriosi of the section.
Accordingly, I like to see my scctiens plastered from head to foot with
orderly arrays of labels {PI. Ill), which serve three iniUn puiposcs: they
dcniitnd clear and decisive thought on the pan of the superviaor who
invents them, they show on the ground and on the drawing precisely
what his small-find labds tucoD, and they make it possible fm the direc¬
tor or a stihsiituTûpervisor 10 undeTsiand at once the diagniHis up lo
date, Ind dentally, they enable th,it diagnosis to be checked—always
with the provisofriiat any material altcndon in it will probably mcoo
either a relabelling of the relevant ‘finds* or at least a record^ Of the
original as well as the coireaed diagrwsis. As o general guide to the
young, it is wi.scr to insist upon the uver-stradficaiioii ihitu the under-
stndficadon of a scctiun in the first instance: ii is osy subsequently to
group layers and thdr ooDteim but it will never be sale to subdivi e
them. - • u
Now a word a* to systems of numbering- IJiias or strata it is ob¬
viously necessary to number downwards from the top of a cuiting, so
that the numbers arc mostly in the revetse order of occumulatio^the
latest (ropmost) layer betng layer i. This somewhat fllopcal p
is Avoidable since it is nreessary to la>xr-miinlKrs re ^-finds
as they come to light, withoui vraitiog for the completion of the secuon.
The same disability docs not apply to odlWH, which emerge as recogniz¬
able entities at a later stage of the work and are not u-sed f™ J^b^g
purposes. Here the logical system can, and certainly should, be fo -
lowed: nameJv, to number the earliesi culture or phase os num^r T,
with II, III, *c,, in sequence above them. Unfominatcly, usage in this
matter is chaotic, A number of AiesopotamiaD and Iraniim stws (Arpa-
chiyah, Uruk, Uquaif, Oawm, Giyan) have been numbered from top to
5TRAT1GMAPHY
SO Uiat, for oLimple, Gfyun V ^nd Gttwra XX are airly, and
Gi>^n i Md GawPia I arc late. On the oihcr hand, Sialk VI is hit, and so
with Nineveh, Hissar, Susa (new classificotioE) and orher tttlomHy
classified sites* It should unqticsiionably be hud down as standanl diat
c^turcs or Ctiliuxa] phases are numbered (with Roman numerals) froni
the earliest to the latar (Fig. 9), The only cCrcuinstancc which xcuiy pre-
SCTt dilfiaiJty to this method is failure to reach the bottom of a site.
\tTicrc this hmppcnft, the earliest phase uacoverod may not in fact be the
^liat Oil the site, and therefore not truly its number I. ChanJiu^dam in
in an instanec of this kind, where the cxca™nr might justify the
top^nwnwanJs numeration, which he in fact adopted, on the pita that
the waief-tflble stopped him from teaebioj; and dtagnosui^ the lower
eve . Interim reports on uufmishcd excavaiions may provide another
excuse for this proaeduTC. But in general, the imwer to tlie diffitailty Is
tiair: no sue should be dilg unless it b seoioned and diagnosed m the
lowest level and a complete culturc-sequcno: recovered. Water should
not noisily be accepted as a bar; at Arikamedn in 1545 we had w dig
11 feet below scii-lcvet, and this depth we achieved, with some difficulty
It IS true, but withour elttborjte equipment, whilst ot MohcnjcHdaro ia
1950 we peaeirutcd with the help of piimpa to a deptli of in f«t bdow
tJit iMtcr-l^l under peculiarly difficult conditions, and with mote time
could ccnamly have dug lower still.
pc pud^may complain that two contrary systems Intvc here been
ncommcnd^ the numbering of layers &om top to bottom, and of c«l-
pli»m ftom fettom to top. In fact, tbem is no ttmilict whatsoever
niirnn^^ ^ lopcsl luid pracDcal for their several
1™“^ “Mnsisieuuj- is their only demerit, and it is of a
^ that only a could oup at. ‘Do I coniradicr my^lf?- cries
WjiJt Whii^n; \crr weU. then, 1 coiiindicr mj-Selfr Or rccaU tJic
wotds of a less bluatcnng fellow-countryman of Whitman's • ‘A foolish
SW M when .he magisterial
Ralph aJdn bmciMn op« his mnuth let no dog bark.
Our secuon, then, has been dug, diagnosed, and labelled. The task
rmat® ,0 uxonj jt accurately and expressively. The mechan-
ra w Sht d«
KE^’h! ^ “^“-^bber, a sale-nile. good thin strinE
IZ f rartowx’) or frinch^s, a m«sur?
Jng tape, dips for £jcmg the upc, a jToot rod, and a plumb-bob.
56
SXRATtGKAPIIY
At ti convenient pouu> gencraily at the top* the string is siretched

firmly dCtoss the scKlion utid levelled dihcr by means of a bubble-kvd
or by a surveyor** level hud on the tcimimil points. As usual, dcuilj are
impnrtant. String of the best quality should be used; inferior siring
breaks or, worse stiH, stretches and sags- In any ctisc, the atring should
be supported on carefuUy levelled pegs at hwizontal intervals of ten
feet. Moieovec, to avoid mors ftoin strciehing or other causes, the let'd
of ihc string should be cheeked once or twite during the day-
This string should be marked off in feet or metres w, îiM still, a
measuring-tape should he stretched (nor nw tightly) alongside it. Thetij
otevciy foot or m otiieT sdcutcd poinrs, verriod aicasuremciits arc taken
with a weighted measuring-iapc above or below the daium and are
uanEfcTTHl by the draftsman, on the spot, to suitably squared paper.
As to icah. the smallest scale at w hich the detaO (if an average section
be adequately reoaidcd is | inch — i fuut (altemariscly, s cm.
I jn.)j and this may bn regarded ns standard for large secriona. Wheneter
possible, however, the larger scale of i inch = i foot (alternatiwly,
10 cm- = I m.) is prcfetahle; it is more accurate, and better lends iiwlf
to onnoiadon- -
In the process of drawing the section, there is a cotmnun tMdenej- on
tlic port of the mciperienced draftsman m eaaggeraie mequdit^ m the
surface of a stratum, so as to absolIC its general contour and charactn.
He should be reminded that, on the normal sailt uf i ^ ^ ™
pktuie will be onc-twenty-fourth of the shse of urip - m
equality, therefore, rising actually a inches above the average l^d of
a scrotum will, in a normal sodc-drawing, vary by o y °
inch above the level, and so will fotm a nearly hcbJ^Wc bjrak m Je
surface- Almost invariably the draftsman, seeing (he ^ rather tkia die
wood, will over-emphasira the obstacle, albeit that the feamn: is excep¬
tional and misrapiesenw ihc geoetal mraning of the layer,
Furthermora. there Is the question, nm merely of ihem ct-cmpliîsof
accidental frames, but of the over-emphasri
whole layers; ttauliing, in fact, in a permawnt z^mterprmiwn of tfe
aecUon. I can best dcmomtiate this by three dagratm CP«-
diagram A, the draAsman is assumed to have deUnrated the
sttaui corteetly and fairly, with ttfera.^
But this delineation, although mdtatmg the pres*"* r.,^ liis
little or nothing to indirale their vaiying character and
an flhDOSt OK^cgkSS cnUcction of lincs^ ptocesswu of Icttera nut
57
S8
STRATIGRAPHY
yet divided into words. A more ambidous draftsman may attempt to

indicate something of the individuality and diversity of the strata, and
diagram B illustrates such an attempt. This fails in two main respects.
First, the general evenness of tone throughout the section produces the
unmeaning monotony of a sentence spoken without inflection, and so fails
largely to convey the intended impression. The draftsman has not realized
the varying significance of the facts which he is recording; he has again
failed to see the wood for the trees. Secondly, he has not realized that his
rendering is not, or should not be, merely a transcription of accurately
measured or even accurately emphasized lines: it is, or should be, also
an accurate picture of what he sees. Not only should lines of demarcation
be transcribed from measurement, but also the size, shape, and position
of brick-bats, bones, sherds or other materials which, by their character
and quantity and by their ‘angle of rest’ in the soil, combine to indicate
the nature of a stratum and the method of its accumulation. An intelli¬
gently drawn section is far more than a diagram; it is, as I say, a picture,
representing not merely the skeleton but also something of the vital
flesh and blood of its subject. Diagram C is diagram B corrected in this
sense.
It must be confessed that a well-drawn, i.e. intelligently recorded,
section is relatively a rarity. But it is nevertheless a basic necessity of
modem field-work. The published sections are the readiest index of the
value of an excavation-report.
In these diagrams, and elsewhere, I have used certain symbols for the
easy conventional representation of different types of soil or deposit.
They have no special merit but are reasonably expressive. Some measure
of standardization in the choice of symbols would be u^ful but has not
yet been attempted in this country. (Sec, however. Fig. 13-)
Before leaving the delineation of sections, reference must be made to
an alternative method which was occasionally (though not normaUy)
us^d long ago by Pitt Rivers, and has from rime to time found favour
’ ethers, amongst whom Dr. Gerhard Bersu is notable (Fig. 14)*
TOs may be called the ‘pictorial’ method in a sense somewhat different
from that in which I have used ‘picture’ in preceding paragraphs. The
strata arc not outlined (as in Fig. 12) but arc, to a greater or less extent,
differentiated from one another by what may be calW a sort of chroma¬
tic shading. The result is an impression of the section, more akin to a
photograph than to a diagram. The different ‘tones’ arc not consistently
framed or demarcated, nor arc they numbered.
59
SYMBOLS
FOR SECTIONS
ASH
MUO BWJCI ^jjvnr

iVtuoar SAHifJ
iOOS£ fAUm % f i» h *mîk ■ ! **

6î»gt ,/C, j//s>/i'#»\\>
POTJHLUDi
.oas.^.r. MIM I.
7 ^ ■ * «i T **
PA#
C&MfACrÂ^H rAA/o
LOOSE CLAV
BRKXSATS f/f, S
ailmliliiiiliiii) AgÔc^
COMAACTEtAY
SiWACi mMiii
^ R KPsIrt £11=41
Fiq. i3p Chart oC iTrnbolfl adopted by the An:hMeIogieaI Suncy of India
6o
STHATJCkAPHY
Granted thaij in tilt bands of an arcisi^ this Lmprci:ii(iiii&tic ttdmiqnc

has mcrita, ii h equally plain ihflt in the hands of one who b a mert de¬
lineator rather than an anist chaos will prevail. A technique whkh is
beyond the reach of most caccavutors is, on that E^ronnd aJonc^ ruled out
where a workable diemative odscs. 'llie impressionhitic leclmiqtic esn-
rwt be defended on the grounds chat it is ‘less conveiititind"| in iis own
fashion, it is no less convciuioaal ihQB the linear technique? OBly^ ii is
v ns PÔ f
Fjo. 14. 'Ficloiiar method of sccdon-drawing. ^Vfter G. Bcmi
a convention. And if bs* this cruwiling disadvantage: it Icn^

itself 10 nebidousiMSSj to a blurring of detail ood a kek of prediion id
diagnosis. Hven some of Dr- flersn's skilful drawiegs show this, aod a less
skilful performer is liable to be uncrly los&t. Th& is a grave defeen If 1
hive tried ID cmpliasiDe one ihing more thim another it h the need for
considered precasion id Our work- Any mediuni or conveutiou which is
likely to aicotiiwge wiOoUy Thinking is to be deprecated! und, saê ac ite
rare best, the impitsslOfliarie technique Is wooUy. I do not corameod it
TO ihe average worker.
One more (kmil. Il is usefiil, in ihe drawing nf a huge sectiOhi 10 in¬
clude a human figuic as wed as 1 linear scale. The figure provides an
ctfonlcss Indioidon of sipproiuiniite size^ and the saving ofeff&n on the
pan of The reader b worth a little extm effort on the part of the drafis-
man.
61
V
The Layout of an Excavaiion

^cavaLioD is Habit y> dtv-clup into a tihausi of pkn.
and trcQchtfi dif^culi to suptrvist und rttord. And ofrcii cin-
jL by Intrusive spoil-tips thai eventually citlicT canrro] rhe
ti'ork Or me In a constant and cosdy prut^ of secondai^- remnvaL Oti
approaching an excavation tli^t trained observer enn at a glance cvaEitaie
iis efliciency+ It is on axiom that an untidy excavaiion 13 a had one^
whether the untidiness reside in the general layont or in deiaited execu-
Ticm. The guiding principles ate not dliELult: they are "tla™ a planV a
carefully thought-out scheme^ and execute it in orderly fashionn
'fwo contrasted examples will dJustiate thia. The fix^r (PL IV a) is an
ofhctal photograph of a wêll-known exaivadun in rhe Easrii conducted by
an archaoologtstof copsiderable repute and long hdd-cxpericTice, Kever^
rhclcsS} a mere notice might guess, and guess comccfJyp that chaos reigns.
I jmIc at the crowded ^'orkmen, piddng and shoveiling ruomlruously in
lU directions; the absence of u supervisor or indeed of any possibility of
supervision; the absence also of'smallsfind' or pottery reccpiades; and
of coufw a complete Jack of any systemitie idcnrihcaiioii of Strata. Need¬
less TO say, ihe subsequent itpoit fGuthfliUy redacted this cOflccntratcd
cenfusion.
The second ilhistradon (Pi. IV b)j Efom the same subcontinent^ un-
bluahingly represents an excavation of my own, on the principle that
the professor may properly be expected to practise. It shorn a site
neatly parcelled oin in readily controllable areas; small groups of work¬
men are directed by stipcrvisots (disliuguishablc in the photograph by
their sun-helmcEs); the boskei-carriers are w'oridng in orderly pro¬
cession along dear pathways; and in the middle distance on the lightj
the survey-party is cmvcnierLtly at worit at a table shaded by an es$en-
tkl umbrella. The subsequent analysis and detailed pnblicnriofi of this
excavation was a relatively easy task. For ii h not the lease of the merits
of orderly field-work that ii is ediy. Every man knows what he is doing,
iuid records are almost ineviiably clear and sensible, the considered pro¬
duct of sevemi pairs of criricil eyes.
Plan your work, thim, and cncthriKljcally pursue ymir plan. Tme^ there
must always be an dcnicm of chance and of opporturusm In on excava-
THE LAYOUT OF AN EXCAVATION
rion, huwcvcr Cftiufully planoe^l. But scitdrific digging in nut un that

accuimt 'A {ÊUibk, Tlic cxperkuced excavator, who ihinks before be
digu, sucueetb in reaching his objiective in a majoriTy of caica.
The mturc of die plan necessarily depends ujKin the charaeicr anil
needs of the site. For purpose of discussion, wt may here dasdfy nur
problem under duce caicgoflcs: trial'irenehifig or sondagts, aita-
exuivatJoii, ktuJ wliat I Shall call substaniivc iraudiing. The cstaivation
of burials will be cuOSidwed Mparately [p. 95), and isvcsj in vicw^ of
the slightness ef my cssperiance in their ocpioraiion, 1 propose 10 omit
altogetltcrj tlwy require a monograph \¥y one who knom.
(i) TRIAL-THEKCHrNU UF SOf^lDAGES
The eld practice of cutiJng mal-rrcnehea, of making as a

preliminoty to, or even in Um of, anca-aGavadon was frequenLly n Sub-
siiiutc for lutelligeiii chinking and dear aiming.Tt was to a large esteni
^sheoting into the brown' on the ofT-cham.^ of bringing down a bird.
‘ Trifll‘treochcs rarely prove anything, save of the niosi gmeral kind. 1
liavc in mtnd a long and wide irial-tmich cui by an eminent archieoloî
across a famous town-sire without apparent results whilst subsequent
systematic escavarion, initiated on an oliogether different basis, proved
that the ueneb had in fact passed thmugb and utterly £iiJcd 10 reveal
a building of unique ohaiacicr. The answer on a site of that hind is, not
a trial-trench^ bur a methodical arca’-cjicavation, which can always be
discontinued if unpruduoivc but at least rctÛ rahcrcntly within the
area seleered.
'frcnchcs in goicrab save of the somewhar spedaJ Substantive" kind
dcAcrihcd below (p. 63), arc bad for mom than one rcaisom Tliey ^mess
up^ a site. Unless very wide (when they am^ ui effect, cumbersome aren-
ccovadons) they are liable ar any coniidcrable depth to become exces-
rively oordiaed and difftcult to work in, their straiihcation caunor be
viewed comprchcnsivêly and at adequate range, and, obovt allj lateral
cnLugeincnt complicates the record to an caient which endangers its
accuracy* At iViaJdcn Casric in Dorset 1 dug the northerp portal of the
eastern entmnoe in 1935 by trenching, and can still recall the appalling
complexity of my record m the work proceeded and my trenches wi dened-
In the followîog year I dug the southern portal is ^ area by the 'square^
method, to be described below, and both cxcavaikin and record de¬
veloped easily side by ride, without risk of error and without hcadichc.
f On the Other hand, prefudicc agiinsl ireachkig cannot be enended ro
63
TH£ LAYOUT OF AN F3^CAYAT1DN
where ihe prdbnimir): problem ii> soldy and simply to search for a
superfrcially smietuTv. If a Ime ofsndent eoirejidimeniSY for
eicaniplet h Thouj^bi to have passed somewhere thruu^h a cetiaki Udd^
a rnal-îrench rbe fieEd is die obvious mcdiod of proving or dis¬
proving The theory,, and it would be pedanny to pmtsit. The cfaimplc
of Napoleon TITs Colonel Stollcl in rhis respect has already been died
(p- 9). I'he principle here adv-ocated is that trcnchiiig should be em¬
ployed^ HOT m a nonniil method^ bur only when very spcdal drcuin-
(tranecs demand it. Too often does ir form ihc basia of an cxeav'udon to
which orher merhods are more îrable^
(ii) AREA-BXCAVATHJN
. If a site in îfoeeif to have been occupied^ an nrea-cxcavsTion^ nota maJ-
troich, is usually The cffeciive answer. Bur ler us ftrai consider the pre-
roquisires.
An arca-eacavarion must be:
(3) canvcnienily and clearly subdivisible for record and codtioI;
(h) capabte of ca^, prognssive expansion in any direciioa wichom
breaking down or impairing the preliminary datum-LIncs;
(c) capable of preserving for consraoc refereoce at a uiaximum num¬
ber of points complete verdcal sections ttndJ the last phase of The
catcavarion;
Cfl eapahlc, ultimaicly^ of easy inreErarion into a ooncinuously ex¬
posed tegionoJ cxQvanon ^
(e) readily accesriblc at all poinxx for the removal of soil, withouc hin¬
drance from iutrrv'cnmg cuttiligs or tnflic aoosa excavated sur^
faces; and
(f) suffidenily open to the ^|cy to ensure the easy inspeetiaa of well-
lighted secdom at all required depths.
Only one tj'pe of laymn nomiaEly supplies all these needs: namely*
a layout bsued upon a square (PL Y). A scries of squares^ a grid^ dug so
that a balk U left between each pair of adjacent squotes UDtU the extreme
end of the w'ork, supplies oil the six preTequJsttes+ TbeindividuJd square
is a clearly deined ûb-utdi for record and supervision^ supplementary
squares can be added in any direction in accordance with devdoping
needs* withouE afiecting any previous damm; the supervisor retains in
each square [until the end of the wiark) a complete section on oU four
sides of him* tcgethci with such addidorud sections or pait-scctious os
I^LATS IV
A. duc»: ia ifac
B Pi%dplii»i cxcu'iiiciti flt Afffcamcduj Siiiulh Indii^ 1945«

{Sit ^.61.}
i
PLATE V
Layout of an nrea-eKcavation fTflJsUa-Bhir Mouiid,

(SiM p,
THB LAYOUt OF AK EXCAVATION
he may care lo add wiihin the eDcnpusv of [he 5^1101^3. tht scnidficiitiOB
of adjacetir Equami and tbendbre accumuhitivciy of the wtiok site, can
easily be corrdafed and rcoordtd along a nuxnbef of aneriiL IkieSi so rhai
altimaicly ihc harriers between Lht squint can be removed without
1ds$ of v^nical cridcnDc and the whole plao liiid baie> ievd by level; the
bamers or halls provide ready paths of access to the various sqmins,
and Froin them in the &poil-duinps^ and the Kqnnicf, unlike most
tfCTichea, are sulBricntly spacious to Jei in ample li^ht and to provide
elbow-room for interpicladon and records
Hxpcricnet shows that in soils of average stability the horiiôntal
dimensions of a squ&rc should approKimately equal 11:1 auUciFmted
maKimum depth. Thus^ if it is inicndcd to dig to a depth approachtng
20 feet, the squme should be Said out with lo-fuot sldeSr This ratio
allows for die necessary smiis and halfics. SimilaiLy^ a yytoOi square can
be dug lo a depth Of 50 feer, whilst a depth of 10 feet orles^ t^he normal
range in Great Britain) demands only a ichfuot iûare. The smaller she
superilda] area of the square in relation to its dtpLli the better, pmidEd
always iJiat there is ample light and iirorking-toom. With a due regard to
these factors, 0 lo-fooi square may be regarded fls the minimum sub¬
unit.
■'These meaEuremenrs indude provision for the balfcs bero-coi squares.
In most soils a balh 3 fret wiik will ewy any amotim of traffic^ and that
width may therefore be regarded as standarf on dl but the ahnllowTsi
sites, (For itHfoot squares a 2-fOOt balk is usually adequate.) The 3-foot
balk implies that the string marking the top edge of the actuai cutting is
11 feer wihin the perimeter of each square as peg^ out j for acamplc,
the acrua] cutting of a 20-foot square will be ty feet square, and of a
3a-fooc square it will be 27 feet- SiimUrly a io-(bot ûarc with x-fooi
balks will be cut with G-fOOt sides.!
The pegged squares will form the bids of record sad survey, and the
supervisor must emure their inirial exactitude; other wise* oU wm of
complicstiuns and errors will arise- At ach comer of the square a strong
peg, not less ihiLH t| inches square in scaniling and I foot 3 inches long,
with one end pointed, is lirmly driven into the ground, its faces set dia¬
gonally to the proposed square; but the exact corner-point is marked by
a i-inch nail driven vertically into the top of the peg and left projecting
about an inch for the oocasdenial affixture of string Of tncasuring-tape iu
eonnexiod with Survey. On one oaaMn, where the sufl’ace-snil was too
sofi to hold die oomcr pegs with complete security, I found it worth
£b1« 6j f
THB LAYOUT OF AN HKCAVATION
while TO have ihcm sei in concreie, bo imparifinE arc they lo The ^ccurdcy
of ihe liTirh.
Fmm These fiTtd poioTS} the mîrking due of the actual cutdog with
string as a guide to the diggen^ allowkig tbr Ehc necessity balk (see
above), ia a simple sccondaty operation.
The squares ihus pegged out are conveniejatly uiiued by means of
letters in one dtreciion (say^ cm lo west) and by mambers In the other
direction (say, north to south). They will thus be known indivyuaUy as
At, Az, Aî &ic,j HT| 'Fhe appeoprUuc designation must be
painted dearly on the ncamt face of each of the four corner-pegs^ wliich
for this purpoac have been set dlagonikUy in the ground. Thus a peg set
at the juncrinn of four squares will have a differem dtsignadoii on each
facc^ E.g. All As, Bij, Ba. The need for dear and abundant labdling can-
not bo om-emphasized if emt Is to be eliminated from die rccurdsj
pardculajiy oTi a large excavatioiL
In the acTTial digging of a ^qmcCf a piindplc of universal application
in aidiAcojogical cjrcat^don may be stressed: namdyi the use of the
centToi-pii, ThLs is the supervisor's own spcdal durge, and upon it the
accuracy of the general digging in large mcasute depentis. It is a small
cutting, about ai feet square, cut by the supervisor himself or by a
trained man under his eye, to a depth of 1 i-l feet lower than the average
Jevd of the wTJTk. Its purpose is to enable dae supervisor, with a mini¬
mum disruibancc of the strata, to anticipate the nature and probable
vcnicai extent of the layers which arc being deated by his Tnain gang. It
is a glimpse into the future of his sttatigrapluoil wmk. Without it*
adlher the supervisor nor his diggers^ blindly from the lop,
can avoid ihe coufusion of the low-er jKitt of One stratum with the upper
part of the next below it. In other wottbj stradilcafion must, by its
ua turCi always be controlled from the i, e. from the side of the con¬
trol pit, sLccc it obviously cannor be controlled pruphetitally from the
top: v^tkald^ngfint, aftcrwardii must he the rule.
ContiOl-pits must be suffidendy numerous to minimize risks arising
from the uncvoiDess or iatciTupcion of stmia, and sulEcieiitly sm^l in
area to restrict the confusion of evadence which is inherent in their ex¬
ploratory ctalure. The eontrol-pit is Lsdeed a means of conecntratiiig
errors which would otherwise be spread over the whole area. Its evidence
in dctaEi must be used with proponionare dreumspection.
And, above ah* cousEant rcfomce must be made to the straTificadon
revealed by die sides of the r.quare as the digging proccerb. The four
66
sitks nciust be Kmstantlv CQntlated wiUl one another. Any marked dia-
crepancy between th^TtL^ or between the etûiiâleni secrioija in adjaeciiL
sqtiMeHj uimi be CDOsideied nod an erplanarion sought. During the
scucb for an cjtpkuatfon, the supervisor may find it desirable to Kmpend
or restrict the actual digging.
Ttiis lftise$ a further poinin aîn ckf univer^ appUeadot^ In view of
the occasional necessity for a ttrnporary suspeiuiou of iliggiiig in one
square or anoiher^ the director cf an cxcas'adon must alwiiys have in
readiness a sufficieiicy of * reserve-joba^ to meet ccmungcncies Of this
kind. Such TMcrve-jobs may indude the removal of unlmpotumr
suit froin a new square, or the Tddforuement of a gimg en^ged else¬
where upon the diggtt^ of a deep stfatEUU. Eleiiieniber thnt;» when a gang
stops diggings from rwu to she workpeople^—pick-Uiao^ shovd-man andj
in the East, probably four baiket-carrieri—are thrown out of acrion.
And idleness is both costly and infectious.
In the last three paragraphs matters have b«n dealt vrith which arc
not peculiar to the 'square", although they are liable to pmmt them¬
selves on a busy area-dig in an acute form. 1 turn now to the actmi]
iccording of the square.
Be it repealed that a gr*« merit of the ‘square^ mcihod is that it
luolizes both control and record. 'Vhc supervisor's rcspoasibilides are
dearly deliuedi and the area covered by his Edd notebook Is precis.
The b^h of bis record i$ the careful idendfication; embodied in an
acairate measured dratvingj of the stmtigraphy of each of the four sides
of hfe square and of such supplcnientary seciion5 as may be required- As
in all excavadonsT the Laj-cn ait demarcated and labelled with a se™1
number whilst ihe cxaiôn proceeds (see p. 55). Each ride of the
square is also Ubdied with its compass-polixtj which is added to the
index-number of the square i e.g, indicates the nonhem side of
square B3And ride is carefully mciisurfid and drawn to a minimum
scale of i inch to 1 fool, or to b maximum scale of i meb tn 1 foot* or
the incmc equivalents (sec p. 57X A sketch (or even measured) Miction
should also be iududed in the nniebook:; and rdcrence should be made
to relevant EectiDiis not so taduded.
In The proc^ of excavation^ it may be found conveniem somclinKS to
work diafûaliy across the squatcj for smtance, if diagoiial walls are
found, since sections at righl angles to the hne of a wall are necessary TO
avoid a distortion of the ftlevani umita. An example is iliustnilcd in
PL IV Bj where diagonal balks can be seen within the grid-square^^ In
67
the layout Of AN EXCAVATION
tuch the diiagomkl season must of course be corrckLcd wath tlie

side-sectioos and* if accessary^ dr^wn scpararely^
AH "finds^ will be recorded by strata with rcfcraice to the □oucst re¬
corded section* normally with the nearest side of the square. Structitrcs,
pits* or imponont objects* together viith the position of all n]ca.Hiiricd
sections* wilJ be planned cartluliy in the supervisor's notebook in rela¬
tion to the four stm'eyed comer-pegs of the square. 'Finds' will be
classified In the notebook by serial numben* section-labeb sti^tum, and
sketch (see below).
(111) SUHSTANTIVE TRENCHES
lltc epithet ^substantive' is here applied to trenches which are not

merely tentative cuttings made Lnse^h of some lU-defiued obicctive but
are in themselves a definite objective. ]Ui the category are included the
cross-treiKhing of a line of fortification^ to establish their stcuctorai
sequence and eo link it up with the sequence of occupations within the
enclosure, Examples are Olustrated in PJs. VI and VIIA^ one of theca a
revealing section which* in 1944-5* related the stone defences of the
third city of 'faxila (Sirkap) in the Punjab with a dated 'paJace' in the
vidniiy and so threw a new light on the chtonology of this hnpomiit
sirc.^ A funher use of the substantive trench is shown Ln PI. VIIB* where
the method was employed to esEablJsh die stratigraphicil relationship of
rwn ceraciena, represeoiing different cul turesôn thefanikOus prehistoric
site at Haiapp^, in the Punjab, tn 194(5 the two cernecerie$ were linked
by a carefully cut and snn'cyed trench nearly 450 feet long* and the re¬
sultant scerdDH (which has been pubtished) was conclusive. In the photo¬
graph* one of the ccroetcricii h beside the figures in the foreground* the
other Itca amongst the trees in the background* and the lines of datum-
pegSj together with a regular series of crosi-baDts designed to equate the
scLtioQs DU the two aides of the trench* areseetL
In all these auunplea* the principle is the $nme. The ktcral extcol of
the trend! h defined at the outset* and the ddium-lLncs oie
accordingly filed tinec and for all. There is no risk of ihcix obliteration
by any lateral* cmb-likc movement of the cutting* and the recorder's
task La predictable from start to finish.
The Established method of recording objects found in a trench which
It is not iniendDd to enlarge laterally is represented by the foUowmg
extracts from a leoordcr's notebook;
‘ Iftiha, no. 4 fNew Delhi, x^), pp. 41 ff.
68
PLATE VI
SccEkm for ihrcc’4iiii<ffwîsmfl( fecofilinf (Ntaidcn Cmi\i,

i&Mt p. 6$ )
PLATE Vn
A. f&r rei»rdira CTfi3(lln-Sb'itip^ K94S)

p, rfJ.)
E. Similar wtion Unjunc two cinnctay'Srcai (HanppJlr Punjib^

p, firf4
No. Measurements Layer Object Remarks

1V6'X3'-5'6' (2) Brown sand Iron knife (Sketch)
blade 5'
long
III i'4*X2' 5'-6' 2' ® Loose reddish Sperical In a local

earth with oc¬ agate bead patch of
casional patches wood-ash
of ash
The figure in xht first column is the serial number of the find. I make
the practice of enclosing it in a triangle to avoid any possibility of con¬
fusion with layer-numbers, &c. It is inSnediately written on the envelope
containing the object and/or on the label attached to the object, and sub-
sequendy on the object itself; also on an index-card in duplicate (one
for the site-index and one for the index of objects classified by cate¬
gories), which will include likewise all the other particulars of the record.
The figures in the second column are the essence of the record, and are
derived as follows. Before the digging of the trench is begun, two parallel
lines of pegs are laid out, each at a distance of i foot from one of the in¬
tended edges of the trench. Thus, if the trench is (for example) to-be
10 feet wide, the two lines of pegs will be 12 feet apart. The pegs them¬
selves are not less than i J inches square in scantling and i foot 3 inches
long, with one end pointed. They are driven firmly into the ground—
firmly enough to avoid any risk of accidental dislodgement—and are
set diagonally with reference to the proposed trench, i.e. with one angle
facing the latter. In each line they are placed with these edges 3 feet
apart, measured horizontally, not along the actual contour of the ground
(unless that be level); and an imaginaty cross-line joining opposite pairs
of pegs in the two niain 1in<^s must be at right angles with the latter. See
Fig. 15.
Every peg is then numbered clearly on each of the two faces nearest
to the proposed trench with a serial-number in black paint. With the
addition of the Arabic zero, Roman numerals (I, II, III, &c.) are used
for the double reason: (a) that they are easier to paint than Arabic
numerals, and (i) that there is no risk of confusing them with the actual
measurements of the record. The figures on one of the lines of pegs are
distinguished by a dash (O', I', II', HI', &c.). If subsequently it is de¬
sired to extend the trench backwards from zero (e.g. down the reverse
slope of a rampart), capital letters (A, B, C, &c.) are used for the
69
ths layout of an kxcavatiok
successive pegs of the eKTenKion:^ the Imm on one of the lines bein^
similariY distbguished by n d^h (A ^ C, &Cp).
Along the front edges of each line of pegs a Atnng is then tightly
stretched, and is pegged down where this is fieciawitaiod by the contour
of the ground. These strings arc the base-lines freun w'hich rncasure-
menra are subsequently talKn.
P£c 0 nc 1 naa peq b

' 'S-a_a
r-ar ^
1 !'
[
TRÊNCN
. . , -..r.,....
WQHT \ i
; ANQIX \ \
^-tf -i-. ^ ^ f-
) ^ ^--
■ V
4
Wf£10* pm f w a' KQ m''
Fig^ 15^ Layout of a trcoch for thrcc-dimenuemal rreordioff
So much for the preliminary layuiit of tbc tTcnch^ The supervisor tniuit
now be provided with an auglc-meajnirei made lightly bin lurongly of
IW'O ^-fbot or picfcnbly 4-f(K)t aims gtaduaicd in feet and indies :ind
fixed firmly to eauh other at right anglEs fPI, XVI [ b, 11). On each arm is
fixed a bubble-level. The other uonnal measuring-instrumeiits—lapCj
5'fit)0t rule, attd pltimb-boli—cumpletc the equEpment.
Wlien the digging of the txoich is in progress, the position of ev'^
significant find b jueasuxed in the following sequence:
(A) LoTigitJidinal fmassiriimfH* The point at which a line at tight angles

from the main dacum-slring do the object cuts the former is measured
along that line ftoui the last preceding peg. The point in question is
obtained by meani of the angle-mvasuie, with such extemion of the out¬
ward aim as may be required and with the assistance of a plumb-bob.
Thus if the point be at to feet 4 inchea from ihc zero peg^ It will acmally
be measured from peg 111 fo fm) and will be recorded as 111 4*.
{B) OtiTKimi mtraifirrtmnl. The distance outwards^ at r^ht angles to the
darum-string, to a point vmieally (by plumh-bob) above the objcci is
with thfi Erngle-meaitire;, |e%'elJed by |u bubble-level and exten¬

ded by mentja of the 5-foCit rule If necessary. Note thar ibe rt^fiasuremeni
l9 recorded Lioni the diltuiii-iiCnne> iwf from the ftctuid side of the trench.
4. Tht^j if the measuremenT is 2. foct ^ inchca, the distaiicc of the ubjcCl
Ijojtii the side of the itench will be ahcmi i fam 5 Inches. Tbc recorded
fncasumncnl h aillxed to the lousitudliial measurcfncnr with A mulriplEca-
tina sSfin: thua in tbc preaent ciuimple the jacaswemcats so far will be
ni
(C) nuasmfffifftt. This tepresents rhe vercicaJ depth of the
object beJow tbc Icicl of the datuai'-string at the intciêcrioa cstthhshed
under (A) obave, Ir li obtained by tape (or rule) and pliaab^bob frotn tbc
Jevelled itrin of tbc angk-measiire, and it ii added to tbc rcccid with a
minus sign. If the depth is 6 feet 2 inches, the toal record will now read
III r 4" X 1/ 5*—6' 2'-
In the thkd admm i$ ooied ihc number alkrtcd to the layer—ii num¬
ber best showm in a cifdflj a symbol which h in practice useful fur dis¬
tinguishing laycr-fliunbeB froia other figures—and the descriptive woid
Of plirase by wbicb the layer is named. These facts sire of impoctaneCt
Since the wtaferih/ in which an object is found k usually cenain and pro¬
vides a cJiecX upon the meaûrernents^ which may sometimes mislead in
ixicguIiXT strata or near rhe junction of two strata.
Tlic /(wriA tfijiwmn k sclf-cxpknatury'. The fifih i$ useful for
additional inforULatioii snd, above ahj far a sketch of the object- tven a
bad sketch k better than none at aU.
It is scarcely neccssaiy' to add that the utility of such a r^Td—or in¬
deed of any stratigraphical record—^1$ propomonate m the accum:y af
the measured secrion or sectioti^ with which the record is subsequoidy
m be equated. The twô sides of a trench are rarely identical, and it will
nearly always be desLrablc to prepare an arairatc drawing of bothi to¬
gether wiih occasional cross-seedons. Moreover, during the actual dig¬
ging it IK sometimes useM to project cenain entcgniies of objects (from
the thTcodimtnsional record) on to the aenial sidQ of the trench by
m^s of kbeUed or oolouKd pegs. 1 have known occasians upon which
such a visual tepresenladon of 0 distribution in tbc actual trendi has
been illuminating and con^'incing.
71
VI
The Excavation of a Structure

tOM the more general questions of technique we turn to specific
problems. How would you excavate the buried remains of an ancient
J- building? ‘Find a wall and follow it’ might be the obvious answer.
But the previous chapters have been written in vain if the novice who has
read thus far is prepared to accept the answer without protest. He will by
now appreciate that to follow a wall in all literalness would be to destroy
the related evidence upon which its interest in large measure depends.
Let us consider in some detail the nature of that related evidence.
Unless a structure is dated by a contemporary inscription or by un¬
impeachable documentary evidence or (exceptionally) by its intrinsic
charaaer, our knowledge of its date or cultural context must be derived
from the stratigraphical association of objects of recognizable types.
Furthermore, the specific character of the strata themselves—^whether
resulting from construction, destruction, decay, or other causes—will
throw light upon the vicissitudes through which the building has passed.
Only the most careful excavation and observation can recover such
evidence with suflBcient exactitude for u§e. The dating or cultural setting
of a building is based ideally on three categories of objeas: (i) those sup¬
plied by strata which accumulated before the building was constructed;
(ii) those supplied by strata contemporary with the construction; and
, (iii) those supplied by strata subsequent to the construction. Categories
(i) and (iii) bracket the structure chronologically or culturally, whilst
category (ii) defines the point within the brackets.
! To imderstand the full meaning of a building in decay, it is necessary
to understand how it was originally construaed. I have accordingly
urged elsewhere (p. 133), for this and other reasons, that a knowledge of
building-construction be included amongst the qualifications of an
excavator. In its simplest form, a masonry wall is built normally as fol¬
lows! A trench, known as the ‘foimdation-trench’, is cut along the line
of thfc proposed wall in order that the foimdations of the latter may rest
upon the solid sub-soil rather than upon the relatively unstable soil that is
commonly found on the surface—whether natural humus or disturbed
‘occupation-earth’. Rarely, when the surface-soil is itself solid, or
when the building is a slight one, or when the building is jerry-built (an
72
THE EXCAVATION OF A STBCCTUHE
AfauDimstl crime in andcoi rimes), the rouiidarioa*trench may be abseiti;

an the olber hand, a foundBdan'trcneh muy actually be Ctit even inia the
live rocit> In order to give a. level jicatiiig for tbc wall< The jpteliiniiiaiy
BSjumptioii in all iDsanccs must be tliat a ibuodarioTi-Trciieh is
present.
This feature is of great impDri]iiu:e> and oU care musT be rahen in its
idcatificaticiii. Obviously, layers into which it is cur antedate it, and must
be disririgti^thed bcjtjnd ull doubt fioni layers which aceumulated
agatnn the struemn: and ihemfoic pt^tdate or are oontcmporaiy with it.
In loose sandy soU, the foundation-trench will be V-shaped, with on
appreciable space between its sides and those of the initial wall built In h.
But in a stiff day sub-soil tlie sides of the trench may be venical and may
have been completely filled with the foundarions of the wall- Much may
depend then upon the character of the wall-fadng when cleared of ihe
sodi 6. trench-built wall will necessarily have an uturinuned, rdaiivcly
rough face, whilst a wall built 'free', and tbotefore oeceasible to the
mason, will genersdly show a mote careful emming and trimming. Kced-
less to say, a wall with any sort of plaster ‘rendering' Of facing, or with
‘struck* or ncariy finished mortar joiming must have b«n buih ftcc, can¬
not have heem Lrid in a foundarion-^Tcnch. A wide experience of difTcrent
miTeriats ia necessary for the idcntilifaiion of a foundarion-timLh, and
the mull of the mvesrigarion mu-Hr always be noted orcfully, with dia¬
grams, in the notebook.
After the building of the wall to a suitable height, the Sides of the
foundation-ucach tif V-shaped) will be packed and tbc adjacent surfecc
levelled, if necessary, to carry the floor, Subscqucmly, the utilizarion nf
the floor will result in wear-and-iear, possibly patching or renewal, smd
probably the accumulsUon upon it of an 'occupajio^layw', consisting
of extraneous ruatefial'—heaith-ash, mud, fcod-debtis, broken pottery,
lost Dinamcnis or coinage^'S^hicli may help to daw the period or periods
of usage. Later asain, disasrer or decay may cover the occupaiion-laycr
with building-debris (indiciiiflg, inddennilly, somflMng of the diuac-
tcr of the supeistructiire)^ and eventually the whole rite may be catpeted
with vegetable-mould.
Let us take an example. Fte- >6 A illustrates the accimnilaaon of
stram under, around, and over the wall of an ancient sinimure. On the
right-hand side of the wall-section, over the lUrurU sod, two layers
(9 and lo) represent vrllagcHiccupaiioii of Culture A, wi^ posi-^l»
iudicating wooden huts and associated with potsherds, spindle-whorls,
73
(
.1
s
THE EXCAVATION OF A STRUCTURE
&c. Into these layers is cut a shallow foundation-trench to take the

footings of Wall Y, and the flanks of this trench arc filled with layer 8,
which is also spread (on the right) as a basis of Floor i. Layer 8 contains
only relics of Culture A, but one or two relics of Culture B are imbedded
in Floor i; and the superimposed layer 7, resulting from the occupation
of the building, represents Culture B exclusively. Over this occupation-
layer, a new rammed floor (Floor 2) is laid, and on it is a further occupa¬
tion-layer (6) of lesser extent, still containing objects of Culture B but
in a somewhat evolved form. On this occupation-layer, a cascade of
bricks mixed with burnt timber and clay (layer 5) indicates the destruc¬
tion of the building by fire. Thereafter the stump of the wall is used as
the foundation for a mud-brick wall (X) of lighter construction, associ¬
ated with an earthem floor (layer 3) containing relics mostly of Cult^
C. This new structure represents an intrusive culture of inferior quahty,
immediately preceded by the violent destruction of Culture B, and may
(if the evidence is found to be typical) be interpreted perhaps as a semi¬
barbarian supersession of Culture B in an evolved phase of the latter.
On the left-hand side of the wall-section, the same two pre-wall strata
(9 and 10) are continued, but are superseded at the wall-level by a well-
metalled street (Road i). This metalling is renewed at intervals (Roads 2
and 3), the upper metalling being inferior to the lower and suggesting
a decline in municipal standards. Finally, in association with the mud-
brick wall (X) of Culture C, metalling ceases, and traffic gradually weare
the street into a hoUow, removing earUer strata in the process. This
process of road-deepening may be seen in many Eastern villages of the
present day, and serves to remind us that the passing of time may be
represented stratigraphically by denudation no less than by aggradation.
Fig. 16 B shows the unhappy consequence of the wholesale clearance
of the trail along its two faces. The relationship of the waU with the
adjacent strata has been lost beyond recaU, and the sequence ideated
above is irrecoverable. Excavation has devolved mto irreparable des-
So important is this matter of section-interpretation, that I wiU give

another example, dug out of an old report on tiie excavation of the
Roman amphitheatre at Caerleon in Monmouthshire. In the early suges
of this excavation, conclusive evidence as to the date of construction was
hard to find, for the good reason that an amphitheatre, being a pUcc
merely of visitation, was largely devoid of the sort of
may be expeaed to produce dated ‘finds*. As the work proceeded, this
75
7^
Pl.A’l E VlII
A. *GhfMi' wa.II Kt Verubtnium, HenfordiWtt, i«i.

7^'i
B VrfuUrtJlum, ponl. jj»m: p«t (rf the pbn by ekerina trenche*

nmsk by will'robbw.
(^4i ^ 7^)
THE E3iCAVATI0K OF A ITRUCTUftlE
dcfidcQcy id a variety of way&j and Fig. 17

illuaixatej one of them. ^ . ■ 11
In thii ilJusirsilion, the waU in section dd ihc left is the extwwl wall
ofiheainphiiheairc. It Is hnilt ondMpWtifis, the upper part of wJudi
is munaied whilst the lower part Is <Jiy-bui1l. The fuouags <»^py ^
foundaiion-trench, partly vertical and Ûy V^hspi^,
three dark occupatiun-loyeri which therefore antedote the consimctiotu
We ihni come to the faced waUiiig, the base of which is m wnwet wdi
a road-surface that had dearly been laid down in eoimexio^n with ^
The section goes on to show- tlmi this mad was subsequently t™ew^
on at least fourLasions. Now logically, the date of
the amphithcatn: plus the cailirtt mad should ê b«n >
the datfi of the maierid fam tJie three pn^mad occupation-bjem Ji^
U the subsequent road-tepaits. The p^mad
fated their funaion admirably by produaug
ditnting tltat the amphitheatre w^s not earlier ^
mad-râirs were unproductive, and the later end of the h^ ^
Tm r^cd open, ê section did not ou that account
be observed that to the right of the aoiphitheaK-w^l “
secrion’l built in a tienf h which has been cut through the s^ thi«
™ 2
therefore exactly oontemporaiy with ihe amphiUiea ,
preUraimiy “ riftoinM inw"'*. cr»»-«cfi«n bdiiii

bur m its neighbours. Only when an
dosely cxaimned is achieved and i««ded
agreed stramin by stratum. Even SO, h is wi« to
ctcuvarioTi.
77
If we ium frtjm gcnc:nil prindplc^ to f hdr pâcticaJ application, thjcre

is l™ toom for dogcm. In aH cases rhe ^square* mcibod wbJeh I have
Fecoroxnended fur Hrcu-cxcaviiiinns (p. 64} is suitable, and on sites
where binJilIiigs of varying pkns are piled one on top of the Other, as oa
maay Ensiem sitca, no oihci: rnerhod lends itself so reodUy to clear
systemadc record^ But an 3 shallow site, where the excavator is con-
fjoated wiih an -casendally urJiary strucrure) as on $onae Romano-
British sites, it may be found convenient to mke the individual rooms ofa
building os die baiiis of record. Even so, it Is safer to begin on die grid
plan until the dimensious of die probEcm are apparent; and heviag
begun on the grid 1 should mysdf hstitatc tn change horses.
'fwo or three other poinis may be dealt with briefly.
‘ghost'WALLS
^ The excavator may have the mhifomine 10 light upon a site where the
walls ha^-e been completely dcatroyed by stone- or brlck-robbcrs in
ancient or modem limcs^ A notable example of this is provided by the
Romaoo^Bridsb site of Verutamium where Matthew Paris records ibai,
amongst others, an elcventh-ccxitmy abbot ' turned over tlic soil to a
considerable depth that he might Iiqd masonry structures" which he
‘reserved for the fabric of his L;hLlrch^ The excavator of the site found, in
fact* evidence cvcTywhem of sysiemaric brick-robbmg, and Itemed to
recover the plans of bmldings by carefully following the mbbm'
trenches (living inteimlttcnr cmss-secdons as dcscril^d abovn). Par
the most part, the robbing had been so skilfiil and economisd that the
trench was scarcely wider than the wall that it had contained; so that
f^^hen, after the removal of the wad, the imwomed debria was thxown
back and the site levelled far cidiivadon, the mixed filling^ provided a
perfect negative of the plan (PJ. V [ 11 a). By cutiing across the debris-
filJed trench Qt frequent intmals, it was often possible to recover ihcltnes
of the former building with accuracy, and even, in fsiVDiirablc Instances,
to equate sonae of the siratihcatiori with thenL Perhaps the mosr remark¬
able example of the recovery of a plan from "ghost" walls of this kind was
afforded by the north'We^t gateway of Vcnilomium where, in ig3i. Miss
Kathleen Kenîmi found that a great parr of the Structure had been
almost completely removed bin was able, by the aueful cemoval of the
debris from the forinCT wûll tFcnchra, 10 reveal the whole laynm in
negarh-e whifex piêrving the stratified floors intact (PL Vtll a).
73
THE E3tCAVAT10N OF A STRUCTURE
S^TRAJaHT JOINTS
The jiiDCtiOn of two mlfes is oficn a key-point tn the anieiTfTEtaiioa of

a plan. If the wails arc well and truly bonded together, with idcniJad
ma^nry Eind coursingi xhey are ob\'lousLy of the &amc huiJii and date.
There b, hawevcTj a *caich* for whieh it behove# the e^vatot io keep
u vigilant watch. Uy rhe tchiovhI of#ome of the factng-stoncs of an Older
vîJl it b possibleio 'ipoth-in' a younger bmdCh-'ft'all with sotne tsemblancc
of bonding i but the difference in stnictiual period is usually betrayed
by differences in marcrial or coursing, Ot by ibe iin:ompleiciie$5 of the
bonding.
On the other hand, the absence of bondiog—a 'straighi joint*^—does
not neecssaiily imply a differtnee of date. The medieval builder, it «
true, was mually eoPCemed to that all biS walls wTre properly
bonded; the close iotegxation of a Gothic building, rhe interplay of
srresscs and stiainSp denundicdl xhis. Accordinglyji in a rnedieval stnlC*
Turo a straight joint is notmally aignificaiit chTonnlogicaLEy. Bui in the
more stadcarctutecturt of earlier periods a straight joint may be a matter
of IndifTcience or even of choice. Where adjacent walls carry very
different vcitkal loads, it may not be iaadvisabk io leave dicm to 50ifie
ciicni free of one another. Or, in the daj^ when the plan ofa building
consiated, not of a blue-print, but of the acnial foundations planted
on the ground under the direction of the architect or master-mason, the
main lines may have been laid dnwn fir^t and suhsidiaiy feature# added
as the work proceeded. One or other of these cuplanaiiofis may be sought
for the fact that the rnwora of the Roman fortrei# known os Burgh
Caadc in Suffolk show a smughl with the fortress-wall up m a ccr-
turn hdghtj above which they are bonded in; ofj itmllarly, that the
toweri of Vemlajnium air built free up m ■ height of 6 feet and are
thereafter of one build with the town-wall. In both imianoca, lowciS
and wall are csscncuilly parts of the mjhc design. A inc^re remarkable
instance is fhc sTsicinatic use of the stmighi joint in Kebuchadnewar s
great Ishtar Gate at Babylon {sixth century^ S.t). The fimndariom of the
siruaiire are of a varying depth, and (n quote Koldewty:
It li concdvabLc that ibose parta of the wall where the funindutlcmi ore
ftpecially deep do not iink w much m the courac of time ai thou of
shallower fmmditionsj imd Knlancnt is unavoidable eves with these,
standing aa they do upon earth and mtuL 'flius wliere the foundatknu are
diislmilar there must be deavoges in the wdlj* wM^h would leriouily
endanger the itabiliiy of the building. Tlie BahyloniEiii toreww this and
yy
guarded against it. They devised the expansion joint, which we also make
use of under similar circumstances. By this means walls that adjoin each
other but which are on foundations of different depths are not built in one
piece. A narrow vertical space is left from top to bottom of the wall,
leaving the two parts standing independent of each other. In order to
prevent any possibility of their leaning either backwards or forwards, in
Babylon a vertical fillet was frequently built on to the less deeply rooted
wall, which slid in a groove in the main wall. The two blocks run in a
guide, as an engineer would call it.*
TIMBERWORK
In airtight peat or day or alluvium, or on the other hand in a very dry

climate, timber will retain its form and something of its substance for
thousands of years. On the other hand, a solid timber such as a railway-
sleeper begins to deteriorate in the British climate after about ten years.
Andent woodwork found in Egypt or in the deserts of Central Asia is
dated in millennia, and the rough timbering which Dr. Grahame Qark
imcovered with great skill in the peat at Seamer in East Yorkshire in
i949"5i ^ much as 9,000 years old. Dr. van Giflfen, a master of
timber-excavation in the sympathetic soils of the Low Coimtries, found
remarkable vestiges of wooden buildings—posts and wattle walls—of
structures of the first century a.d. in the Roman forts at Valkenburg
near Leiden (south Holland).^ But for the most part, the excavator of
andent timber struaures must content himself with less tangible results.
When carbonized, their vestiges may last indefinitely; otherwise, they
are generally a mere stain or a more or less loosely êd hollow in the
earth, and have to be treated substantially like the ‘ghost’ walls of which
I have spoken above. On the whole, they are not an easy problem. The
stains and hollows may sometimes be missed even by the expert eye,
and a timber-structure should never be explored by the novice without
the closest supervision.
We may set aside carbonized timber as a relatively straightfoiward
matter. An example of the more subtle kind is provided by the first-
century sheds uncovered within the Roman fortress at Richborough
about 1930 Mr* J* P* Bushe-Fox. The sheds had been built on
sleeper-beams, some of which had been laid on uprights, and Mr. Bushe-
* R. Koldcwcy, The Excavations at Babylony trans. by A. S. Johns (London,

1914)* PP. 36-38.
* 2S-28 Jaarverslag van de Vereeniging voor Terpenonderzoek (Groningen,
1948), with English summary.
80
PLATE IX
A. Roman timber-building at Richborough, Kent, traced by coloration of the soil.

(^See p. 8t.)
B. Burial of about 2000 B.c. at Harappa, Punjab, showing outline of wooden

cofhn.
(Set p. 8t,)
PLATE X
Tti£ss <i>f hurJk-cotuirucdiifl in ih£ tuoUihk long-barn^w m hlunElIcby*

LlnooLnihirc.
p.
THE excavation OF A STRUCTURE
Fox first ■wiig-fiiji'H them in the faintly diiscoIouKtl lines and pits which
bej^n to show up lA the light-oolouicd sand of the Site ■when the gtouud
was cleared under suliubde weather cundJtions. The colourod lian were
formed by the earth which had occupied the ntatrices of the vanished
Tintbcnnut, and vertical hollowa, similarly filled^ showed the funner pre*
scncc of tlic posts (PI. IX n). By uireftilly soaping and scoioninE the
Kite, the excavator was eventually able to unoover a scries of tlwse build¬
ings sud to reconsmjci them (on paper] with a substaniinl measure of
oertainty.
The same mcihod, with viuiationsi must minutlly be tbllowed m
work of this kind. Cenaio soils, notably gravel, ate more nciuctam than
others lo yield thdr evidence. A sandy sub-soil is tisoally the most
responsive, and at Hitrapp$ in the Jhinjab, In a cwnciety dating from
^ 2000 B.U., not only was the outline of a coSn visible as a brown
Mdine, but Dr. K. A. Choudury, of the Indian Forest Research
Institute, was able to phoiosraph and identity with complelc oeiiai^
the imprasiimi of a speerfic ritoher (deodar) in the tiutm of sand. *1 he
coffin (Pi. IX b) was tbond firsi aa a dark stain in section and was then
carefully dcaied from above, the Smal Clcanmce being effected by gentle
soaping and lightly brushing with a sioall, soft bruih. Tlw somewhat
similir outline of tt coffin was traceable round a Lite Roman burial at
Venilsmiumj’ and Other examples ore recorded from Britain.
Ir would, however, be difheuk in End more biiiliant cxposiiiom of
timber-cxoivaiioa than those affoided by the w'oAof Mr. C. W, Phillips
at the Skeodlcby long-bariow b LincoliBhire and the f^nus Suiioo
Hoo ship-builal in Suffolk. The general method adopted in the caiava-
lion of the long-barrow will be died b a later chapter (p. loi), Here we
arc concerned only with the hurdlC-work which was incorporated in the
mound and was reveakd b a titnch dug dose lo and parallel with its
lun^lLadlool aJtis. in tbe cxcavaiofs mrds.
the irciich WM 5 ft. Ml width, and an oamlnaliim of its aidrt soon showed
a senrt of what were oJlcd for the sake of cijnventKias ^^'crtiailidca’ at
irrcpilar Lntcrvals, Th^re vaûc vcrtic*l hrcaki m the lUhjmnci: of ihc
barrow could ordy be Intcrpmcd m die tmees of decayed barriei-ft of some
kind, prc:rumibly of wood. A havy min fell on ihc mght after these were
nOTcdi £tnd bnmght away a thickncis of between 6 In. and l ft- of malcrial
ffi>m tilt side oflhc irench towardi ihc upper side of the barrow. Tbb fall
had taken place alon^ a definite division In rhe matexk! cif the harrow along
< WbedcTj Rtpert (London^ i93^j Pl' ^
an St Q
THF E3tt:AVATrON OF A STKUCrURF-
ita iruJdk and it was at once apruutnc, from die ImprnîioiiS uf ixp^
flghi poau and pjc£n of wood Ibccd between tliciPt fhac Mine sort of a
fence had been ict up here heTore the barrow tuul been builtt ontl then had
been buried in iL. . » » The reason for the 'venicaUdea' bceiimc de^ at
nnee^ for they were tracca of simiXiir fences setting off ax «bout right aeigics
ffojn the centnd fence In the direcdon of the lower side of the barrow
m. xL^
All this evidence was of course in "negative^; the soil was toualiy im-
IkYCunble in ihe presm'âiioTi of rhe amial wundwork.
Aeddenr helped in this disetivcry but in no way detraers froin the
asruicness of ihe cxeavatnr^s □bscrvaiion. The reatuic thus revealed was
substantially an nddidnn to know^ledge and will doubdcss In due
course^ imo a new stnictunl and ritual jMllcm of tlie neolithic But at
Sutton Hoo accident cm daim no share of the ctediL The general dr-
cumsTances of this astonishing di^coveiy ate well known: about the
middle of The seventh cenmry a.d. a omn of high rank liad been interred
with hhe richest treasure ever dug finni British soil^ in a ship SOOae ^
feet long) which Imd been buried under a mound in a gteat Ittiich cut
into the ^nd*
Since the whole ship mud its conicnts had been involved in sand for
ftome li30o year^j it Is nor surprising that there were virtually no tetnains
of wood except for soitdl frdgtuent:) which were found here and there
preserving their form if not thdr eharmetcr through ix7Ulmct witli rusted
iion, .,, Tmeea of the eiclsience of wood could frequently he seen in the
form of Uiin layers of discoloured saiiJi asS4Xiaied with blinds of more or
IcsE the same miccrial leached white by the actiQii of ihc melds liberated
in the decay of the wood. The normal colour of the sand on the site was
yciloWa except where it weu the product of the decay of turf or had been
under decaying wood. , . . No dilheulty was found m distingublliug the
blhog of the irench from the undisrurbed fiond even where the colour was
the âme tn each case. The fdllng dropped uwiiy easily from the sides of
the trench and foUowiug it was merely mcdumicaL^
In tli«* ciraanstanew, the escavodoo of the buU was a parricularly

delicate operation, and esttacts ftoni Mr. PhilUps's acoooni of it may be
quoted. The work which had been done before he took over dirccdon
ahewed that none of the wood of the ihip rotuined, ihough all the nunier-
ems cleneh-noiEt which had held Ini pljnklug rogecher were exactly in
^ hexxv (Londofip I93£)j 6a.

^ Anti^rnty^ xiv Ct940)i 10,
THK BXCAVATlOK OF A STRUCTtfRK
pluct, the ship having been tvBtpIctely filled up with u muvli uf the sajid
excavated fjonl the rrench u oniild be pot in agoio. These coadtliens WcfC
ideal for the pirwcfvtidon of iitA form, and It was dear rhnr this only way to
carry ooc rhe woiic woa cut a very wide ircn-ch right ilirough the mound
which would give ample waika un each of ihc buriaJ-ttenizh. The
faces of ihc cutting went stepped backwards in HrfaoH and their facci
tunberKl up^ and the bout was careTtitly emptied uf uil corifctitj begmninji
with the burial chamber. . . .
Since all but the burial chamber am hud been filled in directly with
Band while the ship was still whote, none of the cicflch-naili which held
ihe ship togcdier CMuld move from rheir ptnce even when the wwd which
they scimrcd had disappetued. By careful work from rhe inude it wait
possible to remove all the ™tcnE of the bciar wiût displfldflff any of
the naiJsi which rciTUdncd in their placiea on the sides of the cicavafions.
This process wes aided by A change in the cunsistepey of ihe sand which
was to be found where the boat's tifubers hod onec been. A diuty blacklah
layers utccoinpiinied by some Ictiehed sand, could be felt for cirefully by
slnwly shaving down the Band, and wuminfi Riven of its approach by
the appearanec of the bfifibr red patches signalling; the near presence of
tlench-noils. In this wTiy ill die boat whicii survived wai emptied so that
the face of die excavation evcrywJiere was rhe sand which had pressed
agauiHi the ttnhm of the boat from the outiddc, and which somerimes
still bore In rceognixubk form rhe imprint of the grain of wood. . . *
n,
ITjc ship had twcnty-Hihc ribs_Hie preservadon of the ritM was bad.
All that remained... was o dSrry line of Hind ninning across the ^tiip from
jfuiiwalc to gunwale keeping close lo rhe huUj oiui frequendy still rcteming
a marked rectangular craas-seojon. In wme cam the sard had fonned u
rough east of the derated wood so ttiBb if this w-as breached* discolcni^
sand tended m pour out of the hole leaving b rectangular-sccuuncd cavity*
In excavating The ship these rib traces were left cflclowd in strips of sandn*
These examples of timber-cxciiîidoti must suffice. The lOOli most

generally useful are a biMd-hladed knife, a small spike of sortie son^ and
a soft palnt-brnsh. In the clcar-Liicc of the Suxtod Huo burial-chamber,
a stout shovel mounted on the end of o long ash handle: was used fbr
shaving down and removing the iand on a horixontal plane in a search
foe the si^ficant changes of colour.
cîht huriil chamber, ibidH p. t5a. A limplcr but norwtheleii tjumienve wt of

ptoWrtns atUiidrd the cacivitiwii of the Qaebcrg stop in Sweden. TTie student
is referred to the account of theic (in Engtlth) iii A. W. Brug^tk Mii rxu ,
IfukoxL Sehnclig) UititTg/ftn^ti 1 1917)1 3<59 B.
83
the ElCAVATtON OP A STRUCTURE
MUn-SKSCIC
In the. djy -diinaic of p9Xt$ of Aftics. and bricks of onbaked mud,

ofren integrated by an adtnirrurc esf straw, Imve long b«n used for
building, and may pr«^t some difficulty lo ilK catfavîor, though
grncraily of a fonnidablc narure tJum that pici-eiitod by ranished
dmberwork. A writer on apchawtogical technique ha$ wrinen:
jThtt principal difEtTjlty faced by an archaeulopst in dealing with mud-
brieks lies in the fact that the nwlerial used in their oonstnicUon is ikjt
unlike that of the earth in which they have been buried for crentuncs. As
a result tlicrc have b«n a number of cases where long narrow rooms
nimed out to be iic^vy walls; in Mh« words, the tacavatort reeognired
a Elighr change in enlour or copsbtency of the ground during tlw digging
but removed the wailR and left the loonti uncaca^'aiedl i
1 used hardly add that this uohapi;^ expexittiC^ was derived Tnainly from
that land of arehacolngitail sin, Pêstiiic> and reopens a vista of income
pctcace which is by now sufficiently fanuilar 10 us.
.The pfobkni once mote h thai of imelligcni supervisioni though the
trained w*orkraan is here an asset of olfROSt eqtiial worths The texture of
the earthj the fed of it, the sound of it as the pick or shovel aiiikes it*
are all feaors which, almost equally with direa visual evidence, tell the
eitperienctd digger when be b or Is not on a mud-brick wall. The in-
stjuoted peasant learns to think through the point of hts pick or the
blade of bis knife. Decayed mu J-bridc^'ork may fill the inierior of a
room and present on plan an undiferendated surface where it and a
mud-brick wall^top abut upon each otlier. But tapping carefully and
obliquely with a small pick wiU often produce a disringuishabk hard
note when the point, pemnrating the hiling^ strikes the face of the aetuiil
wall- Scraping the sur&oc caitfuUy with a knifOi sometimes aided by
damping} may reveal a shghi but significant line w^hcre filling and wall
meetj or between Individual bricks. The composition of ihc mud wall
wfi.1 sometimes mdJeate wbciheir it represents brick or filling: ficcks of
charcoal and ncraps of patiecy rarely occur m bricks bm are not infre¬
quent in filling, and 1 imvc sometimes found them La ihe coarser mud
ivhich may be used as the equivalent of mortar. Their presencej there¬
fore, ia a lucful guJde^ tliough thdr absence proves nothing. Petrie adds
that ‘in ihc list resource the stuff should be tearched with u magiufier
to SBC the hollows left by deeomposed straw: iti kneaded brick these hol¬
lows He Lq every direction; in blown dusr and wash they lie nearly all
horisontal*-
84
Tracer ifw Suiion IIm SufTalk-
PLATE XI
PLATE XII
Section thnnt^ the muii^hridi dcrcnect cf the Keraji^ eitadclj Punpli,

built befort 3«o ilc
t.W/i tfjf j
the excavation of a structure
Tht tradng of k rniul-hrick wbU imy Llius be 3 Icagdiy
process^ iovolviDg carefuj mspccriaii and difq|[kiS in dlHcTent lights and
di^CE^t degrees of dampness nr drynic^ of soU^ 'When the wall-face iuw
been definitely located, further eurc xnifSt be taken to aseeitain wberher
It andenily ^nmdcied' wiih a mud or plo^tCT fadng. If nni^ it can
be gently brushed hoiizoiuaLly or scraped with a knife^ When thus
newly cleaned, the mdividiuii bricks will frequently show up dearly, and
they must then be eDcnpLetdy recorded tsize^ bondi quality, icmitiiig);
for the chances are that^ as the will dries out and hleadiia in the sim^
much of the detail will be lost. In order m remin the general pictiir«i It is
sometimes possible to reduce the mud-joints slightly by gentle brushing
or scraping^ and so to leave ibc separate bridyi in shallow relief. t>n the
other hand> if the joints sue of harder^ coatser indterialj as they some¬
times arc, it may be preferable to sentpe the smoother suriaec nf the
bricks^ thus leaving the joints in relitf (PI. XIIJ^ In ciihcr ok, this
work should noinvally be dene after record^ ijot be^te^
In all other respects, a mud-brick wall should be dug exactly os a
Qusonry walh with a similar system of CrO$$’'Secdons through the atten-
dontsuatau
VIT
On Digging Tnwn-siics
T He rtafitml $equci fo 2 chuptCT ODBfnjctuifcs is iJk oonsidcTLiitjn
of the iUiiCCtifal oamplcjt of a mwn-aiic. (The pioblaiiM of it
tnilitary site are icchmcally similar,)
The site of un uncient tmni may today he a lolembly level inii.t
of countryside, or it may have piled itaclT in the couisc of itg« into
ihc form of OO arcifidal mound OF tfH np to itJO feet or more in. hdgiit,
Somethtn* has been siud of the liner ptoceSS elsewhere (p, 53).
Here the question is one of UtOkal approach 10 the probtem of exca¬
vation.
Ixt us consider what mfontiatiOD we want. Our primary need is w
asccnam Uic chtonoloBy and cultuml Miring of the Site. Without these
rwn basic facts, iw smicnual layout will mean little. But they in turn
will mean little wiihoul a knowledge nf the plan of the lown, of its
donKadc economy from period W period, and the social and pohtical
condition of its inhflbitania. These problems are necessarily interlocked,
bur they cats to somie mdcni be separated for the purpose of pielimitiaiy
iovesogarion. . j
On a level site, the lechniad approach is reladvdy straightforward.
Them the cidtuml and chronological sequence can be rccor'ercd in intr^
ductoty fashion by one method only; by a TiKtnctcd anm-excavaiion in
the interior of the town, preferably somewhere near its ccolie. Although
the rule is by no means invariable, diics commonly grow in a reasonably
symmemol fashion; the origbal dvic centre is likdy w remain the
focus and to tcpcesem therefore the maximum depth of accumulaied
occupation. This accumulaicd oocupotioa must be probed carefully to
the boitom, preferably by means of one or more Rquarcs—0 group of
four adjacent squares is often a useful uait-—in accordance with the
method described on p. 64. on no account should this procedure
be repeated at random. Fciric cannot be gainsaid when be protests
against the multiplication of trial-pits. Such pin, 'if they hit anything of
importance/ he remarts, 'arc likely to iaiure it, and certain to destroy its
connctioia with Other ibingii. Frendi explorers have a love bi/tiirf qu^-
quttJOTTdagat B proceeding which often ruins a site for aystematie work,
and which never shows ihc rpcaning of the positions or the noturc of the
6«
ON DIGGING TOWN-SITES
plan’. On the other hand, a rcstriacd probe, carried out by methods

more subde and thorough than Petrie’s, is a necessary preliminary stage
on a ‘flat’ site of unknown potentiality.
A secondary if not simultaneous stage is the investigation of the sys¬
tem of fortifications which normally characterizes an ancient town. The
moments at which a town built or rebuilt its defences were manifesdy
of special importance in its life-history. The construedon of a fortifica¬
tion may indicate the newcomer in a strange land, as in the mottes of
Norman England; or it may mark the achievement of full dty status, and
may reflea a period of peaceful consolidation rather than of military
duress, as in the walled to^ns of Augustan Gaul or Antonine Britain. Or
it may reflea the advent or threat of a formidable rival or of political
anarchy; as when anxious edias of Honorius urged cities to build or
repair their walls and authorized them to use statues, altars, and temples
for the purpose. It may have been designed to defend the populace, or to
control it; as, p>crhaps,in the almost feudal citadels which dominated the
Indus cities of HarappS and Mohenjo-daro in the third millenmum B.C.,
or certainly in William the Conqueror’s Tower of London. Fortifica¬
tions thus not merely outline the town-plan (or some part of it) but may
focus and express the city’s vicissitudes and something of its sociology.
Further, the charaacr of their brickwôrk or masonry is a fair reflection
of the economic condition of the dty at the time of construction, of
wealthy and leisurely dvic pride or of more slovenly necessity. And to
what extent were the defences maintained?—were there long periods of
immunity during which maintenance was allowed to lapse? Finally,
were they riven by an attacker, as were the mined and countermined
walls of Dura-Europos on the Euphrates?—or did they crumble into
decay as a counterpart to economic decline? All these and other ques¬
tions are of the first importance to the inquiring antiquary or historian.
It is along the lines of the defences rather than in the buried relics of
the bazaar that the cxcav-ator may expea first to recognize the major
moments, the framewôrk, of the story of the site.
Let him therefore, early in his work, cut across the line of the forti¬
fications at scleaed points where it seems likely that the evidence will be
most comprehensive. And let his cross-trenches be both wide and deep,
no mere oner’s bites—wde enough to escape acddenial features and to
provide ample room and light for observation; deep enough to reach
down into the natural soil and so to ensure that the story is complae.
Furthermore, let the trenches be carried far enough into the town to
87
relate the defensive system in its various phases with the successive
occupations of the town itself.
The accepted three-dimensional method of recording such a trench is
described on pp. 69-71.
To the sectioning of the defences should be added the careful excava¬
tion of a gateway, where successive road-levels and guardroom-floors may
be expected to amplify the architectural evidence and to lend it precision.
The completion of these works—the examination of the defences and
the central area-excavation—may be assumed to have given us a reason->
able conspectus of the site. The next stage is less prediaable in detail,
although its general purpose is clear enough: namely, to recover a repre¬
sentative part of the town-plan at various periods or, at any rate, at the
topmost period, and to ascertain the character and economy of its
various types of building. The best course will probably be to extend
the origin^ area-dig at the level of some specific stratum, to link it up
with the excavated gateway, and to recover the intervening layout of*
buildings and streets at the seleaed level or levels. Thereby an orderly,
coherent development of the work is ensured, and future excavators will
know readily where they stand in relation to it. This was approximately
the method adopted by Sir John Marshall at Taxila (Sirkap) in the Pun¬
jab, with the most revealing results (PI. XIII a). The extent to which
such work can or should be carried subsequently downwards depends on
the funds available and the current state of knowledge. Although I
would not accept Marshall’s methods of excavation and record, my view
is that he was perfectly right in imcovering, and preserving uncovered,
only the two uppermost strata of Sirkap in the area tackled by him. He
was dealing vnth an entirely unknown culture and polity. On the other
hand, in a Romano-Brirish town I should today normally advocate the
complete clearance of the available area in depth, even at the expense of
extensive destruction. As so often, no general rule can be postulated;
each instance must be considered on its merits.
If we turn from a ‘flat’ site to a telly say, 50 feet high (PI. XIII b),
complications confront us. Probing from top to bottom in a mound of
this depth becomes a major engineering feat and involves a dispropor¬
tionately large opening at the top. Moreover, there is not infrequently
the risk that such a probe will over-emphasize the latest and possibly
best-known phases at the expense of the earlier and less-known. Let me
give two examples.
Some 18 miles north-east of Peshiwar in the North-w’est Frontier
88
PLATE XIII
A. Air-phoiograph of part of the Parthian city of Taxila, Punjab, in the first

century a.d.
{See p. 88.)
B. A typical re// (Tape Sialk, central Iran). (Sca/e indicateJ ^ t/ie sma// figure on r/te
right shou/der of the mound,)
{See p. 88,)
ON DIGGING town-sites
Ptoîticc of Pakistan, near the village of ChEnada, is a famous graup
HunmdB which represents the andem apiQl diy of Puskalis'ttil, set
athwart one of the great trade-rowMS of ancient Asia. Its strcngili was
Siich that cveo the veteran army of Aleiatider the Great took 30 days to
capture it ia ilS b.c,, and sudi was its prestige that AJettatidcr went out
of his way to receive its surrender iu person, tu ilte seventh ceniiityA.i>..
when the Chinese pilgrim Hiuca Tsang visited itj it wm yet'wc
peopled*. Today the vast site is detelicr, bittm inio and divided by the
Streams of the Swat liver, hui its lajfgesr mound, the Bill Hisllr or High
Fort, still stands rn a maximum height of about 100 feet. Hcte i n
the then recently rtM.-onsdtuted Archaeological Survey of India '=^™
OUT its Erst fflcatatinn. The summit of the highest mound was Mle«M
for the work, and trenches were dug into it to a dcpili of so feet. The
result was the di-wovery of soaps of medieval and later builihngs of
familiar Sikh and Islamic typ« pewiblc ftagments of some¬
what earlier histotic dates. Nothing that was nor predictable w« brought
to light, and the real pioblcms of (potentially) one of the most un^riMU
sita of Asia were not touched. The young director, fresh from AtheM,
his luiDii dOubilcM obsessed with the prestige of a rowenng acropohs,
had drried out the csoivatimi iu ih* One spot wliere n inaxmitim accu¬
mulation of relatively modem maicfial could
Similarly at anoihcr site, far away on the steppe of Afghan Turk ,
there lie the great mounds whkh represent the ancient
of Odes', carfax of trans-Asian nade-routes and outpost of Hcllaujm
m partjfcm. BeCtm was already a diy in the fourth cenreiy
sn^d from the savagery of Qlingfe Khm
A n Today only the dny vUtege of Bolkh lies withm the immCM cuemt
of its deidkt walls, and Mvend square mdes of ^
Asia. mdc«l of the world, lie accessible to the jj
Accordingly, in 1924-5, Monsieur A. Fouchef, to
hononr is due as a student of Buddhbt an.^««
series of excavations on behalf uf the FmiKh
Afghanistan. The results wtmr, 05 he fredy ««1 muvely
apînting to a degree, Why? He chore for h« operation the liUth^t
^ oTSe towerioB tfU which reprerenm die ^
and dug dovm metre after metre into the lums
mosquând a ‘Turkish’ hath! True in a
7 miles there was room enough for eiror: Fouthcr ^ J
‘We compared omsdves. my wife and U to tww ants rent mto ihe depths
89
OM DIGGING TOWN-SITES
nf the futile to carry out an autopsy on tlvc corpse Of an dcplumt.^
Bur vL'hy^ f^m so viidc a pf}asib1c dioice, did he chwse of oU diiogs ifac
uhimare summh nf the dtadcl? In a letter to his Paris correspoiident, he
retls us: he writes^ cannol e^pe lh£ magic putver of namely
and you yourseir if T cnuld have comultcd you fioni iifar and had abked^
as in Hermni: ^' WhErre should I would ineonteatabty have ic-
pNcd: *Ôo the Acmpolt!-*’^'* The incscapuble magieDf nuinesl What a
superbly Gallic reason for ptungin^ imn the une sipot in the whole cf
llactra that a less charmingly Mntimental investigator would unhesitat-
iugly have shunned: the one dommani s~pol which could have beeu pro¬
phesied re bf!ar upon its heights the latent and mos t familiar of lucdlcval
buddings.
Koj a little cold-blooded dclibrradon at these two sites would easily
have pointed rhe proper n'ay. 1 happen to know them both, and min pre¬
pared wiihoui undue Tashness to prescribe. The premisses are that both
sites mreteh back into a rcfnoie hut undchned antiquity; that both have
also been wholly or panially occupied in Islamic rimes; but that^ sinec
our cjcicnsiw krtow'lcdgo of Muslim arehitecnire is unlikely to be en¬
hanced appiedably by the ocvsiiy cTccavarion of ihe buried fragments of
a medieval citadel, This phase may pmfirably be by-passed. The oui-
sranding Importance alike of ChUrsada and Maikh lies in their earlier
phases^ when they were metropolitan centres of Asiatic trade and meet¬
ing-places of orieotai and occidental cultureSn To reach these earlier
smta the approach is oot in fact difficult. AtCMrsada the artaefes of man,
weaihcTj and water have^ over a large part of the principal mound, re¬
moved the higher strata to a depth of some 40 feet. In other wtjrdSj. Sikh
and MusUffl may safely be presumed to have vanished here, leaving ihc
pre-medicval strata exposed to immediate attack. There, at the eastern
foot of the High Fort, is the obvious spot for an area-excavatiou de¬
signed to reach the ancient Puftkallvati and to achieve^who know's?^
another Mobenjo-dJiio at its ba^. At Balkli also ii is not difficult, by an
examinadon of the sides of guUies and other cuttings, 10 Hud a a umber
of plac^ w here the familiar top-stulfis abseui, and where penetradoii to
the earlier Icvds can be reasonably assured. But at Balkb there is one
Dhvinus and immediate goal. Nearly half of the defensive sj^siem shows
only ZsbiTuc wwrkmamhip and may for ouf purpose be ruled out, at any
rate in ihe initial stages (brer, it too should be fully recorded}; but a
■ Ar Feutlicrj fjj VtJiift Ri)yt^ Ptndc Jt Bkik-frd Foi^rTu Ue Ea
Pekf, Ardi. co AliduriiiEin^ Firii, 1942^^ p. 9B.
90
part of the remainder is of composite character and obviously incorpor¬
ates earUer work. In particular, along the southern side the Islanuc vraU
is carried by a huge rampart which is itself, to judge from slight indica¬
tions, a complex structure. Here is a major problem for settlement.
Until the excavator of Balkh has dug a great trench, recorded m three
dimensions, through these southern fortifications and liêd their suc¬
cessive phases to occupation-levels in the adjacent interior, he has not
got to grips with his problem. ^
To turn from these two examples to the general run of tells is to turn
to variable phenomena, susceptible to no fixed rule. Much depends upon
the height of the mound and the extent of its erosion. As a general prac¬
tice, however, it is desirable to bite boldly into the penphery at some
carefuUy selected point, penetrate the inevitable scree, and establish the
successive cultures with, if possible, the successive systems of fortifia-
tion. This desideratum is commonly realized by /WZ-excavators, but
there is an unhappy habit of nibbling sporadically at the flanks of the
mound instead of contemplating a methodical, coherent cutting which
wiU, with ordinary luck, link up and sort out many important faaors
once and for all. An csublished method of cutting such a section is to
step it from top to bottom, so that the gangs have convenient workmg-
platforms at intervals and, above all, so that the lower stages of the ,mt-
ting are not unconscionably deep. Nevertheless, there am few pubhshed
J, wto . of *is kmd has b«» ™ sh^ a
complete section through the margin of the highest te/Zof Aechalcohtluc
IndiK dty at HarappS in the Punjab. Its total depth is 50 feet but, 0^
to the compacmess of its material, I was able to m it
stepping. Incidentally, the recorder and draftsi^ completed th
wo?f in detail, foot by foot, as the working-platfonn
lowered, so that, when natural soU was eventually "^"‘'ed the
and the recording ended almost simultaneously. As usual, the log-book
record was three-dimensional (cf. p. 69). ^ «
Apart from the marginal bite, as on a ‘flat’
gateway is highly desirable, combined with an :^-dig in the adjacent
SLor on aSdently large scale to esublish
text of successive gateways and fortifications. Further in the •weno^ an
elevation may indicate the position of a diadel or te^le and ^ ^
subjeaed to an area-excavation, which should subsequently h^^
up systcmaticaUy with that adjoining the gateway, and thereafter con¬
tinued to the designed limits of the enterprise.
91
The ultimate goals of a /^//-excavation should be (a) to establish its

culniral or chronological range by the marginal sections indicated above,
and (b) to uncover completely a specific phase or phases of its occupation.
I emphasize this point after seeing many incredibly messy /e//-excava-
tions where everything has been attempted at once, with cuttings and
soimdings at varying depths and the whole site permanently wrecked
or obscured by tip-heaps. Let the rule be to finish a clear-cut scheme
which can be transmitted lucidly, on paper and on the ground, to future
generations of investigators. Leave your tell for others as you would
wish to find it. Too many excavated tells of Asia are, to adapt Petrie’s
phrase, ‘ghastly charnel houses of murdered evidence’. Fortunately,
there are still some thousands of intact ones awaiting a more humane
and legitimate execution.
92
VIII
Burials
‘Where necessary, the dolmens were blasted, the
circles of stones were removed and the dstvaens
construacd with large flat slabs were made available
for study.* Mysore State Archaeological Department,
F Annual Report for J94^.
rom the living we turn to the dead. The varieties of burial-ritual

devised by man in his care alike for the departed and for their sur¬
vivors are legion. The body may be buried intact or after cremation,
it may be left awhile to the weather, the kites, and the jackals, whereafter
a few representative bones may be collected and piously interred. Burials
may be communal or individual. The dead may be thrust with no durable
equipment into the earth; they may be accompanied by the treasury of
Sunon Hoo, or by a sumptuous army of attendants as at Ur; they may
go naked to St. Peter or with a harem to Paradise; they may be enclosed
in a humble pot or in relays of costly sarcophagi; their grave may be
marked by no lasting monument or by ‘a mound on the hill, high and
broad, by wave-farers wide to be seen’; or their ashes may be tipped into
a river which will bear them happily from our ken. Fear, habit, a httle
affection, and much affectation go to the making of these things. It is
fortunate for the prying archaeologist that his technical problem is
often a rel^vely simple one. His difficulty begins when he comes to
reconstrua the ritual represented by the particles of ^dence uncovered
by his skill (for an example, see above, p. 3); and if he steps beyond
ritual to its fancied significance, he will scarcely find the answer in the
cartload of books which, says the Chinese proverb, contains all wisdom.
Here is no place for the anthropology of this vast subject. From the
technical point of view only, I will deal in turn with certain broad cate¬
gories such as are known to me as an excavator or observer.
ROUND BARROWS AND CAIRNS
Round barrows (of earth) or cairns (of rubble), ranging in Britain from
the end of the Stone Age to the Viking period and abundant at various
periods in many other parts of the world, provide a useful starting-
point. They may be a few feet or more than 50 feet in diameter, and their
93
BURIALS
present altitude shows an equivalent variation. But, whatever their size,

their excavation can, in a majority of instances, follow one of two alterna¬
tive procedures. The best introduction to them is to emphasize how not
to proceed.
The day is not long past when barrow-digging, or rather barrow¬
hogging was a polite amusement on a summer’s day. Something has been
said about this in Chapter II (see also PI. I): and it will suffice to recall
how in 1849 Dean Merewether ‘opened’ thirty-one barrows in twenty-
six days in the Avebury district. This is probably a world record, but
there arFseveral runners-up wliom it would be waste of time to cata¬
logue. The method was to plunge downward into the centre of the
mound in the hope (often, al^, fulfilled) of finding a primary or secon¬
dary burial, or both. The crime lay, not merely in the neglect of all the
structural and ritual minutiae which are a large part of the interest of a
barrow, but that, working downwards in a central cavity upon the
strata of successive structures or burials no investigator, however ex¬
perienced, could correlate them owing to the simple fact that they were
largely removed in the process. Not, of course, that Merewether and his
kin had any notion of stratification. Nevertheless, even after stratifica¬
tion had begim to be dimly understood, the practice of driving a narrow
trench through the middle of a barrow long continued. I remember see¬
ing an excavation of this kind which in fact revealed the primary burial
but—so narrow and dark and dirty was the trench—failed to discover
a surroimding ritual ditch 6 feet wide, although the trench passed over
it at two points.
The central pit and trench are therefore alike ruled out of court as
methods of excavation. Nothing in archaeological excavation can be
more definite than that.
Now for the more positive side of the matter. The imderlying prin¬
ciple is that the whole barrow shall be removed and, whenever possible,
subsequently replaced. Tins principle was laid dowjiJby-EittRiyers and
cannot be gainsaid. At any pomt m or about the moimd, structural
features or secondary burials are liable to occur, and nothing short of a
complete record can nowadays justify the excavation of a type of monu¬
ment which has in the past suffered so grievously from piecemeal work.
A satisfactory method to this end, adopted, for example, by Sir Cyril
Fox at Ysceifiog in Flintshire* and by myself on Dunstable downs in
Bedfordshire* was as follows (I quote Fox):
* Arch. Cambrensts, 1926, pp. 48 ff. * Arch. Joum. Ixxxviii (l93i)> ^93 ff*
94
BURIALS
Two parallel rows of numbered pegs were driven in i ft. apart on either,
side of the barrow, at right angles to the face on which it was proposed to
work. As the removal of the barrow advanced foot by foot, the face was
maintained in a straight line between pegs similarly numbered, and on the
discovery of a deposit its position was fixed—firstly, by measuring its dis¬
tance from the appropriate right-hand peg along a line stretched to the
equivalent peg, and secondly by measuring its vertical position above or
below the original ground-level. In practice it was found that positions
could in this manner be fixed very easily. The centre of the deposit marked
CII on plan [Fig. i8], i ft. 6 in. above the floor of the barrow, was thus
quickly recorded as 25': i8'-f-1' 6''. The scale of feet on this plan indicates
the position of one row of pegs.... As the work proceeded the floor was
tested at frequent intervals and holes dug when any indication of distur¬
bance presented itself.
In this clear description it is of course implied that a carefully measured
cross-section of the barrow is drawn along every face which revealed any
feature of interest. Fox in faa published seven sections of the Ysceifiog
barrow (Fig. 19).
At the risk of repetition, I would add the following details. First, the
datum-pegs must be substantial and firmly driven in. They must also be
broad enough to bear a clear, painted numeral (I prefer Roman numerals,
both as distinctive and as easier to paint). Thirdly, a two-inch nail must
be driven into the top of the peg at the exact point of measurement. In¬
cidentally, this nail is useful for holding the loop of a measuring tape,
and the transverse string for each successive cutting. Fourthly, a con¬
tinuous string along each datum-line from nail to nail is useful for inter¬
mediate measurements. Fifthly, it is convenient to record in the fashion
already recommended for three-dimensional recording (above, pp. 69-
71), with ruled columns arranged across two pages of the notebook.
The great merit of this system of excavation is that the excavator is
working throughout in the fullest possible light and comfort, and the
recording is rendered exceedingly easy. Its demerit is that, unless special
and rather cumbersome arrangements are made, the records arc ex¬
clusively parallel with one axis of the mound, and the other axis is
not adequately recorded. In a majority of instances, this pardahty docs
not matter, but it might, .
The alternative method in general use meets this objection. It is
known as gucarteriT^ or the quadrant method. The mound is marked out I
into four quarters by two strings, laid preferably to the cardinal points of
the compass and over the approximate centre. Opposite quarters are then
95
BURIALS
excavated in turn, a balk i H feet wide being left between each quadrant
in such a fashion as to give a complete transverse section across the
mound in both directions. The example here illustrated (PI. XIV) is, for
convenience, taken from a level site (a bunal-pit at Brahmagin, Mjrsore
Fig. 18. The Ysceifiog barrow, Flintshire, showing the layout for ex-
cavation. After Cyril Fox
State), on which the system shows up dearly. It will be observed that

the balks are staggered so that the two halves of each transverse secnon
are in the same plane, although one half of each is necessarily in reverse
in relation to its fellow—z detail easily adjusted in the final drawing.
Recording by this method is rather less simple than by the other. Each
quadrant is numbered or lettered or named by compass-point, and it is
96
PLATE XIV
A. Excavation of a burial pit (first or second ccntur>' B.c.) at Brahmagiri, Mysore

State, India, showing the ‘quadrant’ method.
{See p. 9^.)
B. Silver coin of Tiberius, dated a.d. 26-37f from Chandravalli, Mysore State,
(5« p. t2S)
5eCT10hf5N^'lTb6 AR£ PARAtueU;N9^ 15AT RKJHT AKS-tSSt&tTlKC
% SBCTIOK at 20 FEET A^B ON Pl-AN
2>. SECnOK AT Z6 T££T C OM PLAN ctfr\An««Z

rcatmr,a#«s: t>m>Tn»enM«
3. SECriOK AT 33 FEET E.-F OK PLAK
.3ECTroKAT 3r TEET G—HoKItAK

OMIm OfuvtfftMT f i tsnonKio
Fig. 19. Selected sections across the Ysceifiog barrow (nos. 5 and 6
omitted). After Cyril Fox
H
6670 97
BURIALS
desirable to fix a line of pegs at i-foot intervals along one of the cardinal
lines of the layout. From these pegs, the usual three-dimensional
measurements are taken in each quadrant until, in the ultimate phase
of the work, the balks are themselves cut away. In the example illus¬
trated, it was easy on the level site to lay out a measuring-tape in lieu of
pegs.
Apart from these two standard methods of excavation, other methods
have been improvised in recent years in emergency-work carried out
under a time-limit on doomed mounds. With skilful handling, some of
these improvisations have produced important results, but they are not
on that account justified save in emergency. In particular, there has been
some reversion to the axial trench, subsequently expanded about the
centre of the mound. With a builder’s bulldozer in the ofl5ng, anything
saved is better than everything lost; but, though this obsolete method be
then as inevitable as quarter-day, it is just about as desirable. The two
standard methods are the best to date.
Whilst no attempt can be made here to catalogue the varieties of
structural or stratigraphical problem which a burial-mound may present,
a few of the more common features may be noted as a guide to the be¬
ginner. The primary burial is usually central imder the mound as origin¬
ally formed but not necessarily as later enlarged. It may be a hole in the
ground, with or without a small mound or heap of stones inunediately
covering it; it may be in a wooden (dug-out) coffin or in a stone cist,
which may be ‘large’ (3 feet or more in length) or ‘small’ (under 3 feet)
and buried beneath or standing upon the natural surface. The stones
should be examined for intentional markings. Secondary burials may
be inserted into the original mound, or placed upon it and covered by
added material, or even inserted into the added material; the shape and
extent of the cuttings containing secondary burials are therefore vital
pieces of evidence, to be looked for and recorded with the utmost care.
The area containing the original burial may lie within a circular ditch,
which may either be covered by the original mound or lie alongside it,
and may either be continuous or interrupted or, significantly, approached
by a ramp. The shape and character of the ditch—whether flat-bot¬
tomed, and whether beaten and hardened as by ritual traffic—must be
considered and discussed. And the bottom of the ditch must be searched
for material which, on analysis, may show whether the ditch was ex¬
posed for any appreciable time, or whether it was covered immediately
after the ftmcral (cf. p. 47). Alternatively, a ‘processional way’ round
98
BURIALS
the primary interment may be built up above the natural surface, as,
apparently, in the Pond Cairn at Coity, Glamorgan, brilliantly recorded
by Sir Cyril Fox.‘ Here, the burial and, as it seems, an initial infant-
sacrifice, were covered by a cylindrical stack of turves, and around
the stack a circular space about 5 feet wide was enclosed by a sur¬
rounding stone wall 15-20 feet wide and perhaps 5 feet high (a south¬
western feature in Britain). This interspace appears to have been used
for elaborate ritual, and the excavator’s discussion of the phenomena
should be studied by all barrow-diggers in whatever part of the world
they be.
The edge of a barrow or cairn may be retained by a stone kerb, or
revetment or ring-wall, up to 18 inches high,* or may be anchored by
earthfast buttress-stones.* How far, or in what manner, timber may have
been used alternatively round British barrows we cannot say until many
more of them have been completely excavated under modem conditions.
The practice is known from Holland, where Dr. van Giffen has excavated
examples and in Holland also timber circles were sometimes set up
within the mound. Three barrows in Britain (on the Yorkshire Wolds, in
Lancashire, and in Montgomeryshire) have produced analogous evi¬
dence,* but whether, as one excavator thought, they represent the dead
man’s hut is mere speculation. The whole matter deserves further in¬
vestigation in the field, and timbering should be very carefully looked
for, not merely at ground-level, but also in the stuff of the mound itself,
in all future barrow-excavation. The timber inside certain Indian
Buddhist stupas, which have been compared with round barrows, is
an analogous problem.
LONG BARROWS AND CAIRNS
Long barrows in Europe represent a group of burial-customs pre¬

valent in certain regions approximately between 2500 and 1500 B.C., i.e.
at the end of the Stone Age and the overlap with the early Bronze Age.
Like round barrow’s they vary widely in form, size, strucnire, and usage,
but in general they were intended for multiple burial, normally by
* Archoiologia^ Ixxxvii (1937), 142 ff.
* Ibid. Ixxxix (1943), 108, 110 (Llandow tumulus, Glamorgan, excavated by
Fox); and Froc, PreMst. Soc. 1938, p. 112 (Breach Farm, Glamorgan, excavated
by W. F. Grimes).
* Ibid. Ixxxvii (1937), 134 (Simondstown cairn, excavated by Fox).
* A. E. van Giffen, Dia Bauart <Ur Einzelgrah€r.
» Grahame dark in Proc, Prehist. Soc, 1936, pp. 30 ff-
BURIALS
inhumation though occasionally by cremation. As their name implies,

they are distinguished by the fact that one axis is longer than the
other, but the difference may range from only a few feet to a third of
a mile.^ It follows that no uniform procedure can be commended to the
excavator.
The first long barrow scientifically excavated—and one of the few—
was the famous Wor Barrow on Handley Down, Dorset, excavated by
Pitt Rivers in 1893* The published photograph of the fimshed work, in
which 'the figure standing at attention in the middle distance marks the
site of the central interment’, is itself a monument of the Geoerai’s
scientific discipline. Save for four pyramids of chalk and turf left to
indicate the original height of the mound, the barrow and the Ming of
its peripheral ditches have been completely removed. The pyramids
represented the axis of a broad trench, 45 wide, which was cut
at the outset centrally along the length of the mound, leaving the flanks
for subsequent clearance. Objects found in the ditch-filling were recorded
three-dimensionally, and in the report were projected diagrammatically
on to two schematic sections of the ditch (Fig. 2, pp. 12-13).
It is no detraction from the outstanding merit of this classic excavation
to suggest improvements of method. In the first place, it is a primary
desideratum to record a complete longitudinal section of a structure of
this kind; i.e. along the original axis of approach into the chamber (here
* Thus a long barrow on Thickthome Down, Dorset, was 90 feet wide and
only no feet long. On the other hand, the monstrous ‘bank barrow" in Maiden
Otstle, Dorset, was 60 feet wide and no less than 1,790 feet long. True, a writer
in Arch. Joum. dv, n, has disputed the identiheation of the latter as a barrow
and prefers to call it (as indeed the excavators nicknamed it) a neolithic race¬
course or cursus. But (n) more than 500 feet of it are stiU a mound and cannot
have been used as a ctfrrui; (b) seemingly comparable long mounds in Dorset
(e.g. in Long Bredy and Eroadma^me parishes) are stiU mounds from end to end
and were never ‘racecourses’; (c) the fattened portion of the Maiden Castle
structure coincides exactly with the extension of the Iron Age Camp and is sus¬
ceptible to explanation by that fact; and (d) the Pentridge cwrji^, which has been
dted as an analogy on the ground that a small long barrow is incorporated in
one of its margins, bears no resemblance to the Maiden Castle plan. The Pent¬
ridge cursus is over four times as broad, and to say that *if the Pentridge curstis
vs?as reduced in scale and the offset long barrow placed between its ditches
centraUy, the two structures would not differ materiaUy one from the other’ is
to exceed the limits of legitimate hjrpothesis. To assume at Malden Castle a
racecourse completely blocked for at least a third of its length is like assuming a
bottle of which one-third is corkl Admittedly, we know very little about any of
these structures yet, but it is safer to stick to facts as they present themselves,
however astonishing they be.
100
BURIALS
of wood). This is particularly necessary in chambered barrows, where

the record of successive entries can best be shown by a section from the
entrance or forecourt inwards. (Supplementary cross-sections are of
course also necessary.) Secondly, under the greatest height of the mound,
i.e. along its longitudinal axis, internal structures (as at Skendleby,
p. 8i) are most likely to be best preserved, again demanding a central
longitudinal section. Thirdly, it is likely that the two longitudinal halves,
of the mound will be approximately symmetrical in structure, and hints
or evidences in the first half dug can be followed up and verified in the
second half if the halves are dug in succession, not as at Wor Barrow
simultaneously. Fourthly, in an extremely irregular neoUthic barrow-
ditch a series of carefully drawn cross-sections is needed. Pitt Rivers s
projection of all the finds, some drawn twice the scale of the ditch and
others only one-third the scale (Fig. 2), on to a purely diagrammauc
section is useful but is no substimte for an accurate picture.
There is indeed no reason why a mound such as Wor Barrow should
not be dug by the quadrant-method already commended for round bar-
rows. Only when the length of the barrow greatly exceeds its wd*
does that method tend to become cumbersome. Whatever the method,
however, the first step is to mark out, and preferably pegout,thebaiTOw
longitudinally into equal halves, and then to tackle one half only ata time.
As an example of the excavation of a longer barrow of this kmd, I ^y
again refer to the very successful work carried out in 1933 4 Y •
C. W. Phillips on the Giants’ Hills mound at Skendleby m Lmcoln-
shire. Mr. Phillips describes his method as follows:
The first step taken was to lay out a rectangle 180 by 80 ft. roimtê
barrow and drive in a strong post at each of the corners A
was run from the nearest Ordnance bench mark, smd a local bench i^k
was cut on one of the posts. A contour survey of the barrow was then
“fn the case of barrows with large ditches there are two main lines of
investigation which may be pursued. First, there is the actual barrow with
interments^ any original structural fea^- « may have^
Thfs gives information about its builders, its value depen^g on *e
See of preservation of the contents and the absence of disturbance
Sefondly, there is the evidence of the part which the
played in the life of the distria since it vras completed. Some of this may
be recovered in the form of stratified traces of casual human secernent
in the ditch before the shelter afforded by it was destroyed
the case of a long barrow ditch of normal size this may usuaUy be rehed
lOI
BURIALS
upon to give evidence down to the dose of prehistoric times. There is the
further possibility of intrusive secondary burials in the body of the barrow
or in the ditch.
The season of 1933 was devoted to establishing the second of these two
points by digging a large section of the ditch on the upper side of the
barrow. In 1934 the barrow proper was excavated, and as much more of
the ditch as was necessary to prove its character all round the barrow.
In opening the barrow a central spine was left intact right down the
middle to show the original height, and this was carried across the ditch
at the east end. At the same time 5-ft.-wide balks at right angles to the
central spine were left at stated intervals.
It was originally expected that the east end would contain the burials,
but, when a complete excavation of the eastern third of the barrow had
shown that this was not the case, a 5-ft.-wide trench was driven along the
central spine right through to the west end. In this way the burial area
was approached from the side, and, when found, fiiUy opened out. Also
the remarkable hurdling in the western part of the barrow [see above,
p. 81] was revealed and studied, further cross-cuts over the whole of the
barrow being made at various points to establish the section and find out
about the hurdling offsetting from the central ‘fence’. Another investi¬
gation was the careful clearance of the revetment-trench found across the
eastern end on the old ground-surface. Trenches had also to be run
along the sides of the barrow to follow up the post-holes which were
found to flank it along the greater part of its length on both sides. The
whole of the extreme west end was also excavated to find out as much as
possible about the eight posts which were found to have stood here,'
In all this admirable work, the excavator would probably himself be

the first to admit an element of opportunism, due to the novelty of the
evidence revealed, the immensity of the task, and the consequent neces¬
sity for selection as the excavation developed. But the whole project was
well and truly based upon the elements essential to a long-barrow exca¬
vation, namely; (n) the provision of firmly fixed datura-lines; (i) the
preparation of a contoured survey before excavation; (e) the maintenance
of a longitudinal section or spine during excavation; (d) the thorough
exploration of the ditches; and (e) the careful removal of the (greater part
of the) mound. The structural hurdles, first identified at Skcndleby,
and the marginal posts will now be looked for in all long-barrow excava¬
tions, and the search for them can therefore be legislated for in the
primary layout.
> Archaeohgia, btsxv (1936)^ 42-44.

102
BURIALS
A further word in regard to the contoured survey of an unexcavated
long barrow. The contours may not adequately indicate the actual ex¬
tent of the mound, particularly on a slope. It may therefore be nêssary
to add a basal form-line, differentiated from the contour-lines, to indicate
the approximate line of junction between the structure and the natural
surface of the ground.
The two long barrows which I have chosen as examples are both de¬
void of the stone structures—chambers, passages, peristaliths, entrance-
complexes—which characterize many of the kind. To these the general
principles apply: above all, the restriction of excavation to one half of a
feature at a time, and, wherever possible, to work upon the longitudind
axis or ‘approach axis’ as the base. Thus in the excavation of megalithic '
‘port-hole’ dsts in India—i.e. large dsts with a small circular entrance
through an end-stone—half the dst was excavated first, in such a fashion
that the section bisected the port-hole. By this means it was possible to
ascertain exaedy what portion of the filling had been inserted through
the port-hole, and so to reconstruct a feature of the ceremonial usage of
the tomb. The procedure need not be further elaborated; nor need the
method of digging and recording the sections through the ditch, where a
simple adaptation of the three-dimensional method described elsewhere
(p. 68) is applicable.
‘flat’ cemeteries
Where burials, for example in um-fields, are not nmked on the sur¬
face of the ground, their investigation becomes an ordinary area-excava¬
tion of the kind already described (p. 64). Within the squares of the pd
it will obviously be necessary to leave additional ‘keys’ or cross-ôns
to interrelate certain of the burial-groups vertically. Otherwise, Ae mam
problem will be that of isolating the individual graves and plpning thea
oudines by very careful horizontal digging, often with a knife or trop .
In Haling with inhumations, the excavator may advisedly l<»k particu¬
larly for discolorations or impresses which may indicate the former pre¬
sence of cofiins or shrouds. See Sir Leonard WooUey’s account of ê
cemetery at Ur,' or the reference above to a coffin-bunal of the Indus
avilization at HarappS (p. 81), all or mosUy dating from the third mU-
Icnnium B.c. It is manifesUy important to determine whether burials
overlap (and so indicate perhaps a culture-sequence) or have otherwise
• Ur Excavations II: the Royal Cemetery (London and Philadelphia, 1934).
pp. 137,165,184, &c., and pi. 14-
103
BURIALS
been disnirbed anciently. Care must also be taken to discriminate be¬
tween authentic grave-goods and objects which may happen to have
reached the grave with the infilling. No rule can be laid down for this;
much must depend upon the precise position of the objeas in question,
and upon the general chararter of the infilling Where any shadow of
doubt exists, it must be stated in the report.
104
r
IX
Watch-makers’ Jobs
HAT be a watch-maker*s ]oh\ remarked the andcnt, after long
and meditative observation of an archaeologist on all fours, plying
JL a pen-knife and a water-colour brush upon the reluaant soil.
And watch-makers’ jobs are not the least amongst the employments of
the director and his supervisors. *The archaeologist, who the moment be¬
fore may have been writing notes in a different pan of the held, has to
take his turn with the pick, or more probably with a kmfc, and may
spend the next few hours crouched in the same hole, in the same uncom¬
fortable attitude, engrossed in the cleaningj recording and safe removal
of some one particularly fragile object*” I do not altogether applaud the
dispersal of effort and the shortage of skilled assistance implied in that
quotation; save in the direst emergency, the general does not have to lay
his guns or repair a breech-block. But the principle is right enough; the
extraction of delicate objects from the earth demands the highest skiU,
patience, and knowledge available, and is not lightly delegated.
First, however, a word or two of warning may be uttered. Avoid any
semblance of excitement when an objea of some special distinction first
begins to emerge. I have seen a director of excavations leap excitedly
into a trench on such an occasion, commtmicating a false and emotio^
atmosphere to the incident and interfering therefore with cool, objective
workmanship. It is essential to check any son of exdtement instantly,
and to insist firmly on quiet routine.^ In particular, let the adjacent gang
get on with its allotted task v^dthout intermission. Nothing unusual has
happened. Discipline is now more than ever necessary if proper \ alucs
are to be makitained. And secondly, the removal of some special objea
is very rarely an adequate justification for an ugly untidy piece of exrava-
' Sir Leonard Woolley, Digging up the Past (Pelican Books, 1940), p. 40.
* More charmingly expressed by a French writer on excavadon, lijaut laister
Us vtstiges ett pltttt aassi loitgientps fue possibU, ne jamais se pressw « amserver
SOD calme. Le moment le plus dangereux est cclui d’une vraie gnmde ddcouyerte;
I’exaltation risque dc devenir tcUe que Ic fouiUeur oubUe la moitii des consignes,
poussc desexdamanons admiratives ct s’aper5oit trop raid que sa documwtation
cst incomplete. II est vraiment grand s*il a le courage dc s asscoir et d burner
unc cigarette pour riflidur.’ A. Leroi-Gourhan, La Feuilla prihistoriquts
{uchiûe (t mithotUs) (Paris, 195®)> P- 7*
watch-makers’ jobs
tion. The rules of stratification must be adhered to; if possible, a part of
the object should first be uncovered in such fashion that the remainder is
still embedded in, and projeas from, a clean-cut local section. Apart
from the importance of its relationship with environment, this section
may reveal unexpeaed information, such as the stain of a vanished
casing which on plan had evaded recognition. In other words, ordinary
rules may not be jettisoned without very exceptional cause. The warning
may seem unnecessary, but experience shows otherwise.
Having said that, I turn fix)m precept to example. And first I propose
to turn to the ship-burial of Sutton Hoo (p. 82), where Mr. W. F.
Grimes has described the ‘exciting and exacting task* of removing the
treasure firom the burial-chamber.* The following extensive extracts are
reproduced by permission.
The bronze bowls... were standing one inside the other, with a number
of iron objects in association. Three angons [long iron javelins] were
actually pushed through the drop handle of the larger bowl, with spear¬
heads and other implements and weapons near at hand. . . . Corrosion
was so far advanced that it was unlikely that any free metal could survive
in any of them. . . . The weapons were corroded not only to one another
but also to the side of the bronze bowl with which they were now in con¬
tact. ... It was abundantly clear that they could not be mechanically
separated on the spot. .. . They were therefore cleaned up with care and
freed from sand—a task which took some time because of the large amount
of undercutting and the generally restricted conditions—ând the whole
complex of bowls and iron objerts was lifted intact. . . .
Beneath [a silver dish] was an assortment of articles, most of them in
a fragile and parlous state, the recording, removal and packing of which
took the undivided attention of all working on the site. The most urgent
were the orgamc materials—especially a number of small cups which
were thought at one hasty glance to be made of wood. But there were also
leather and other materials, all of which owed their preservation to their
having been more or less encased in fabrics and a flock-like substance
which had kept them in a state of perpetual dampness whilst shutting out
air. Delay here would have been fatal. The cups already showed signs of
distortion and were damaged in other ways. And a hot drying sim poured
into the bottom of the ship. If they were to be preserved for future treat¬
ment it was obvious that quick measures were called for; equally obvious
that the best hope for their preservation would lie in a reconstruction of
the conditions which had already preserved them for so many hundreds
of years. The cups were therefore closely packed in damp moss in boxes
* Antiquity^ xhr (1940), 69 flf.
106
with well-fitting lids to exclude as much air as possible^ and stored in a

cool place away from the sun. The leather and fabrics were put temporarily
in bowls of water.
There is little to be said about the removal of the remaining objects
from beneath the dish. The small silver bowl was taken up with its con¬
tents untouched but protected with a thick layer of moss. . . . The fine
leather bag in which the small dish lay was our despair. It had become
ragged with decay, was cracked and fissured, with all its native toughness
gone. The remnants were cleaned and photographed in situ; after that
interest concentrated on the size of the largest piece that could be detached
unbroken from the underlying wood of the trough in which all these ob¬
jects had been deposited. . . . The trough and the objects surrounding it
were left until later; the wood was protected from the drying action of
sun and air by a thick covering of damp cloths.
In the western end of the burial-chamber, was revealed
an iron object which proved to be of quite unexpected size and character.

As with cleaning its various features became clear we called it a lamp-
stand. It was much corroded but appeared to be fairly strong, and in spite
of its size was lifted quite easily. The surrounding sand was completely
removed and the weight of the stand evenly supported by three people.
It was placed on a wide plank of a suitable length on which its unsuppor¬
ted parts were propped with packing covered with cotton wool.
Near the lamp-stand was an iron-boimd wooden bucket in an advanced
state of decay. The iron binding was badly corroded; the wood had be¬
come friable. The vessel had therefore collapsed upon itself and become
a more or less shapeless mass. Even so it seemed possible that careful
removal might enable some sort of reconstruction to be attempted later
and there was the further possibility that the bucket might contain other
material of interest.
Surplus sand having been removed, the bucket was bound firmly round
with webbing, particular care being taken with the lower part, upon
which the greatest strain was to be placed. A piece of thin iron plate of
slightly larger area than the bucket was then obtained and gradually
introduced beneath its base, the sand being cleared away in front of the
forward edge of the plate with a trowel. With the plate completely inserted
the bucket was ready for lifting. But the iron was too thin to sustain the
quite considerable weight of the bucket without buckling which would
of course have disturbed the already broken pieces. A flat spade was there¬
fore inserted beneath the plate, which prevented any lateral pressure
being inflicted on the bucket while providing a large surface for its support.
Bucket and plate were then lifted together on the spade and placed on
107
strong boards so that the spade could be withdrawn. In this way no part
of the bucket was seriously disturbed.
Mr. Grimes remarks parenthetically that he has ‘used an almost

similar method in removing very fragile Bronze Age pottery. The con¬
solidation of twô Bronze Age urns from Coity, Glamorgan, was only
possible because they were lifted intaa in this way and the crumbling
pottery hardened and treated in the workshop before its contents (which
helped to reinforce it) were removed*. I may add that in similar circum¬
stances, in the removal of a considerable prehistoric clay oven,
I have likewise used the iron-plate method with success. The prelim¬
inary binding of the fragile object—preferably with an ordinary surgical
bandage—is an important stage, and may sometimes have to be supple¬
mented by a surface-coating of plaster or melted paraffin-wax to give
additional solidity. The particularly fragile clay oven was coated with
a considerable jacket of wax, and survived the operation of removal per¬
fectly. It is now in the Dorchester (Dorset) Museum.
Enough has been quoted from the Sutton Hoo account to indicate the
general methods used. Mr. Grimes observes that the account
is not a story of new and elaborate technical methods, but rather of the
way in which simple readily available means were employed to deal with
an unexpected range of materials, each . . . with its own set of problems.
In some cases one could visualize an instrument or method by which the
solutions would have been the more easily obtained. I longed several
times for a broad-bladed implement, a kind of fish-slice, which (in various
sizes) would have been ideal for lifting such things as the horn mounts,
the silver bowls, and even the wooden bucket. But only a specially made
tool could have combined all the necessary qualities: strength to sustain
weight and pressure, thinness and a sharp edge of a suitable outline for
easy insertion beneath the object to be removed.
The absence of some luxuries had to be countered by patience and
perseverance. Our tools were of the simplest kind. Here we were fortunate
in our soil: the sand yielded readily to treatment, and especially, when
dry, to brushing for the more delicate objects. For these the usual proce¬
dure was a sequence of alternate phases of brushing and drying, the sur¬
face sand being removed to expose a new damp sand which in its turn
was allowed to dry. Paint brushes were most useful for this purpose. In
actually carving sand away from heavy objects calling for no special
treatment of the surface—and also for much of the finer work—I was very
gratefril for a curved bodkin or packer*s needle, especially as long as its
point remained sharp. The curve of the needle was particularly useful for
io8
negotiating hoUows, angles, and tindercutting; it provided a

‘feeler’ for unexpected shapes; and since it was not mounted in a handle
it could be used in a restricted space.
Of the general practice of cleaning it remains to be said that each
specimen, whatever its material and character, was freed as completely
as possible from its matrix: the importance of the first step côt be too
strongly emphasized. (Even sand when damp is sttong^y prehensile, an
the extra strain imposed by its adhesion even to a smaU ar« might weU
result in damage to a firagUe specimen.) Care in co^g the aiwy
without damaging the surface and in lifting the object so that aU Ac
weight was equaUy distributed, wiA no strain on any one part, completed
Ae process. Such precautions may appear to be obvious, but were espea-
Sy taportant on a site where aU except Ae most precious metal was
badly decayed and corroded.
In Ae circumstances, dealing as we were wiA a wide range
and materials all calling for laboratory ueatment, and not kno^g wtot
else we might have to face, it was sound poUcy to remove ev^ Ae most
Other meAods have now replaced more direct mechanical treatment of

antiquities; here mechanical treatment was reduced to Ae minimim
neewsary to remove Ae object from Ac ground. In ^
muchôken and apparendy worAless objects.
were expended m attempts to maintain Ae various parts and f^g^ts
m Aeir relauvc positions, as a very valuable first step m rccons^ g
Ae original or obtaining maximum information from it by
laboraS examination at a Uter stage. This was not always success^ ^d
was made to see that no scrap of evidence from Sutton Hoo was lost Dy
lack of care and patience m Ae field.
By way of comment on Ais account of a Afficult and ^“cc^^
of work, I would reinforce Mr. Grimes’s plea for a broad-bladed im¬
plement’. Independent experience has shown me “mtant necess*^
for a selection of implements of Ais kind. They can be
wiAout undue difficulty and should be a routme *
equipment, beginning wiA Ae broad-bladed kitten lê ^luch ev^
s^rvisor shoSd carry (p. I54) and remold
na^cd flat instrument (‘turner’?) with which fried eggs are removed
from Ae adhesive surface of Ac frying-pan. .
From Sutton Hoo we may travel to Ur of the Chalde«, wh^ S^^r
Leonard WooUcy had much experience of the .
has something to teU of it. One of his masterpieces was Ae recovery of Ae
109
WATCU-AlAKJ-liS' JOBS
haip turn Sliub-ad's tomb, and he shaU teU the #i(ny in Ms own
wends J
11k first thing ta come to w&s rht^ gold cap the uprighti which
«ciecd to be loose In the soil and gav'd ns no hini as to what Jyy below- As
the worlc veent un there were found two or three gold-headed ndli und,
Harehlng foe their pwsible connexior, we found s hole running down inm
me earth aerosit which could be seen the ahafts of more nails obviously
m ptuiijhn, he. the hole represeoicd some wiooden object which hitd de-
ajfed Dwwy altogether but the naiJs once (bred in it were beioir kepi in
place by the soil ogaMit their heads, A sdek wns therefore iiucrred into
hole for so far as it wotiM go und plaster of pam was poured in round
iii when dwt had hod lime to set ihe digging continued and there was
I^nd the Jotm part of the upright modelled in plaster with the gold-
hwded nails in ji^ measurenieni of die soil and the calculation of the
disance apart of die dislodged nails gave tJic fuU length of the upright and
enabled us to refia ihe gold cap at the original height. Below, the plaster
had «|«ndcd into the ‘shoe’ of the instrument and its How had tl.en been
Kopped by the nor alrogcLhcr decayed bitumen which Imd held die shoe
TO the base ; this, wiili the line of shell and lapis inky wMeh emphosued
Its curve, was at once hardened with poralSp wax. This broughi us to
the wunj^-b™ which, being of wood, had complctdy perished, but
he hr^d band flf musaic along its edges was for the most part in positioa,
though rather distorted, and could be cleared little by little und secured
M ir app^ by wiced muslin. Tiie wood had opparcntly been painted
ack îh a hM of red paint running parallel m the edge a Utrle inside ihe
nlay border. The top edge was first treated and then the side eoidd be
hud and tJic foim of n aseertained; the rectangle of the near side was
ete (It ^ indeed lifted in one piece) and the back of the inky of the
far «Jc «™ld alM be clcanM and secured. Tlje gold and lopis-kiidi calf 's
el I? “^trument Seemed to bo in re,her
bad contht,™, lor the whole of the top of the Jiead, cooslBing of kpi,
tesserae t^rBenciag Mur, Mid fidien down mio the hoUow left by the
dcoiy of wood™ core, and the merel wo* □ good den] b™r, but nothing
was nussing; m the end it was restored without much difficulty.
prewi^ and an acceptable reroimrnciicni of the whole icstnireent

renderrf possible. Tlic whole succcm of the operation depended on the
recogmtioii at the outset of the poienikl significance of a hole running
down mto the cinli. Similarly, ^
no' 74 r.* Hid Di£^ up iht

pp. (PeLoin SiKih, ‘"'I PhlkJetphk,
^ isis),
IiO
WArCH-MAKFHS" JOBS
of our ricfteni Rravca at Ur, tint whjcli contaiutd ihe tamoiu ffotdca
hclmcti w^s iocaEed hy ihe diacovery of a copper spear-hcfld ^ricfcuiH
point upwards in llie The mil dcarcdl i'ram roLind ih and ihcPt
emnfi to Hghr a Icnfith of thin gold tube which adorned the top of the
shafts below thii there waa a hale in the pound left by the wooden ahiJt
liaeli when it tuiuotl to dusr We followed the hole downwardii emd it led
ns 10 the pravc, asainat the comer of which ii had been Icamujs when the
corUi wm throwii hack inm the pit^ with this forewarning we were able to
trace the entire outline of The p^vc before we shifted to lay bare in con-
teiir$¥ and bo could record In order all the olfcriiiga heaped and crowded
roued die cofin.
Where the atnual woodworlt i$ pfKcrvcd, other mcihodi uiUii be
adopted in acrordance witli ckciiffwtanceah Thus In a La Tiae
$word ifi ins bronze-bound wooden iveabhard was found hi the ^'aicr-
logged clay at the boLLom of an ancient ditch at St£Qwi<^ in Yorlwhire.
tt wan cnrcfiilly cteared to a rough ouiline with trowel and penltiufe, and
sufhdent mud wsi^j then washed from it widi a small soft bru^h to enable
a phoengraph and an approsiirLaTe life-size drawing 10 b* made of it tm
the s^pnt. Moinwhile y local Ctiqwntcr vms preparing a b03C lo rcocivc it,
with an ample allowjuice for packing matcrULl, When the drawing wu&
enniplctctl, u siilT 3-p3y board was carefully ickserted beneath the object,
whi[^ was then ijnpiediaicSy placed without further cleaning in the boi
in a thick matrix of wet nempapcTp The package then taken by the
next train to the British Museum laboraiotyi cart hemg lalccn to nudn-
tflin the dampnew of the packing. The subsequent croiiinent, applied
with great skin by Dr. H. J. Pknderlcltli, lies oumde the scope of the
field-worker.
Lastly^ ir iu fair to iiidude iWsiOUgst 'waich-makm" iols" the removal
of a Roman tnosaic pavcmciit^ the oondmon of which usually demands
the utmost ddicacy of handling. The surface of the mosoie may have
sunk into hollows: many of the tesserae, even chough in position, may
be fractured • and the fine monar in w'hich the tesserae were originally
embedded may largely have losr its grip. The conventionaJ procedure
has been described in categorical form by the late Mn. T. V. Wheeler*
VSJ^.j who herself applied it sueccssfuliy/ Her accouni is hem tepn>-
duccd.
I. Dig a tremh i foot wide and iS uiclies deep round the area of the
pavemcni to be lifted some days before the work of remos-al is smned.
A considerable amount of raoistiirc In thereby drained iway.
1 rht Mumms Jitumal, E30C (Loaduo, IQ4
111
1
watch-makers' jobs
2. BwJtl a siinl or ihelier of suJTidcnt sisc to rover the pav'CRienr acd

the
3- Rrtnovc all dirt îcmpultiiuiy' fmm the fhavetnent and from the joinm
between the tcH^rac. iTie condition of tJn? pavement will diertaH the tneana
to tx ttoploved. Three may vary fmm Kembbinjt ^îih bmsh and
wiTtr aod/or stTBping with c blurif knifei to the ddicutc use of an
dnmife-ijtiek or bel^Ks.
4. Place coke-buckets ou bndu m the pavemenr and keep ftres buriung
day And nfebt untd the moBslc, jrs heddinj and the earth beneath are
absolutely dry. iTie heai not only drives ylf any remaining moisture but
uho dkiniegrares the Romun mortir. This drying-out is tJw jtio^t lEnpor-
mm pan of the whole process and may take fmm two to five days accord-
Lng to local condidons of dampne^s^
5, Before or during the drying of the pavemertt, prepare u flat boardj
a feu? Laches larger in each dlrecrinn than the fragmciit or secdon under
treatment A plaiform of j-inch boaird^ reinforced with ems-battensj is
adequaic.
fi* Remove the fires and clean the mosaic fiiuilly.
7» Brush a ooai of ordinary glucj prepared in a dixie nearby and sufE-
amtly thin to run Incv and fill up all Intcrsticcis^ over the surface Md
leave Lt to set.
S+ Brush a second coat of thick glue heavOy over this hardened surfi|jC^«
9. Press dot^fii into this hot coat a piece of stuut canvas newly wrung in
Ixrfling water. The canvas should be squec^d down wirh the fingers until
it is certain that all air -bubbles ore ejiminareti, A final coat of glue on top
of the canvas TvfU ensure adhesion to every tessera*
10. Replace the fira in the drainage-tfoiches for an hour to drive off
moisture.
11. Remove fitifci and. allow the glue to harden. The rime that this will
take will again depend on local circum-sloutCdij but upwards; of eight hours
should b< allowed. Agaln^ it is essential to nmlce sure that the glue is act
before attempting the work of removal.
12. ITm mosaic can now safely be released &ora its bedding. Cut into
the mortar of the bedding with trowcli^ slate-ripperB or^ if available, bar-
chisels^ leaving a good clcanuicc of a-3 hichw below the lower nirlace of
the tesseraoH
I j. S^b'p in planka undeiticath » the bedding is removed. These should
project beyond the odges of the paTîemienti for use in re\'crsing in
14. Lay the prepared board on tti the surface of the pavement.
Using the lower plank?i as levers, turn ihn pavementi now sand-
between wood» on lu the prepared boards which new bccomnii the
working-plarfqrm. Ttie pavement, or sectiud of pavementk can now ht re¬
moved to a work-ahed on this platform.
115
16. Remove all Roman mortar. If the heat from the fires has penetrated
properly the greater part will come away if gently tapped with a wooden
mallet. Care must be taken to scrape away any obstinate mortar from the
side and back of each tessera, and bellows used to blow fine dust from
between them. The tesserae are now in position face-downwards on the
glued canvas and are clean.
17. A slight sprinkling of water applied here will soften the glue suffici¬
ently, without releasing its grip, to press the now convex surfaces of the
former hollows into their original horizontal position.
18. Nail a two-inch frame round the working platform.
19. With soap or thick grease fill up all breaks in the pattern, so that
the modem cement will be kept back from the ultimate level of the face
of the finished pavement.
20. Mix a sufficient quantity of modem cement with broken Roman
material, imitative of the mortar used in the original pavement, to pour
into the joints between the now-isolated tesserae.
21. Tap the frame on every side, to ensure that the mixture is filling
all joints and that all air-bubbles are eliminated.
22. Lay a reinforcement of wire-netting and steel rods within the frame
and fill up with cement. If the section is of considerable size (and it is
possible to handle a section 6 feet by 3 feet by this method) a second layer
of reinforcement may possibly be needed. The cement should be brought
to a smooth surface within the frame and left for two days or more to set.
Where large pavements are to be removed, sections of a workable size
should be determined upon, and marked with tell-tales on the canvas,
which, initially, may cover the whole surface. The canvas may then be cut
along the lines laid out and each section lifted separately.
23. The frame can now be knocked away and the pavement reversed,
canvas-side up.
24. Flood boiling water on to the canvas until the glue is sufficiently
dissolved to allow the canvas to be lifted, not pulled, away. Any remaining
glue should be removed with boiling water and a nail-brush. If, in this
process, any tessera is found to be loose, it should at once be reset.
25. The breaks in the pavement may be treated in accordance with the
policy desired. If restoration is envisaged, the newly-cut tesserae will
declare themselves as modem and so prevent deception. If restoration is
not desired, the breaks should be brought to a smooth surface at the
general level of the mortar-bedding.
The pavement should now be strong enough to withstand any reason¬
able handling and ready for exhibition.
6670
II3 I
X
Tactics and Strategy

T hk main inbiect ef thix chaprer is the overall plynnin^ cf beM-
nswarch, a phraic duguiKcd above under iJjc sonicwhai jjaJiatod
(crm ^strategy’. In icimspeo every arctuiculugii^r is constaudy
arranging tiis tocu iuiu a pancm, his iabodausJy cnllccied words into
a onheient seatence. Less ftcquectly does he place his inquiries In ad-
vance wiih an tûivalcnt regard for sequeacc. There is some excuse in
rhe fact that at the present lime held-arcJiaculogy is still domuiated in
Kricain by special conditions arising out of a hard «'oranda hardex peatT
11 is condidonedf loo oftaij by the incidence of a bomb-hole ora houiiing
scheme or an auodramc, or by sh«r economic duresse, rather than
by long-ietra plmming. Yet even between wars, strategic pkiming,
though (Ometimo tallscd about, was rarely carried into effca. Othtr
fanors—personal or international oompetirion, sheer lack of imagina-
rion—obstructed it. Sites were dug bireause they ‘looked good* or
because they JwwAt produce infonaation, rather like canying out a sur¬
gical operatiou at random ou a palitni in the hope of finding somewhere
the cause of on undiagnosed disease. It was thus that the neolithic sijr-
getm used to cut a hole in a man's skull to the hope of letting out a head¬
ache; but not tints, 1 need hardly say, is the orderly way of sdcncc. As
arehaeologisu we arc excessively prenc to opportunism: we do not
sufficiently plan or create out opportunido!. Admittedly a happy chance
has from time to tiinc added unmcpoctcdly and even dramatically to
knowJed^; for instance the finding of that famous seventh-century
boai-buria] in ilie mound at Sutton Hoo, Nevertheless, the pmgress of
sdcncc depends less on these luucaids than on the methodicat, logical
use nf the disciplined imagination in the evaluation of eausc and etfcci.
It depends io no small measure upon careful Strategic planning.
The nearest oppreach to plan^ during the past hslf-centuiy or
more ix probably to tx reoogmited in the dlscovety and entargemenr of
the Atiooan QvilLaiion of Crete. The vivid quality of the avilLzarioii
itself, fh* imaginative genius of iis prime discoverer, and its ccfttral
geographical position have combined to give g certsim unity to the multl-
fariniM bur more or lc» related work which has been cairied out by
several nations in the adjacent hinds. But even there co-ordination Ws
114
TACTICS AND STRATEGY
been notoriously inadequate and the results proportionately scrappy and

unsatisfying. A great opportunity has been missed. In the circumstances,
as illustrations of what I mean on the one hand by tactical and on the
other by strategic planning, I may be forgiven the presumption of taking
two examples from work in which I happen myself to have been con¬
cerned, where it is at least possible to speak at first hand of the conscious
and careful sequence which was present continuously from the outset in
the mind of the direaor. The first of these two examples is an example of
tactical planning in the development of a local problem. That problem
was the unravelling of the history of Verulamium in Hertfordshire as a
Roman and pre-Roman dty, with its potential bearing upon the Caesa¬
rian campaign which culminated hereabouts in 54 B.c.
I may recall that Verulamium, in the valley below the medieval and
modem town of St. Albans, is a site of some exceptional interest in the
protohistory of these islands. There or thereabouts in pre-Roman times
reigned, as his coins tell us. King Tasciovanus, the father of Cunobelin;
and since CassivÛaunus was a predecessor on the same tribal throne, it
was assumed/4hat there also Julius Caesar had found and stormed the
headquartets of his great opponent in 54 B.c. Later events included the
martyrdom of Albanus, a Glmstian citizen of Verulamium and our only
authentic Romano-British m^x^, and a visitation by St. Germanus of
Auxerre who thus provided ode of the few unquestioned fifth-century
contacts between Britain and the Roman world. Altogether, the site
was fraught with archaeological possibilities of an alluringly miscel¬
laneous kind. The historical and geographical sequence of human
occupation in the St^ Albans region through the five centuries presented
a challenging probl^.
The visual starting-point was the large walled and embanked area,
some 200 acres in extent, which manifesdy represented the developed
Roman municipality. The conventional theory before excavation was
that this area represented also the pre-Roman capital and one of the
first functions of the excavators was to check this view. Accordingly,
deep pits were sunk to the natural surface within the walled area, and at
the same time sections were cut through the defences. The former failed
to reveal any pre-Roman occupation; the laner showed that the defences
were not earlier than the second century a.d.
That may be regarded as Sugc I of the investigation. Stage II was
direaed inevitably to the examination of the next most evident struc¬
tural relic in the vicinity; an unexplained earth-work known as *The
115
TACTICS AN1> STRATEGY
which projfxitefJ on the hiJl-$idc from under a part of the circiiii

of the sccond-ceiiLiiry defence?, in md under die nunparr, was
found a considenible quannty of native pottery, preûnuibly indiciring
the proxkility of the pre-RoiTLin site; bm an adoilxturc of c:arly Ronm
imtciiiU indiciiicd a pasr-cnnc|iiC5C dacen Siagt II^ then, had idcndlied
$iieof an earlier Roman lawn^ but left die pre-Rciman sctrlcmciit still
in the air.
The early portery recovered from *Tlic Fosisc* served at least os $
pointer* The eauh-worlc Eay jmt below die brow of the plateau tian^Tn^
the vaUey whieh carried the main Romm site, and the obvious next step,
Stage III, was to pursue the inquiry on to the plateau itself. Here, in
the deptlis of a tangled game-preservêj hes a mmplcx group of unmapped
and unimpressive carfh-iAt3rk$. Some of these u^rc disused field- or
'WDodJacd-boiindaries^ but amongst them was a Ducleos which was
proved by excavarion to be a portion of the pre-Romem city* A part of
otif quesLion was thufl answ'etedj the associated remains, however, wêre
of posE-Ciesajiim date, and CVidcnee was still lacSdug for the Qiesanm
epbodCi CFijs- ao^)
There wus sdR an unexplained feyture which appeared to rcLite to the
pre-Roman site. A mile to the uurtb of it were trace? of a crossôunEry
dyhe or battier fadng aw-ay &om Vcnilamium. Excavation, whith may
be d«$iJkil BS Stage IV, sbowTStl that («) this dyke wus likewise of
pPC-Ronwo dflic, imd it luid oiigutBlly barred iut open gup of chslk*
land betn-ccn ibc hiU-top woodland on the one hand and the maishy
vallty 00 the other. It was presuioBbly a boundaiyof the ptc-Roaum civic
temioty at a poini where nattual obsladea were inl5$iog. Iis aujor oilier
io out task of eaplonrioo was, however, as a pointer to draw fresh
anention to a secndagly related but mysrerious doss-cuuntry dyke of
&r more fonnitkblc site m the opposite dde of the valley.
This fiiftber dyke, known to faiae as Beech Bottom, is a. remarkable
consmictioCL, It is loo f«t wide and upwards of 30 feet Jeep, with a
bank on both lips. It is sited aJoij); ihc bonom of an cast-wcsi valJi^, and
there is only the peacer size of the niore southerly bank lu
show that the iiced oorth. It vraa dearly 0 maior tiafiic-baiTicf
niiher than a nubuiy woric in the nurmwer sense. About a mile of the
dyke is ^ tisible, hut a ftinber half-mile has been tiacni by cacavatioo
or safe inference. Its Dcaminaiion had already been as Stage V
when a happy chana: supphed mon: dduiiw evidcocs than unaided
sctence could have expected. Wotkmim digging a sewer at measured
I Id
TACT res akh stratecv
deprhK ihimagh the dyke, a poini where k had been filled up, found
a coin-ho3Jid of the second eeniury A.tip deep in the filling hot still a doicn
feet above the battom of the ditch. The coins—^ shovdfijl of than, it
Fig. 20. Suctenivc silts of pre-Ronran and Rn<mAn Vcnilimiiim

H ibtr OrjiHfliit ûTTrT cuPh tlni*a XXf^% br pfrniîiflp Ow
GmCrtlkt af iCm. :Sud4MTr Ofi« 1
was said—were immcdktcly sciittertd aoudsi; the ready hinds md

pockciB of [he cosmopcIitaD gang ofwarkinciiCODCetued^ Bmistrtnuoiis
search that night in a niiiltitudc of local pliOts of refreshmcni and cnitr-
fainmrnt produced aboui forty of them, enough to Lndicitc the gtnml
117
character and date of the hoard. It was clear enough that the ditch in
which they had been found was of some considerable age by the second
century a.d., and that solid faa, combined with its entirely un-Roman
character and siting, placed it fairly and squarely in our pre-Roman period.
But this was not all. If the dyke had served any logical purpose at all,
it must have barred a fairly open stretch of country between two river
valleys and fords: that of the Ver, beside which Verulamium stands to
the south-west, and that of the Lea to the north-east (PI. XV). Along this
line the eye is carried to a height above the ford across the Lea, where
are the remains of the greatest oppidum in this part of England, that
which adjoins the little country town of Wheathampstead. The present
fame of this oppidum may be measured by its appearance in an historical
novel; but it was not in fact until Stage VI of our planned advance that
the significance of the mighty earth-work became tolerably clear; it had
previously attraaed little or no attention. The profile of its defences is of
the same kind as that of Beech Bottom, only more so. The enclosure is
some 90 acres in extent, on a platform above the Lea, just as pre-Roman
Verulamium stands on a plateau above the Ver. A significant difference
between the two sites began to emerge, however, as excavation pro¬
ceeded. Whereas the Verulamium site began in post-Caesarian times,
when, in the age of Tasdovanus, Roman things and forms were already
beginning to penetrate ±e backwoods of yet unconquered Britain, there
was no hint of Romanization in the equipment of the Wheathampstead
oppidum. In other words, this oppidum preceded the last two decades of
the first cennuy B.C.; and, that being so, it was an easy further step to
ascribe its great defences to a phase of inter-tribal or even international
rivalry such as marked the age of Caesar’s British campaigns. Here, if
anywhere, may we locate the headquarters of such a one as Cassivel-
launus. At least we are approaching dosely to his time and environ¬
ment, and no rival site is known. The name of Wheathampstead has now,
with no great questioning, entered the books as the scene of Caesar’s cul¬
minating British viaory.
Thus in six progressive stages, developing gradually and logically
from known to unknown, the \TSta of a formative phase of protohistoric
Britain began to unroll itself. In reverse, the picture is that of the initial
overlordship of a powerful Belgic king, competitively exploiting a
countryside which had previously, it seems, been but thinly inhabited.
Thereafter, during the Augustan Peace on the Continent and the centrali¬
zation of native rule in south-eastern Britain, the emphasis changes:
118
tactics ASU STHATKÎV
ffjniGcatioi:^^ bereabouis ar aiiy me,* arc ntiw of cnvinl idaCj and con-
liucnULJ trade petcohm with increasing freedom. Tlic phase coimmates
in the formal metension of the P&x Rijmann to lowbnil Britain inA.D. 4li
thereafter fottiiicndon Vi^s disoounred by over-eoiiEdcaoCp whkh wa$
in turn couriered by the Boudiccan revolt and by the coosequent coa*
struction of new defences of no mean order ("The Foaelj and finully^
In the $padous dap of the second ixfitiuy] the town took shape on
evolved, conveniionat continentaJ lines. We arc not here conoenied wiiti
the deinil$, only with rhe sequence of thought and action—nn elenien-
taiy pustr^tiofi of archaeological tactic on Jialf a doocen utiles of land¬
scape.
From the Herrfnrdshirc pamh-pumpr I turn to the wide horiMns of
Asia, 10 planning on a Large scale lo which the tcrai ^strategy" may pro¬
perly be applied* Again, the exarnplc is from my owti experience, on the
reoew'ed plea that hrit'^hand witness has on actrtalii^? which may sulfi-
ciently asumcrhalancc defects. As Director-General of Archaeotogy In
India from 1944 to 194S, I had, in spile of war and acute pnlincai dilii-
cuftteii an npporrmiity such as few archacologiSK have had of strategic
planning vriihiji the far-fltmg boundaries of a whole sub-continent. The
nonhem and southern parts of that sub-condnent differ alike in the
character of their respective probletns tmd in the amonnr of w'ork pre¬
viously done upon them. CcMial aud south India, wHrh whieh I am here
concerned, had scarcely been worked at aU. In the whole of that great
southern regioiti some holf a million square miles in cxicitt, tJjfiJC was
in 1944 no linn ardioeological darum-linc prior to the Middle Ag«.
In the north of India, contacB with prehistoric Mesopotamia and, later,
with Persia and the lands of the classical West had provided reasonably
fixed pegs upon which to hang some part of the prehistory Of the fron-
ticT-hmtia. In die south^ most of rheac pegs were absetiL Sporadic ex-
plomiiun of an unmethodical kind had there revealed great quantities of
material; um-flelds had been pihcrcd up like ricje-crops, megaliihic
tombs had been ransacked (somerinics, aa Erecntly as 1942, with ihc aid
of dyujunitc—p, 93), lown-siis had on Dccasion been broken into and
miscelLmeous debris recoverod from them; but now-herc had this
archaeological loot been inieirelaicd E^'EiematicalJy with itself or corre-
kted wiili any cxtemal datum- Of the rmiscun^ of the region 11 wnuld
hive been suHicient to quote the words applied by Petrie to niDetecofh-
ccotury museuim in general: 'Our museums art lastly characl-
* At Qjlcheitcr thrir imijainiM! by dyzuitic wealth and ■mbiiwo.
119
tactics and strategy
hinises of murdcriKl cvidtiice; ihz diy boaes of objeos arc (Jicre. baie
of aU (be facts of grouping, lotaJity and dntiag which would iilve ihem
hfetorical life and value/
fn ibis world of chnos (here was, however^ one jwKntial fixed point.
Since 177J, Roman coins, generally in hoards of gold or silver, have
turned up from tune to time, mnstly in south India where they rroresent
the Tmpenal iuxui^- trade abundauTly vouched for in classical and Indian
hfcramra. Hero was a cbaiMS of securing rcipiired datum. Could w
bur hod a siguifirant assodadon of Rontm coinage or other datable
Roman unpoits wiih a native Indian culture, we should have ar once a
firm base froni which to advance upon the general problem of the
ancient chronology of the Indian peninsula.
Accordli^ly it was one of my first acts as Director-General in roiA m
draw up o list of si|„ to have produced Roman coinage and to
simd vm members ot my slaflf nn a i,ooo-inile toin of the listed sites
with a view to scleenngoueof them for iovestigadoii. My envoj-s worked
Steadily down the west coast of India and round Cape Comorin without
ciûnrenng any veiy* hopeful sign. But the solution was in fact im-
miDcnr troeq another sottrce.
In July 19441 visited the Madras Museum and found in a cupboard
tticie a pan of a Graeco-Roman amphora of a type fomitiiir in the Medi-
temiDcan region about the beginning of the Christian era. It hod, I was
formed been dug up recently at 5 coastal site near Ptmdtcheny, die
capj^ of French India, some Eo miles south of Madras. 1 shall not
hghdy foiget CtD me) dramatic tnomeut when chance put into my
f!? w "I ””"" f^ a InJim siimoier an answer which
I Mdd aWt beheyc th« ftte had ant me 6^ miles from my DitJe
Rom™ Britamra discover: an answer which I had indeed sought but
h^ s(^y dared 10 expect. By arrangement with the French authori-
™iw ™ r c""”® PoadicIiMiy, and saw in the public library
there a collection of ub^ which had been recovered during the pre^
two or three yearn by French anilquarics at a sire known locally os
^medu. 1 milB »uth of the row. The collection comprised a mo-
™^ble qtnmt^. of Imhan i^tiny, beads and other obiecoi. togethM
Sin* ^soublage of material of Mediierraaean origin, ia-
k™ amphorae, fragments of glasswuic. pTof a
^ rcpnscnibg Cupitl wirh an
^ S'” '^"S «II«oly =, tad of
Aodiwiu but 00 lo,,,cr availiUi;. ;Uor ioponim ihu oil droc,
130
fLATE XV
51 ALBANS AND WHEATHAMPSTB^VD HERTS.

WITH FORMtm WOODLANDSi^STORFD ON CEOLOGiCAL BASfS
PREi NiirQglC WORK3. BLUL- : tiO^MAN WQPK5, RED.
OPEN
CUAV LAND CHALK lAHO.
LKymYlJiiOJOED
ALLUVIUM
CJWVEU irre ;
Mip gf ^ onhwcrla of WhcuhAmpufid jmd VeruLuAiucai with mtmxidtif

{Sup. NfJ
TACTICS A{S2|) STKATEGY
howcvc^J^were several ^hEmdfiof ned-gtotd woft of Uieldnd nmdc at Am:-

lium and other Italian cctiires prior TO A.p. 45. The m**^rifng of Lht&e
things had not been suNpecred by rheir finders^ but to m eye tJ^cd in
rhe West it wna insiantly and easily apparent. Dated pottery^ by vutue
of its rdatwcly li mited dumhiliiy^ was better evidence even tb^n dated
eoins» -with their 1™ cnmpurable anrvival-valtie. It temamed to ascer¬
tain, by Qiie^l digging^, ihe predse reladonship between this ArTttme
ware and the Indian culnirn or culturK fonnd on the same site.
Our subi^quent (1045) asca\'Biions have been ftiHy recorded. It will
suffice to noto hrcfc that the relationship was readily established^ and
that the site will dietcfoirc jô down in the ^tnty of south Indian ^rdmeo-
logy as that from which tlie classification of andem souih Indium cui-
tutes effectively bcgaiL With die aid of ihe dated Ltnpotts lefencd to, it
dehned for the hrst time the chronoLogtcal position of an extensive com¬
plete of south Indian pottery and other equipment datiuj^ from the Orst
Vivty Centuries a.d. y and It was not long after the conclusion of the work
that the importance of Artkamedu was found to extend far beyond the
vidnity of the site itself. This extension wb$ the result of carefully
pianDcd^ ptogressive research^ proceeding steadUy from known to part-
known to unknown^ and Its main stages may be recounted In the present
contect^
The first step was to discover the distribution of the nwre distinccLve
of the Indian warts thus newly dated. A tour of south Indian museums
met with immediate success- Amongst the distinctive produas of Arika-
roedu was a type of dish (FL XM) dceofated on the internal base wîh
coaccdtric rings of a rouletted pattern otherwise foreign to Indian cera¬
mic but familiar on certain classical wares including Arreiine ware itself
Some of these roujetted dishes were doubtless iiriparrsftani the Mediter¬
ranean,, others were local Indian imitations, A search of museums showrd
that dishes of this kind bad been found in soutficm India at a number of
places seprsuated by some hundreds of miles from one anodio- (Fig, 11).
The famous Aniaravacr, for exampte, which mpplictl the BucMhistscdp-
Tures foTtnerly displayed on the main stairoac of the Briiish Miisimm,
had produced sherds of them;! now in the Aiadi^ MuêuitL At other
places, such as jMaski and Kondiipur In Hyderabad State and Qiandra-
valli and fimhmagjri in the nortliern end of Mysore State;, tins dis¬
tinctive decoration, type and fabric had been fouhd bt ^-published and
ill-conducTcd cxca^-atiuns. Moski^ KondapuT^ and Chandrnv^li had
been. In parr ar least, of the central Indian Andhra Umpire which
TACTICS AKt» STRATEGY
Boumhcd at tJiE end of R.c ând the beginning of and ar Chandfi-

vaiii sitisy denarii of Augustus and Tiberim had abn been found. The
histoiifsl Etnd archafiological environment wa^ thei:efore conslsieai îth
DisrmmoN of
ROULETrEDVftR£„.*
wmtD ANDHRA Wy?E.o
SfTESWm aOTH,'...®
CAti30E»f OTTSS o)
FlCr 21, Map of tndii showing; ^hicU have produced 'roulettcd

ware’ of the first cenrury a,d.
tlic Aiikamedii (lutmg. At all ihest sites the newly datKl nmlened ware
introduced at once an element of chronolo^cal precision;; and on oU
of tbem it WBS iujinii«tly associated willi an elaborate and disttnccive
TACTICS AKD STRATEGV^
ccraniic industry which difTcrcd Jargrly from rhoi ai ArlkamMu btiL

cQiiJd nowr pic&uinublyj be regarded 2a cnn^rrnpoi:aj[y. Aln^dy die
fixcd-polnc of AnKamedu had indifsted the poîbiliiy of dadog^ by
memsi cf a litdt iâicfiiJ exet^'^den, a widespread and impoimnt TTidian
cuhiirc mott ihan 300 tsilca a'v^'ay.
A IQ one of th&c siie^, Brahma^H, in 1^5 indicated oiher and
wider possibilities. Trial excav'atiDQs of a somewhat unorrhodox kmd;^
assisted on occasion by higji explosive^ hud here been ennied out by the
State Archaeological Deparlment and had re^^calcd remains of an cxeen-
sive iindeiic Township. Bin the autstandingly intere^dng feature of the
area was this^ that adjoining the tow'n-site n large cemetery of
tnegaiithic tombs of a kind widespread in peninsular India but iies^cr
hitbetto adequately dated. Further, the Brahmagiri cists were marked by
the dJDCular entrance-opening or "port-hole" which is present on many
other Indiim examples and on comparable tombs in western Asia,
northern Africa^ and Europe (including Britain] and may be a slgnihcanc
common-factor. Indeed, these Indian tombs greet the eye of ibe visiting
European prehistoiian with n friendly familiarity and at the same time
with a chaUenge^ their date and origia at once become matters of more
than local impormnee^ Their general character has indeed been familiar
since the work (referred toonp. 8) by Meadows 'faylor in llydeTabad
Stale in the middle of the nineteenth ceGtnry> and James Ferguswn
drew renewed attention to Lhem in his in 1S72.
The ctiiiostty of others^ tooj has not been lacking* and in the aggregate
some hundreds of these dats in ccntrgJ or southern India must have been
despoiled at one time or another for the beneht of public or private
museums, from Oxford to Madras. The result has been merely to pile
up the mountain of our ignorance. Until the undated conients of these
lOmbs could be related to a fixed chronological point, their fiinhcr spolia¬
tion was -waste and worse than waste. But here at last, ac Brabmagiti,
was a dear opportunity for the csiablishment of sneh a KlatioTBhip ; for
the equation of the cultun; represented by the tombs with esidcnce
from an adjacent town-site a>ntainuig 3 known fector—the dated
loulettcd ware of which sherds had abeady been uneanhed them, ft re¬
mained 10 ascertain ihe prmse nature of this equation by the normal
methods of sdentific cxoiviitictL
Accordingly, as has already been recalled in Chapter IV^ In 1947
simultancoiis exploration was earned Out on the BmhmagLri Eown-dte
and in the adjacent megallihie cctncteiy. The former revelled three
1^3
TACTICS A^^D STRATEGY
BUCTcwiv'c and disdnctivc culwr^s, &rwbidi tlic uppermost ii:jsod-
aiGd fmm tbe ouEsei witt our rodeited ViTire of die first ceiuuiy ajj.
This fopmosT oiJnire ifl now identified as that of tlie ^ndhm phase to
which Tcfercnce has already twn made. Below it, and partially uvizr-
Ijpping it, was a cwJcuie identicaj with that which was being revealed at
the same Time in iJic ndghbonrmg cists. The overlap bctwiioi
the twu^ assessed in the oianner described on p, 48^ was uinpLy aufli-
dent to Lndicarc rhar the megaiithie culture* with its dabomte ijan
ctiuipment, lasted imo the first Century A.D, Thus fur tht first time a
fixed ehmnolDgical point was obenined for a group of tELe^thtc toinba
of duuractcnstic south Indian lype.
But this was not alL A reasonablecorpputation of the limc-viliu: of the
megiilidijc strata—a computauon which need not be revievied here—
suggested a backward diinidon of some two to two-and-a-half ccniiirLezi
from the texminal date in the first century A,D. j Lhat is, the culture had
arrived somewhere about zoo b,c. Below that was an S-fuoi acLniimila-
doji of occupation mnEeml Teprescadng anallOfiCtliet dUTcrcntaEidiiiurc
primitive culLure, Whereas ihc Andhra pottery liod been tirmcd an the
fftst wheel and the m^lirhie pniiery (it seems) on the duw wheel, the
crude wares of the lowest culnire had been hand-juatlc, w ithout any
sott of whecL Agaic^ whereas the overlying cultures were both in an
advanced stage of tlic Iron Agc^ the lowest wii$ essentially in the Stone
Age (polished stone xxq, mugh miemliths)*. although bronze was not
altogether unknown—the first evidence* Incidemally, for anvthing
approadiing a Bronze Age in Mmth India. Yet* in spite of the disparity
of the two cultures^ there wna dear e^'idtuce that this exude chalcolithic
community had outlasted the arrii^ cf the megidithie intruders; in
other w^rds* that the ehalcDlithie ailmfe* eletneals uf whidi had long
been recpgnijed* without contexr^ in many pons of south ludia, had
lasted iuio the second Lxnuiry ax.* and erteuded backwujds from that
date through S feet of deposi t.
It would be irrelevant here to eonsider the details and impheauuns of
this happy coitcurroice of evidence, although in fact it W'as to the wider
inrerpretatiOP of the cultuxc-movcmenTS ttpreêuted by it that our
work was dirtctetl SuSce ft to observe that we bid recovered a piotnrc
of n ntdimentaiy stoneûsing sodeiy os’emio suddenly (during the
Ttnac of Troubles following the break-up of the great AJKokan Empire)
by an elaborately equipped invasion of uoo-users and megaliih^btii tdernt,
followed in turn by the sophisticated cisiliKidon of the sutgent Andhra
124
PLATE XV J
RouteiLCii lA-Are of th& flfiE ccnrur^* a^. i^mn ArtkamciJu^ Sauth tntlii,
^Shp, iir*}
empire: ihrcc hitheirLo imdstîficd btiL widespread culiares nrvn- for the
first riinc arranged in dear isequence with a cbronologicaJ danint-linc
tit one end of ihc scqucno: and in ccDsequeoce fl new signifiaince lo
SubcoQimenral prorohiatory. Su unporLimi was it to establish thin datum-
line beyond all ahadmv nf doubt tlut a stnalJ parallel ejccas'arion waa in
fact cairied our simuitancoualy un unether ADdbrd tottn-site somE 45
iXliles avi^y. "This sire (ChandiîviJli) yielded parallel evidencE in the
associadoii cf our rrmlefred ware with local Xodbm coirw> overlying
■jnegalithic' fabrics^ and incidentally produced ftom one of the Andhra
Liycnt .1 denarius of Tiberius dated 2^37 (PL XIV a). It would bo
ddficuk m imagine a more satisfilctory conchision to- our ounpalgnt
JIT one end of onr story, our keyêvidence assodaxed with Roman ware
of the first half of the first century B,C. and ar the other end of our story
identically similar evidence associated with a Roman coin of the same
period.
'Phis story of the ôpening up’ Of central and soiuhcm [iidlan dhtono*
logy by gradual pcn''asiun from our starting-poini at ArduunedU may
here be carried one furthet stage. In front of 11s loomed the great prob¬
lem of linking up the newly found chronology with that of the northern
Indian plains* where lay the great Cities of Indian epic and protohistory^
I'he problem Thrcatctied dMiculty of a special kind. Tlie cultures of
these nonh Indian ddt^ have been fairly eitcnsively 'smnpled' in a very
rough-and-ready fashion^ and are found ro difrer in cv<3y respect from
t hose of the south- Corrdaiion was therefore only feasible OH a site where
south and xionh had actually met aritd had eaisied for rome tiffle side by
side.
Search revealed exactly Such a site* pateniuiUy, at a point where the
norrheni plains bend southwards along the broad COOStaJ strip between
the Bay of fiengu] and the Eastern Ohais of iiilimdbiU-bamcr. In Orissa*
ad}oming the old temple^ry of Bhubautiw’ar and, not far from the
sacred oisi-coost town of Ihirij home of die notorious Juggemaih* lay
the embanked site of an amaeni dty* now feaown as SiAupalgarh, which
fulfilled die necessaTy geographical conditions, l^xcavatbn in 1948 and
1949 showed that it also fulfilled the archaeological lequisiltt which we
had postulated: for In stradficd relationship with a distuictlve and wide¬
spread non hem cemmic lay sherds of our nD\i -fac]i30U$ foulettcd w arc^
identical with that whids only three yEui previously we had found nc^y
700 miles to the souih-we&t in associiirinn with first-eenrury Arrctine.
North and south were now for the first rime fiitnly interlocked-
I3S
TACTICS AND STHATEGY
I Deed BQi pursue this mancr further. Enough been sold Lu UJiiif-
tTftte how ilucc Ktsasons af OLEtiÛy planned, progressive work were
able to draw -j dear archaeoEogicat danun-liEie across many hundteds of
nulcs of a sutM.-ondnoiE^ and ro open up spadous new vistas ofcuiLural
rdatioRship. Gencraiinns of fijmbting ftojiiiite to opportimisL site might
easily have failed to approach rhis result—have indeed ia Lhe past failed
to do so, Tiic two ncêsjeiry fboors wêre prescnl^ Opportunity and
Planning, To these mighi be added the factor of Luck^ but 1 do oat
much bdieve in Ludt. Hard Thinking and sttAdy ciecutioa arc at least
its hon^t and dependable substimtes. and make the oppor¬
tunity, remembering with hope and at the same rime with humility die
dictum of Gibbon that ‘the winds and the wm.-es arc always on the side
of the ahtesT navigator’.
' ' • ^ .
Ijistly, a problem in which strategy merges dosely into tactics may
suitably be cnnsidercd in this chapter sin-K in fact maners of mafor
policy me involved. From ttme to time the quesrion arises: shall stress
be hud (in some particular programme of wnork) upon horizontal or
upon vcrtiQil excavation? By ^horizontal exca^-adon* is meant the un¬
covering of the whole or a large port of a s^pedhe phase in the otcupatiou
of an aiidmt site, in order to reveal fidly its layout and function (cf,
PL XTII), By ^'crtical excavation^ is meuni the accavarion of a restricted
area m d^pik^ with ^ view to aKeitaining the sucoKSion of cultures or
of phases and so producing a ttme-^lc or culture-scale for the site„
The two prucediuiDt ate of course complementary^ not antagonistiCj. and
the mccav'aior may be cipeaed to attempt, if rarely to achievej both
methods of approach. But in a great majority of instances^ a priority
has TO be determined, having regard to the stare of curreot knowledge
and the resources available.
Let us consider the nature of the evidence which the wo methods may
be expeewd to supply. Vertical eECOvation iilonc, whilst supplying a key
tn (he length Of an occiipatian;» to its coatinuJty or imemfutccacyj and to
Mmc pan of its culLiiral cquipmient:> cannot be expected co reveal save
in the most scrappy fashion the signihcant cmriranineiic—economiCi rc-
ligtouSf atiûllistT:ltiv<^—ofa human sodety. In other words, it leaver us
to tJic dark as to those very factors w^hieh fit a past culture or oviJixadon
into the story of human endeavour ujid so nuke its recoveij' wôrth while.
It is tlie taiJw^ time-table without a tnin. On the otlicr hand, the ex¬
tensive horizontal Exca^^dens wiiich were in effect the normal pracrioe
U6
bcfoic stnuifimLiDo wa:^ sdciqiiatdy undcniTond gcii£!rtii]y produced an
ab^LTuctiuD—uftcoi m vcr^' confused and mulcading abstT^ctjQii—iinre-
lulcd witli uny siuit uf picdsion to the sequence of huimn dcvcE-npirtuir.
They Exnlinr without a lime-table. The trdm aûmciimes mn îgor^
ou^ly enoughr but we knew nut when they were running or where they
stanedt iiiteti]iediJA[e btoppii^-plau::ij or dieir detitinution.
At cenain Plages of research both Lhe&e ineomplete methiHL may have
a ûbsutouve value»mdeeUj ±e)^ arc dieinsdves stages in die piOKiess
of research. 1 am CLOt> for ejcampk^ of dtose w lao $coni the hotLtontaJ esc-
cn^-ation (in the nineties^ of the Rotnan tow'n of Stlchestcr, True it was
dug like potatoes^ withom a shadow of the sdcnti^c nicety of the coo-
Tcmporary cKcavations in Cranhcime Chase; and the resuliam phn ts the
uncniical syiithcsif^ of a varying urban deiTlopmcni ihrough more than
three cennirics. But ic gave at nnee, and with a rough accuracy, ihe
general impression of a Romano-Brinsh town such as Efiry years of sub¬
sequent and oficn more ouicEil work have lailcd to cquaJ. More exact
Vertical and horizontal digging on both this and other similar sites has
indeed begun to ici^cai the soooiogicBl eiôludon c jsendal to our histori¬
cal pct^pectivc; but who amongst these later and excavators has
DUi constantly referred back w ith pioEt to the crude ptimltive assemb¬
lage of Silchestci?
So also elsewhere. The GJastoubuiy lake-villugC} excavated uncritic¬
ally with results that ore often Infiiiiadiigly boOliiig^ has nevertheless
given U5 the complete Layout of a smaLt Early Iron Age settloBent and so
enabled us lo assess in bnmd tertus the îoJ and economic signl&cauce
of such a seitlcmcm os no exacting and poitinl probing could have ren¬
dered possible. For chatj even in momentsi when the evidence in detail
completely fctils nsj we may be properly thankful. And kt ni for a
momeat look furtliei tifield. One of the most dramatic and revealing of
all excavated cities is prehistoric Mohenjo-daio> beside the Indus in
PaldstaiL TecluucoUy the methods adopted by a succeston of excavators
there became almost on miemaiional scandaj| and neither rrofess6r
Piggott nor I have beeo at pains to spate the losh^ But the primary
marvel of the great Indus city is not that it did (or did not) develop in
sueh-and-Jtuch a fa^thion heaweenr let us sayv 2500 and 1500 hut
ikai it fflxrzd of afl in the remarkable foim that cxtcnsivcr If dispropor-
lionaidy summary^ cxcayatioii has revealed to us. Its house-walls,,
towering aecumiiUtivety above our heads, its lung straight suttis, its
lanes, its elaborate drainage system, its dtadd—these imd other things
in bulk rc-crentiC o whole phii5C of imimn sodciy even ilinugh in detail

they fiiii to analyse it for us. Aiialysis—by venicil —
should^ of course, have accompimled all this siLunminy bon^nnt^ dcar-
aoce[ but there can be no question tlnit Mohcnjiiîiiin} takes its place as
the leprcsenmtivc of one of the great dvilizaiions of the andent ^'orld
in some measure by \irtue of the edmes of ii^ csqibrers.
And since we have arrived In PakisiMtr kt ini take again one more
familiar eimmple that landr For a thousaiid >xars the dty of Taxila
stood upon successive sites in the nonhem Puniabj and one of these
siiesj of the first ceoturies n.c,“A.n., ■was cxtcmivxly deared by Sir
Juhn Aiarshail so that a oonstdcrable porfiou of a remarkable rcaangu-
Jar lowTi-plan was ultimately revealed (FL XIII). Tl>e clcamnce did nut
confonn with modem technical srandntd^^ and in fact mote ThiiTt one;
phase is repreaentE^ without discrimination in the published plan-
Ncverlhcless, ihc outstanding interest of Pirthian Tajila is the genoal
chdracterof its building and their relation to Astreet^grid wiihouikiLUWti
analogy in this port of i\aia. 1 lad iht excaviator OOncenLmtcd on vertiod
digging on this deep site he would have given us valuable iafuimaiion
for which wc arc sdll w’aidiig;i hut he could scarcely have given ua also iIk
piaurc which wc owe to him of a teemiug dty with its sttetLi and
teiDplesj its palaces and its ahopa^ He would have given us a useful cata^
logue but DOtj as in fact he has, a vivid chapter Of sodal history.
Tlie four esmmpJot of horizontal digging which I have given—Sil-
Chester* Giustonkiry,. Moheri)o-daifO* and Taxila (Sirimp)—arc not
very^ happy iu tlut udiic of them was excavated with adequate shill Tech¬
nically, they eh bdong to the pre-Piti Rivers era* ihoiigh Pitt Rivers had
in fact ei^tablished his methods before any of them bcg^. Needless to
My^ it must not be mfenrd ihar horn;omal excavation is neccssanly
stminum' and unsdeniifie! Ideally, the excavation of a lowmîte would
begin with vertical diggiogi suffideni to establish the time- or t.-iiltuic-
sequence^ and would proceed to the careful horkoutal of stac-
cessitx phases* one at a time. Obviously the process cannot be tci-cised*
and at the three sites memioned only the careful vecticaJ excavation of
jrm not yet touched can pxrtinlly replace the squundcred evidence. A
better example of horixcnml excivation on a small scale is provided by
Little Wcwdbiii)' rear Salisbury> where Dr. Gerhard Bersu cleared ihe
greater pure of an Iron Age farm-stead and was ahk m reconstruct both
lU architectiiie and its economy^ The site was a shallow one, and the
technical problem wâs incomparably simplo^ than on a deeply accumu-
izS
lad VC town-site j ncvcrtlidRSi it is cbasteiilRig w reflect fsow liiik of the
mcaniiig of Little Woodbury could have b«n lecoveitd merdy by
verrical samples of it.
With the pioviso^ Lhen^ that ttli hortEoncal digging muir proceed from
clear and compreiieuslvc v'ertical MCtions^ die quesrion of priority is
fundamenlally not in doubt. Qtrefoi horiH>nial digging hJd;i£j in ihe
long run, give us the full Infonxiation that we idciJIy warn. Vertical dig¬
ging will, by itselL serve a v^nable purpose in esrablishing the geogni-
phicil distfibudon of a culture and its iinM>rdaooiiship with other
cultures from place to places but this evidence atill dîvc^ iis ultimate
significance from a tennwiedge of the sociil enviromnent of the culrurcs
concerned. Wise words arc said by the authors of the Council for Britiah
Archaeology*^ Suw^ and Pcli^ cf Fuld Rctcarch m if
(jramf Brttaifi (194B) in their statement witli rcfeiehce to the Early Iron
Age, that to have at least one siic in eadt main r^nn of ihe couniiy
excavated totally *will now be of more valtte than merely m annpie a
number of sites in the same time Instead*. But they are equally wise
when, on the following page, they also advise that hill-forts be sub-
mined to ‘selective exeavatjoiis to escerwin the dare, ehaiscicr, iind
sequence (if any) of the defences^ and whether or no the fort was per-
mancntly inhabited. This method »* ^ should be continued rcgiocyilJy/
Jr is all a matter of bias, and bias depends upon our knowledge at the
moment of the problem of the moment- There is no essential cooflict^
but the question of priority b worth caêûl rhought in ardiaeolcîcal
planning.
t>ncE more, the moral Is that the exerdse of tacrial skill without a
conmolling strategy b oppoitunlsjn mn riot. Or, to change the metaphor,
the cxcavamr without an inteUigeiit policy may be described as an
archacolo^dl fuod-gathercr^ master of a skill, perhaps, but not Creative
in the wider terms of constructive science.
fsn 129
XI
SrafT
T H£ tirnc has cofii£ to consider the stall and ntdcliioery with which
all the various operations described in the precedinft chaprerj ate
cjuried our. Fiisr, die persoand.
The staff of an archneoiogical excavation on any considerable scale in¬
cludes a dirceiDr) a deputy diiectorj a supervisor Ibr each area under
otcavaiion, a traiiicd foreman, a small-fiad reairder, a pottcry-asristHnr,
a photographer, a surveyor, a chemist, a draasnun, and, according to
need, an epigraphist or nuinisqiatist. In rare cases, cennin of these posts
may be combined, but it is preferable, and often necessary, to subdivide
them, Where distances are short, os in Great briitiin, It may be possible
to dispense with the chemist and the draftsinati in the field, but it is in¬
advisable to do so. Only an ignoram critie could protest dun the list is
excessive. KeverthclcBs, large and rdadvcly costly expeditions in the
past have failed of their duty thtough false economy or Isdc of prevision
in ih» all-ijoponant manej of staffl
To the oUida] and essendal nucleus, a valuable addidun is student-
labour. I have never undenaken an excavation, whether in Britain or
nverwas. without the assistance of nms'crsitv students, and I suppose
that in England, Wales, France, end India jeo students, or more have at
one time or another worked with me in this fashion. My debt to them
has been immease. They impose a constant need for dear exposition
and ihcrelbie ibr cloir thinking. They ask simple, awkward tûesdoos,
which have to be answmd convutdngly or with a frank and wholesome
idmiMion orignotaDoc. You can't fool them, '[’hey axe tlu:6icndljen and
most srimulatiug of critics, and the brat of ihem rapidly become the
most ctHoperadve of ooUcagues. My habit it to work them in pairs, a
senior wi^ a junior, and m give the senior coariderablc, if conmlled,
rrapoadbiliiy—indeed my Htc-supervisoia are nearly always senior
students, A note on ibcir training will begiven below fp. lyg). The whole
problem is almost exactly that of a ctmraanding officer with a bevy of
jpÔunjt tubaJtcrm*
I append a few observitioDs on each of the posts vFhkh I have
nftnird-
130
STAFF
THE DIRECTOR
It would be easy to be trite in describing the qualities of the director.

It goes without saying that he must have the combined virtues of the
scholar and the man of action. ‘Scholarship is by no means all that is
wanted’, affirms Petrie; ‘the engineering training of mind and sense ...
will really fit an archaeologist better for excavating than book-work can
alone.’ I shall say more about this matter of scholarship in a moment or
two. Meanwhile, it is scarcely necessary to observe that the direaor can¬
not be an expert in every branch of his work, any more than a general is
an expert in every tank or gun under his command. But, just as a general
must be exactly familiar with the performance—the range, fire-power,
mobility, and so forth—of every arm available to him or his enemy, so
must the direaor of an archaeological excavation be acquainted with the
exaa potentiality of the various techniques appropriate to his craft and
the nature of the problems which are likely to oppose him. He may him¬
self, for example, be an indifferent photographer, but he must know
accurately the capacity of his cameras, lenses, and filters, and must
thoroughly understand the preparation of a photographic subjea; at all
points he must be able to direa or check his expert. I have been blessed
with a series of exceptionally competent photographers, but scarcely
ever have I had an archaeological photograph taken without carefully
checking the subjea through the lens before exposure, for lighting, posi¬
tion of scale, angle, and background. Experience of this procedure (and
of the converse) has convinced me of its necessity. The direaor, and he
alone, is responsible for the record of his work; he knows, or should
know, better than anyone else exaedy what his record should express;
and he can achieve the best possible record only through a full know¬
ledge of the capacity of the mechanism which he is using.
So throughout the complex operation of modem field-work. The
direaor sets the standard of achievement and must know enough to
impose his standards without question on his experts.
But he is, of course, more than a gauge of technical values. He is a
leader, and it is a truism that leadership is based not merely upon
knowledge but also upon imagination. It is not enough to confront the
situation when it is presented; the leader may be expeaed in some con¬
siderable measure to create and define the situation, to be something
of a strategist as well as a tactician. The present book deals mainly with
the tactics of archaeology, with technical devices and local problems, but
STAFF
iht ummute {jimHly oflbe Qey-iirchactrlogist ts the wide eomprehcmiim

of the uitefrelatt€>mhip of majoT prubleim; in nthcr words^ a disidplIcLcd
iimginjitiOH- Aji out^tofudiiij! exmitple ofthi^ qujility h ai^y to ici:ngiiize
in the w^rk of Sir Artliur Evaos, the disoavcier uF Bd^[: BriuiiD and
Mioo^ Crete. It If a quality which cû ludd-aicJiBculcgy fnnn a tcch-
nicflJ sdeooe into an art.
Abotre the director must cultivate st fcmpulom aeeuxncy and cam-
pletefiefs in the oh§crvatIott and rccniti of hi^ factuaJ evidenes. Accuracy
if 3 fundMueiiial quality of ichclarshipj and chse is a ichalarfhip nf
hddûrchaeology m there is of oilier sdendfic dbdplincf. Let us pause
for a moniait and coiwider this imtter Of scholoifhip a little fiiriher,
Eacai^tion has ban dcfcriljcd by 4 superior person as *dirt archaeology',
with the implication ihat it bears soinething of the sitme reLnion w
'fcholaifhip^ as landfcape-Eardemiig does to boiany. But what is
'fchDlarfhip'> Tt mighi perhaps be defined in the btoadesi temss as
'mccurate and comprehensive knowledge adequately eaepressed". Admil-
icdly, other and narrciwcr dehnidons have found favour. The Oxfard
ErtÎish Dktiorkny dcdaces that it fei Icarmog, crudisJon; cap. profidency
in the Greek and Ijuiq knguiiges and their literdturc' ^ blandly adding
that fonuerly the scholar vfaJi nficn "one who had studied at the imivcr-
lily, and who, not hai-ing obtained any fitted anploymcnt, sought to gain
a lining by literary vrork*. Bven that nice definitioii fails m reach the
pirdfion of J. Al, Barrie's Meet Street professor who, in hut emmination
of budding journalisTs, included the questioii: 'Fahtiium, OAj va f<™
dire^ Par Nf piui utifa. What ate these? Are there imy more
of than?'—the correct answer being, ‘They aie scbolanhip, and thenc
ate twTj more, namely Janr ilf/we and TVttoPresumably the
author of the phnuc ‘ dijr atchMOlogy" would equally describe Darwin's
study of caithwrrrmi as 'diiT biolejgy"; Anyway, wc need not bother our
heads any more cither about the superior penan or about the Oj^&rd
EngUik Dt'chtmary. 'Scholaiship^ as ft monopoly of dassical erudlnon
koi in £ict been dead finoc the xSjio% when Ncaiidmhal Mon, rhe
Bouchcr-I^twich hand-axes, and the Oi^ of Species almwic simul¬
taneously iniDte it hip and fhigh.
In the liberal and only valid utierprtiaijon of 'sdtokiahip', field-
orchoeolojo- and excavation, no Iras than textual entickm, may be a
proper quaUficaikm of the schokr. Pin Rivers wis a scholar. But cr-
Ot^^tioD if by itsclfj of course, no more schoLmhip thou ii the ability tn
irjhMi a AttmU chap. tx.
STAFF
cornmic the Amad. Pitt Rim% wm a u;hekr cu^t merdy became he du^
iatU, but above dJ bcotuse hl^ powerful magimiiou ei^bkd bifn lo
vuualLu: and cteule dCW' stfliidiinl^ of sdoidlSc ciocuiacy. His super*
acenuULy wm the cordmal qifullty of bh schoki^hip. It nor simply
that he wm inethodiciilJy and pedimtisiilly ^cortitei added to method
was a oompithemion of the causative rektierm of things which gave
ttnicturtaad corparoility to his obscmtioiiS aiiJ Tecotds—'comprehen-
sivt biowdedftc odeqwitdy estpressed ^ oi T have imdefSTaTcd it. And
that compTch»^lon of camative fdaiions is a vital quality in sdaUirly
fidd-work. It isa quality w^hidi cannot beinidnted but canbeiMiueiiKly
developed by training. It it in fact a ncerasjjy' pait ofa modero archaeo-
logical educadonr ‘I'hc nxiluTionaiy dn^clopmcnt of flint Implemcniî of
potfciy-iypcs^ of script, of pictorial or sculptural tedd^p^ ti taught
today to the young nrchacologist as a matter of course. But let the bud¬
ding director htm-^f take up such a study in demll on an objective basit,
let him eitplore for himself the piHslbilitici and pitfalls in the varying
categories of evidence bearing upon some selected group of maieiial—*
let him diiidpline hia mind for some months in an exact appreciation of
evidential values and intcr-rckdonships^ and be will acquire a pan of ihc
grammar of hU EubiDct, without v?hicb he can never achieve full hieracy«
[t marim not at all that his iiubsequeat vtudis may take him to Asia
or Africa. A dctailDij study of Bmnee Age wares lU WTltshire will pre¬
pare him for a detaOed study of the chakohibic w^re$ of Persia. After
alli Pitt Rivers chose hre-anns. The procedure Ii in all essenrialt identi¬
cal^ and the switch friim the one to t^ other h merdy tbe rediiecdan of
crtiined eyes^
And while wc oxc diieusiing some of the component clcmenrs nf ihe
director, let us not omit on undersunding of s&ucturi^ rdaiiomhipp
which k nnoihci anpcct of the causative reJadon&hip of which I have
been speakings The properly equipped director is one who hai had,
amongst much che^ some apedfie archiiecturil mining^ Hia under"
sunding of the Pilact of Minos or the Ziggurai of Ur will be none the
wnrw if he has lamied the tort of things that C. F, Mitebell haa long
taught in hii (rath edJiion, 1943). Soooa or kter,
the excavator of neolithic and later cnlrurHi will be coafronted by itnic-
tural evidence in ftone* brick, or rimbcTp and ihia cviLkncc oaay mean
little or nothing unJesi viewed with an understanding of the signibcaaoe
of structural factOif, Butj tn a wider mntm, the itructural seoie i* one
which is invaluable olio cn the undencajiding of the mterrelatioasbip
STAFF
of soil-siniLTr Ic is a sotuxe of comt^t asionishiDcnc to me ro find how

lacking is the av^pge mind id the ODciMtundfrig nf whaT is and what
is iiiJi snaedtraHy pmsibk in the analysis of u sextioD. On aU grounds I
have no hestiation in urgiiig a systemaiic i.'uorae of amhifccrural ttain-
ing fJasting, say, six iD^nths} upon anyone whu propoîns m devote him-
self TO hcld-archaeology.
There is indeed no limit to the desiderata with which one might saddle
the cmbamsKcd director. But before aitempiing to impose an impossible
burden upon him, vi'e may disdogulsb fairly dearly borween two kmd$ Of
tâliiicndDns! those which can be delegated and iliose which cannot.
Antoniî the former arc, as T have aircddy suggested, photography and
draftsmanship. One of the best archaeological cxcav'atura biowm to me—
unsurpassed as a painstaking and acairaie anal^'st of sttatigiaphy—is de¬
void of any son of skill aa a draftsman. The cbtinit.'aJ, physical, or
boranicai aiuilysis of soils is manifestly the task of a separate specialist
with kborutory^ equipment. The extent to wUdx the director should be a
iingnist is more debatable. Technically, the metliods of cxcavaiion are
identicii] whether a liLoatc or an illiteniie culture be in question. Fcfrie*s
view was that *the andem language of a country, all important as ir is in
the study of ceEoaius^ is yet in its critical aspects not so essential during
field-work. But the exca^-ator should at least be able to take the sense of
all written material which he finds.^ "l^hat is a fair enough tstatement.
Nc^'erthclesSj c^cty ^prcdiiicm concerned with a literate andquiiy
should have oa iu staflF or wiihin a few- hours" reach a fully qualified
epigraphist wluj can gauge insranily and in some detail rhe poteniial
value of a tablet or a seal nr a coin. The necessoiy linguistic and cpJgra-
phical studies are generally such as monopolLee the time and cnpaciTy of
the specialist, and they am rarely be combined with the wider functions
and qiuUificatiOm of the director. With Petrie's mcnation, therefore,
linguistic proficiency may be regarded as susceptible to delegadon.
But tbft qiialtdcs which e^nrsat be ddf^ted are the insianf under¬
standing of StmCturoJ and stnuigraphical problemsj the quick and ac¬
curate cottclation of the vanoiu groups of evidence as they appeîr, the
reasoned appreciailon (sometimes miscalled 'intuiiion") of the immediate
needs of the wYirk us it proceec^ from day to day> the capvdty for
ufiddaycd hut well-founded decision, and the ability to ensure even
progress in all the various departmenia and sub-deportmcDts of the
aiterpriw. To these qualities must be added a clear antidpation of the
needs of the nlum4ue report, and the Bipadty to present tl^ report in a
134
STAFF
clcar^ onnd&c, and LnrcEli^blc form. Tn oiher words and in brief, a prcci&c
and Trained mind and an informed and informing imagination arc the
qua li ties of rhe dirccicr of an archacologicaL espedllion. And having
said ibai^ I am av^'am that I have canied the reader :9earccly at all aloi^
the upward path. There arc, however, a few praetico which I have
found of me in an attempted approximation to thxi^ qualities in die
held, and I note them for W'hat ihqf may be worth.
Fint and foremmLi the dixcctor mmi be a free agents ffeefrojn admini¬
strative detail. Hij primary and cemtant duty is to circulate irom $lte to
Site and from workshop to workshop. Every seatan^ in its laiest mani-
festadon^ must be clearly m his mind's e>'e, and he rmiai be familiar with
every development in the hour-io-hour work of hh team. If plans or
seoioos are being made, he must himself ensure thdr adeqiuc>-; if hia
draftsman is drawing poTTCiy^ samples must be tested. Hix surveyor must
be overlooked critically, his photographer uiihscd and supervised.
Above alJ, he must familkriKc himHcif with the gtuups of potiety as they
He on the rray beside the work or ax they etnne froid the wa^h^ and he
must occasional ly check the marking of them. The indexes of small &nd^
miisc be inspected daily. At some oiomeiiL or momenis dixriug the day^
he mast examine the held llotcb^lKJks of liis supervisors^ And he must
keep his colleagues and employees consiimdy> in varyiog degrees, 'in
the picrune", and ensure that they are* each of them, awue of the iia-
portanct of their individuaJ contributions to the peogress of the work*
^rhis last point Lh essential to the rnouiL^maiice of morale tuid efIideiKy.
It ensures ihe fullest and most mtdJigent co-operadoa of ev'cry member
of ihe Team. Once more I tum to soldieriog for sm analogy. Ooe of the
grear quajiricii of a ccrtaiii celebrated commander was his habit of taking
almost every one of his soldiers, from general to private, into hb con¬
fidence before a battle. Almost every Erittsb fighting-man at til ■Alamcin
wax a partner to his gcdctal's plan and could oolbborace imelligcntly
therefore in victory. As often, military and archacologicaj field-work
march alike in this. TdJ your units before you stan Egging, ccictly
what problem you are out to solve, how you imend m rackkr it, and^
when poxsibki, the sort of time-table which will probahly be required.
This lost b poiticiilarly desrimble with a idativcly mexperienced staff. I
recall, for instanoc, how ai Arikamedu, in south India, w here W’c sought
and found for the first litne cxicruivc evidence of i Roman culture In
significant oontaci with o hitherto unknown Indian culture, I explained
beforehand in detail rhe naruTc of our piublirm and our methods^ and
STAfF
ended with the wTuning that îgnihatDi resdts n«d uQi be expected
for Q fortdght. fn fact, the c3i«v3LTioD opened bodl^j the site bad been
much duitiirbcdj and for nine days nothing of importiyicc was found.
Therij on the tenth day, one of my indian srudents emerged excitedly
frora the muddy depth5i of a cutting at sca'Icvcl with the stamped base of
an Arceimc dish in his hand. Thcrcafier the results were overwhehningly
satisfactory^ but in those firsr days, with an imttoined staff and students
and the mercunaJ south Indian tcmperamcai: to contend wiih^ had
been saved from something approaching disaster by Che timely wâminp.
This i$ cot to say that all can be foreseen and thar there is no place for
oppoTTunism in digging. But the w^hole framework of the problem must
be very carefully constructed before the work b^ns, and the use of
oppominit)' is limited by that framework, w'hich must itsdf be compre¬
hensive enough to admit it. HetCj h-owever, I am ag?ijn crLcroachlog upon
rhe general problem of ardiaeoLogicaJ strat^- (Chapter X).
And all the rime that the direaer is attending unresriugly to those
yanous and vital details^ the gcowu^ xecumuliTion of evidences arid in¬
ferences is worfcLflg upon his mind» cieatiog new tdeas^ suggesting work¬
ing theories or probabilities for trials couhnning or bending his plao. He
alone can (and must) know u!i the ktojt cvidimees as they emerge and
can £t them imo a logical pattern. Nor should he defer tins operation
until he readies the remote sanctuEty of his study. Inferences of a general
or far-reaching kind w^dl coDStantJy refer hack anew to detailed observa^
rion in ihc heldj rttiuiring fiesh trini and appreciation of the material
evidence. 'I"hc more thinking that the director can do in tbe actuaJ pre¬
sence of that cv^deflC€!:^ the better, and it goes without Saying that the
denial inteiprctaiioo must be absolute and final La that presence. Says
Ferric:
At the moment rhar a fact Is before the eye—* faer which may never be
seen again, and perhaps never panlJeled'^^t la needful for the observer
to make cerlain of dJU ihe debJUs, m verify evtiy point which ii of fresh
vaiue> and to record all iliar is new with eertBiniy and eiêtltude. . . .
Everything seen should be mentally grasped, and its meaiiing and bcai-
Jngi comprehended at the momeni of disccivery^ so dearly that a definitive
ixatement can be madci which shall be ns certain and ai absolute aa
anything can be which depends on human seme.
There ii but httlc pbee here for the ifcKa/wr, and the direaor
must be sufSdendy hberated by hit staff from routine cates to keep his
W'iti actively and unceaidiigly about Mim A director who knew^ this and
136
STAFF
cho^ hi^ was nuorc than odcc accused taking

credit for the work of otherî. The Hcciisadoo was^ of coui^j, correct^ he
knew his job and chew a sTaff capable of doing much of hh work for him.
Perhapsj afieralli the greatest qualihatioii in a director h the capacity to
select Afld irain a cnmpcTcnc sta£f. Let u£i lum now to thot Staff and con¬
sider some of iw rcqnmics.
THE DEPUTY DIRECTOIt

WhiJsr the director h the opciationfll leader of the expedition, the
deputy director h prinmriJy Ln charge of the administnitiutL He (or
she) deals with Oic w'cU-bcIng and pay of the various grades of staff, emd
is pcTNonaliy responsible for the smooth working of the machine- He
should also be B Limned fieidârchacDlo^t^ in dose touch with the
director and able in cmergeiicy to rqircscnt him^ but is not offidfllly re¬
sponsible for tJie sdendffc aspects of the work. If he is a speemlist in
some relevant branch of archaeology, so nmeh ihe better.
The following are amongst his duties:
(a) Eilklin^ tr hoHtmg. This problcni, of course, varies widely with
local conditions. It ranges from the snenubm task of securing: lodgmga
in a provindal English town to the rclaiivdy easy one ot building a vU'
Inge of grass-huts in Bengal. Whatever its. precise nature, it must be
tackled w-ell before the actual work of the expedition is timed to bepo.
Spcdal utteutioQ must be paid lo the xdminbLrBlive office^ the drawing-
o£&cc, the photographic ^rk-toomj and the pottery-shed* In certain
cxpcdldom 10 the Near East bcrwecn the wiirs^ an lumeosssairy degree
of Injury was sometiitie$ sought at disproportionate cosi in the living-
quancTi of the partj . On the oihcr handi the poioriously Spartan con-
didoDs under which the PerricR used to run their camps reached the
□ppositc extreme. Unnecessary dtsoomfort amongst any DDcciidcrabk
body of people for any considerable time cmi only lower efficiency and is
itself a sufficient symptom of iiicfficienCy+
(b) Equipm£ni. The direcior is imiEtcdiately conceriicd in the choice of
cquipmcntj but its preparedoEss and maintenance arc Largely the duty of
bis deputy. The iheodolite or its equivalent muM be cheeiked, the sur¬
veying-poles must be kept freshly palmed (a shabby sujYêying-polc »
sometimes hnrd to see, and another cmblcincjf incffidency), ihcremuat
be on adequate supply of bubble'Iet'els and wch-conditiotvcd meamriiig
tapes, tlicre must be good string, pegs, labels, cloth Or strong paper bags^
IJ7
STAFF
otiiclxKila with wacer-pFOcfed covcm, drawtcig-NHiids and pajicr, and

so forth. All these are the proper char]^ cf the deputy diietior, and art o
ramtanr drain on his atlentioiL
(ej Aeamiiti, The tingraterul task of keeping the ninnijig accounts of
the expedition ialls to the deputy. Xhia is no Ught luatlerj bur a deputy
who tanodt carry it Is lUi unjiisi^blc addition to the peisniincj. Above
allj he must deal accumtely and promptly with the weekly wages bfU,
including the dresnrae complication of Ixtkshcff/i for special 6nds whfte
the hatcM boniB-systeai is in vogue. Any son of cirof in the payment of
wn^ ttud baki/uesh reacts insUmtly and unfavourably on the iclatLons
berween Riastet and men.
(d) Sufflfcs, The uiganisatioa of food-supplies is the duty of the
deputy, with or without the assistance of a 'bnusefceeper'.
^ (f) floifiiial. Tlie hnT-aid box and| if necessary, a supply of other
simple remedies arc irain:niiied by the deputy, and o daily ‘hospital
hour’ is uotmally ncecasary in the morc remote parts of the world A
know-led^ Of first aid is thus add«I to the deputy’s qualifications, to¬
gether with a Sympathetic firmneis in its application.
StTE-SUPERVISOHS ^
I have generally chosen these from iunoDgst my senior students, that
13, students with octe ur mote seasons' training behind diem. They are
in charge of an area sufficiently compact to enable than to keep in im¬
mediate contact with ali wnrfc done in that area throughout the day.
Scarcely a shovelful of eanb is removed save under their eyes. They are
personally responsible for the small 'contipl-pro^^êti;^ nmmally some
2 Or 3 foct square, wMdi are cut ahead of the nuin vvork to provide a
preview of the sîa (p. 66), 'ITiey write up the area notebook, keep
Sketch-plans and sectioiu, and record Special smajl-fisds. They insure
that each snartim Eus its separate small-liiid bo* or basket, with de-on
labels imd supplemeotarj' loose label (p. 159), Their equipment is con-
rained m a porrable ‘desk’ or box, for which they ok individually
accountable. So Gtr as possible, they have the aame workmen from day to
day, get t& know them individually, and allot thcni tasici suitable to tlicii
various capodtim. They are. In fact, piaiouriâiniiijandcrs, and on their
efficiency depends ultimately the technical adequacy' of the excavadoa.
Wherever feasible, they arf each aasined by a junior student whom they
help to tram.' ,
133
STAFF
FOItEMAN
The forcmîD is the seigeant-major. He ha* pmir«l himself as a

and aheuld be [Jjc besE workman Ofl the sire. He may assist in the un-
rovenng of especially fragile or kdpormm objeos. liur he irmst, above
all diln;^, be by nature capable of controlling his men with firmncsis and
sempulous fairness. OP flrdviiJ in the morning and (In the East) after Lbe
midduy break he coUi the loUj and he maiTshals The men on pay-day. He
should know his workmen indi^ndiially from 3 somew^hat difTeroit anjglc
frum iliai of the site-supervisor, and should show patience and assiduity
in their training. At iHc end of each week^ he b respOP&ible fM" providing
the deputy direcror with all mfonnation required for the pay-shccL
The best foremen are also enthusiiisLs. In EngJandi \i1lliain ^'cdlakc,
my foreman and mUcague at iMuiden Gastle and in Eranoc, Thomas
Hepple, who assisted jMt. J. P. Gibson and .Mr. Gerald Simpson on
Hadrian's Wall, and W. E. V. Yoims. Mr. Keilfcr'i foreman at Ave¬
bury, arc three outstanding examples of the foreman-archacotciguii and
friend. In India, T had one such^ a Fonjab peasanr, whose mind was con-
sranriy on bb work whether he wa:$ on dtiiy or off. and who used to come
and dbeuss the problems of the $ite with me in the evening. One Of
more of Sir Leonard Woolley^s Iraqi foremen. I believe,, used to travel
vohintarily some hundreds of miles every year airoii iht dcicrt in order
to rejoin hift old chief’s stoiFduringiitibsequent work near Atnioch.. Such
friendships are amongst the highest reward thai a direcior can desiter
ITicy bridge the ctaas-room Eind help m link mfened fact with basic
eanhy knowledge.
XhE SMAIL-FIMD KECOnDik
llie recording of u 'Ond' begins ar the moment Of discover}'. Unhas it

requires special treatment Of aitenrion the object is plnced immediately
in an adjaceciT box of tray Of basket to which is tied 1 label bearing the
code-name of the site, the pumber of the square or other sub-unit, and
the number of the stratum, ^ hc Inst I always place within a drd^ to
distinguish it iionihtakably tixim all other numbers. A dupliatc label,
loose, h placed in the bortoni of the reccpraclc. (Sec Chapter XI lit
pp.
At a ^-uimble moment, at the end of the day'^s work or earlier, the rc-
cepuele is transferred to the ^pmaU-hud-s shed, where its contents are
cleaned, SO far as is suitable at this stage, and marked by the smaU-finds
STAFF
recorder, either with Indian fnJc or white psunt on the objects themaclvcn
or on kbels liutl firmly to tbenJ. Sooner or laicr, it is useful to cover the
mart neatly with shellac as a protcctioa.
'fhe recorder's aeai aa is w card-index all individual objects of any
possible Significance. I always maiolain two card indexes, one by cate¬
gories, the other by sites, See p, 164,
Such obiecis as require fini-iLid are then handed over to the fieJd-
hbonuoty for ueatinenT (p, 169). Moat of them, however, can be im¬
mediately packed for trajuportaticio to base, cither in boxes or in ba^pt of
paper or doth fp, ifia).
A high respoasibility nssts upon the shoulders of the tecoider. He is
personally responsible that erery object is dearly and correctly uiitrkcd
on as inconspicuous a surftce as possible; no object should be spoilt
for ^hsequMt photography or display by nn uiuightly code-symbol.
He IS aLw responsible for sioring the boxed nr bagged finds in siidi u
fashion ihjr they an quickly accessible if requited for refncnce iu the
field—a very import flLL dmiderartuu, wmctinicaflvertooked. The fiads-
shed must, to other wurds, be in impcocable order. Muddle means
inaccuracy, delay, incompetence, and the dmtiuction of evidence labori¬
ously pmered. At a small necavadon tile prohlcni is proportionately
simple, bur at a large one the recorder will require a stafi, which must
be subjocTcd by him to the closest supcivisinn and diadpline. Every
addition to the staff multiplies manifold the risks of inadequate or uven
incorrect recording, and the recorder must remember that on his UTICCIK-
ing rigihnee depends much of the scientific value of Jm expcdidon*k
work. He is a pillar of the enterprise,
TJI£ rOTTBBY-ASSiSTAST
TIk duties of the pottery-assistant equate with those of the souU-finds
recorder, and on a small exavarion the two posts may be combined-
But at most exmvatiom, particularly in the East, where pottery is liable
to occur in embamtsing quxntiiica and is usually of the highest possible
evidential value, ir is csseniial tu have a truined assiscanr exclusively in
charge of this department. And a very exacting task is M$. His pretimin-
aiy qualificarions must include an extensive knowledge of prcviuculy
recorded diacovery, so that he can instantly spot analogies amongst his
new material and cm biiug them ar once to the notice of the director-
He must also hive a scuad general knowledge of ceramic techniques, so
140
STAFF
tkAl, (Ot ™inp]e, he t3in protect from the emmitwrous poiier}'-washcra
sitch waxes (c-g. wares painied after batdng) require special rrcarmcDt.
He must be a utrict di^dpHuariimi and possess a clear and order]]f mind.
And he musi have a simple* effedive remtiae for the marshillsn|r of the
fairly considerable wurking staff which, on any l3rge?-scak occaratiunt
must he alloncd to him. No hampering economy ^hnuld ^cr be im¬
posed upon him^ if the cJtpedition omnoi afford an adequate pottery-
dcpartmcni* it ithould dose down at once. ITic hickncycd assertion is a
true one: pnitmy is the alphabet of archaeology. J ra plasiidcy lends it a
special facility in the Imitaiion of other fabrics such as metM-workj
leather, basfcct-wôrk, or other pouery. For esampic, Romnn ierra
imitates meral-work nr glass-work* soine HroTi3rc Age beakers imitate
wt>veij snrne ncalithie pottery ia Wcsrtm FnropK imiiates leather^
work* some poncry both in Europe and in India unitutes Ronaan êrît
and so nn. Pottery Is thus liable lo preserve the influences of
Other wares and maicnaJs smd to represent adiunil or LadustriaJ inter¬
actions of vaiions and often vital kinds which wonid otherwise be lost to
us. Its fragpiliiy limits its normaJ durability and, unlike oietai-work, which
may last for gencrarionii or centuries and is also more liable to be
treasured for inrrinsic vnlucj the date of its dcstruLiion {at nny rate in
m.iw) can be approximated to its date of manulacuire. That same fragt-
lity tends to restrict the diffosion at least of the commoner wanes to st
measurable vidnity of ihcir Idliis; although it Ls saluiary to recall ihar
Graeco-Roman amphorae, owing ns the popularity of the whines which
they contained* found ihcir way la nutnbers m the shores of the Bay
of Bengal* that Roman glass reached India and China where local glass
was almost unknown* ihai Roman Arreline w-are found a sale in Muth
India by reason of its superiority over local pmducts* and iIwt Chinese
celadon wore penetrated far and wide ov^r Axia and AJrica in ihc ninth
tind following centuries a.o. partly becemre of its quality and partly
because of the forceful salcxTFiBnshipof Chinese mctchanis. Bui whether
because most pottery is foiriy loi^ or because some of it is asTonishinglj
and signiforantiy WDoric* it is of the first ImponaRcc to the orchacoloipsh
and the immediate and faukless record of field-evidence relating to ii is
Q primary tairc of ihu dxfecTor aiid h^ admirustratjon.
The pottery-assistant is therefore a very irapofttmi pe^n. liis
methodhi will vary witJi his own idcaa and drcumstaiices. in ChaptcrXIlI
are discussed ptoccdures îriiablc on the one hand to a tcmpuatc (often
intcmpcTatc) climate and on the ether hand m the more predictable
141
STAFF
tiopicaJ DT fiub-tropicfll climates whnE hidisi of his work fiHi be happitj
tloDc out of doors.
THE PHOTOGRAPHER
The ascoual qualificsdons of the held-phutogiapher are (a) that he

shall be icchniially dependable in all liglits and colour^schetnes, aod (A)
that he sboll be quick and ingenious la tiic improvisation of scalToldi^
for his cameni uid, in particular, of a readily scccsaihle dark-ioom; he
must &i all Limes be in a position to produce the devcinped plate or fUm
for inspection by the director witbio 20 minutes of exposure. To these
qualities he must add a ihorough undeis Landing of the preparatioft of a
subject, wlwLlier in the open air or in the studio, aldiough this is ulti¬
mately the personal responsibiltry of the director. Fun her notes on these
marceTS will be found in the chapter On plmtography (p. 174),
An intcUigeni and experienced photographer is a ang gua atm. From
the begiuniag of the excavation, it is iris spare-rime job to mat..- „
study of the site at different times of tire day, and so to be prepared to
advise in regard to %hiing and position. Here again, the director is
primarily responsible, but a good and observant photographer am help
him 3 great deal. like a doctor, the photographer must be iivEtilable and
prepared fli a moment's naricc and at all times. On an excavotioa of any
magnitude his duties cannot be combioed with those of any other mem¬
ber of the staff.
THE SURVETOR
The prepamtioa of a njewured drawing of a seciioa is the duty of the
supervisor, under the eye of the director. The supervisor and the direc¬
tor know more about the atrarificatioa and mcaningof the sections than
uiyone eJse^ and the ti:Lk csjinnr ddegiiLcd,
^ On the other luiod, the preparation of a oantoured stuvey or the plan
of a building, Ihough, like everything else, demanding t-nn want super-
s^sian, must on a large esavation be ddegaicd primadly to a piufcssiuoal
surveyor, I do not ptopuse 10 include here a detailed einir?ns on the
ctofr of surveying: many handtwAs ate available.' But there are one or
two observations which may usefully be made in regard to the
tiem of profesifcnal survey-meihods to orchacnlDgy.
' e.ff. T. Baker ant! O. M. L.en(Ui, IjtnJ and tugittmug fonwriir figth cd..
T.onîi, Tw,j)j K, FOrfx snd W, K, Jcnkiiii, LatuI Surzrtyinr (jib ed., Lunden,
iMWi K. J. C. Atkin-wn, FktJ ArehatmUigy aondon, IM*!); A. H. DctwnlEr,
Aidutidt qf AtiJLiaslêid fNew Haves, 194R),
142
STAfF
Mrst let U5 caiuiidcr the naiurc i:rr the which oie likdy tO awiiit
ihe arehacolijgical ;&urvcycr. Thc^ê (21) tlit preparation Of a
plan^ gcncraJly cuntoiircd and ofica of comklerablc ttx&TiKi and (A) the
planning of a sitnictiii^ or group of sLrucrurc& on a small scale or in de¬
tail. Far (a% ddict a plone-iabk or a tiicodolite may be used; (ft) is nor¬
mally canitd oul by LtmuguIaUoii from a base-line. In bo^ caregoriea. ii
ia impcrtiuii to remember die degree of accuracy required. Here I tread
upon delicate ground. Throughout this book 1 have srreased the vital
need for accuracyi and I would countenance no Âithdrawal frani the
piKiiiiati. But tlicrc is a hind of accuracy which fniAiraEcs itself by its dis^
pioportion to needs or possibiilties* or may ei™ result, paradoidcsdlyt in
m essmiiial inaccuracy. In my es^ricnoe, prafcsslaoal surveyors are
liable to lose the couhLhnnadon of the wood in the plomung of the trees ;
to lavish so mudh care upon the sllghi irregularities (maity of diem often
secondary and acddeural) of an ancicni wall oe to loisrepresait its essca-
tlal choractej: and original intention. I have already remarked upon a
similar 'ovet-accumc)^* on the pan ofibc inexpêritmeed in the tendcfitig
of a section (p. 57)1 In all eases, the degree of "accuracy’ must be com-
mousunite wiili the scale 10 whieh the plan Is to be publishcdi A plan
drawn to the scale of i inch to t foot will commonly bo published lo the
scale of it inch to i foot, and on tkuL scale small escctcsccDoes of flint
workj or suiMl vatiations due perhaps to subsidence or other defor-
matlOtti will cease to signify. Omit them ihetefoce; but do so only after
careful coi3Sideration4 In a measured section it would be uifcrly iwJcss
pedantry to measure crcTy small pebble in a gtave! layer, and the same
need for proponion and commoQ seme is prtseui Sn the ground-survey.
Do not be more "accuraic’ tban your scale permits i do not nvcrloid
your machine. We all know the dangerous pedantry of ihc ulua-cureful
moiorisr wins, in season and out^ laboriously nfee^ every rule of the
I lighway Code. have seen how' aggrie^'ed railw'aymcn or mcat-poncrs
can hold up a community by a dehbemte adherence 10 [he rules of their
trade. With less deUberatioa, perhaps, but wiib «iuai pedantry, the
professional surveyor cou not Only hold up an ercavaiion but urn imperil
the real accutaty- of its records. Watch your surveyor, and traift him in
the difficult but neccssaty art Of disringuishing casentuds without In-
validacing the impenonality of his record- It can be done.
For arehacolDjdoil purposes^ thcreferfi^ I prefer a somewhat simpler
insmimcnE than the regulation thcodnUte- An artiiiciy "director^ de¬
signed for quick but «5«itially accurate survey^ meets much of the Deed
143
STAFF
And j$ Q far hiuidier nuichlacj und imny hiivc cquiva-

leni icstrumciiES ua Lbt miirkci.
SimOarJyi^ the ofuiiiiiiiicrablr ^checks ^QJld dlagoiuib from a ha^c-
lint so a miU-syssein^ with an accumiiladoD of $iiiaU is nUm.
besTor rcplacied by the mrcfiil survq' of rwo termiod points on which
she plan j$$wunj; froio a qdek series of Joed mrasuiTtmeciLs. Thereby, i he
correct local intertekuonship of the walls of a b\iiidtug—in otJiirr w'ords,
m csjCDtiiil chajracier—h often more accurately and speedily recorded
Than by mote elaborate professional methods^
Having Mid iMSjr I mubt end by warning the student agdim theoppo-
site citrezne. A very eminent arehaeologisi of my i^qiiamtance so
Mt out on a surveyingHncpcdJtiDii with two pea-sticks and a visiting
card, iJwr former for 'line^^ die latter (sighitd oloag die edges) far âIsglc^
I saw sotnt of his resulLSr and imr^'cUcd that they were not more tn-
Bcrumte than they actually were. But then he was a geniiiH.
THE FJEnn-CHlMSST
A small ftclddaboiatory and an experienced and resource ftil chemist

arc essential save in the run: insTances where a stade luborarory' is wjiiiki
easy reach, llie work of {X}mcr%'aiion in the field is limited as a mlc to
first-aid, and it h highly unporeani for subsequent nearmenr thai the
ftcld-chcmist shall keep a cnrefliJ and detailed record of all action taken
by him. This record must accompany each object to the base^laboratory,
since subsequent action may be enndidened by it. For cx.impk, not long
a work of sculptiue arrived in Brussels, seemingly in perfect con-
didon; btic shortly afLeiwards ii butst into pieces—the excavator had im¬
prisoned salts within it by imprcgfiadi;^ the surface with ccUuloid, and
had omitted lo inform the redpioiQ.
It is paniculatly importanr that the field-chemist shall be able lo cfeat
coins in any metal widi a minimum of dday, since their evidence wifi be
required at once by the direeTor. For the rest, the chemist's main func¬
tions arc (rt) to pre^'cot funher decay, and {h) tn consolidate J&iable
objeets sujfidcnily far removal and tnin$partition. Thtse fiinctions will
be amplified in Cbapier XIV (p, 169).
THE OlAFTSMAK
The work of the draftsman is vxricnis, involving a range of qualities

rarely possessed by aoc and the same individual. In practice* I have
*44
STAFF
^crally faimd it to employ ibret draftsmen, althou^ 1 have

JcnDwn one or two cxcepLiopal men who have covered the whole ground.
Whether one or ihfce^ ifis e&senliid tb« thedrafisman shaU iSnmugtily
underatand ilic ledimquc of Hoc-bloclt reproduaion (p. i^), and ihaU
Eo be able lo aiiuue hi* style to the machine: lo employ Iiok of the ri^ht
thicknc^^ and QrmnesSj baving regard to ihe extent of reduction requited
id pubticttiion^ and to avoid too close a pmximiry of one line to smother^
EUch » may Ictd to a fusion of the thin menil mrips that will uJiimately
rcpreseni thiKC lines on the b!ock?i. Oeasi lines and dean hatching^ with’*
out an unnecessary coarseness of detail, cj3t2JI a inieocss of hand and
eye tliat^ combined mth expencoce^ are a primary qualification in aE
types of sdentUic draftsmanship.
On that common basisj the wnrit falls roughly into three categories:
the tracing and lettering of map* and jânSj the drawing of obiects and
the semi-schematic drawing of pottery'. In ad tliesc things, the criiicail
eye of the director is a present need As BlwnyS;^ the dLrectcir must be
able to impose his judgement with an acKptahle authority upon his
draftsmao. At the same limej the draftsman's initiatit^ must be en¬
couraged; hiE advice, oommciits, critldsm are^ or should be, always
worth listening lo^ and I hav'C ofieii profited from them. Niat only in
matters of scientific inicrpreiation^ but in. the atiractivc presenEation of
his materialj the draftsman cart help the archaeolngist to^get nvei^ to liis
publit:—his sdenTific no less than his nOd-5pedalisi public. Few scien¬
tists 3« so devoid of humanity tJiat they will ooi look more eagerly and
attentively sr an attractive diagram than at imrclin-cd gcDineiry. The
placing of a plan npoa the page^ its reladonship to ita lettenog^ scale^
nonh-point and key, the character and placing of iia border^ the render¬
ing of a bead nr braorltt in dear line from an exptessive angle, are aU
factors w'onhy of consideration. These aesthetic attributes do not impiy
iny speda) mcaaiaie of clab<>t3tloni indeed timplidty of espressioo Is
as impoTiant in ihustiatten OS in text- Oocasinnally* sdentific Olnstraaeti
has boot cxcmsjveiy elaborate. Ciood laste is the yanistlchi but good
taste, ia a very' poritive quality* partly a matter of Lnstina and panly of
ddibemtion. Let tlic director and his dmftMnan ddJberaie frequendy
togcthcTj profiting from their combined good setae and such lostinct as
may have been vouchsafed tbem^
The diOcrcitce betwtên a weJS-piescnccd and oa iU^considerfd plan
if intlccd immense. I have referred to the placing of the border: l« k be
so placed that no Spoce is unnccEssarily wmted, but in such a fashion u
uTt t4S ^
STAFP
Lu eiuiblc the actual plan to be balanced with the Ictrenng^nd pl^$ing to

Lbc cyc. Whether the observer fiiliy realis^r it or nor^ wa^tc spac^ db-
Iraeiini; and disturbing. Equally^ of eaur^, overcmwdtJig be
avoided^ but this is a less frequent oirrie^ alh Jet evfty Wdrd and
figiiie in the plan be easily legible after reduaion by the bl'Sck-moket,
Far better let ihcrn be too large than roo situtII; nothing can be POore
rmstraiiiig and infuriating than an indiffcrciirly legible coirnnen
ftdiiug. And Jet the draftsman avoid all unnecessary underliae^ and frills
—fl vice to which engineering draftsmen have been particularly prone.
How far he should let himsEif go in the maner of addiUoiml and sdenli-
hcalJy uonecciHary dccaradon is, be it repeated, a matter of taste. The
old map-makers who adorned the seven seas with whales and ships and
wrote ihcir titles upon sccolls sustained fay cherubs^ or Stukdy who
added appropriately romantic mi ns, produced works of an which often
disguised a defidency of sdenTifle information In recent times, Mr.
Hcywwd Sumner has oinclied his pkcis md maps with a tlicur ihui
is often a pleasure to the eye but is sometimes overdone.* Agdn> die
Ro^'ol Cummhision on Historioil Monuinents (England), in its ITej/mcr-
hmi and other recent ^'oiiurfcs^ has made a brave attempt to relieve the
monoiany of small plans of huT-drclcs and the like by ihe addition of
large and highly ornate titles, which are tvell meant but not oJw'ay'a rele¬
vant. On the other hand, a nonh-pomt is fair game for the dcxnraiijr
within modexadon, and has a well-established iraditiou of ornament be¬
hind it. Gcucially, when in doubt simplidty is best.
TliCdpFEfjifffi of an object such a& a brooch by the draftsman invôlvm
it somewhat more det'doped artistry. Ptr cuntra, i have knowTi drafts¬
men who cûJd achieve this but who were quite unable to rule ihe denn
Imcs of a plan or to produce dear and aitraciivc Jctimng. Here wc need
On aitist in the more orihodoic and creative sense of the term, and he is
often one who scorns ihe more mechanical technique O'f the architect's
dransmam The ardsix howeier, is inclined to impose his own personally
Upon the motifs of another age^ and the obieuive securary of his render¬
ing is sometimes questionafalc. He needs watcbung. The choice of the
lioprcs^vc angle, the ment to which his picture caiUs for a diagrammatic
element or supplemcnti arc matters for the director's onosideradoo^ and
the director must, by training and experience, know bin parr.
> For «i»mplet in hji Daenpittn Aâniar ttf PffUny Sim at
&e. ( Lundvo. ^ g^cneral map ho a huvily wer-wBiBhted and uii-
nccemry borderj hut hia oiliei phuri md diasruiLi arc^ in ntv lukiiErmnitb po’lcct-
146
STAFF
M^niion may be adikd of a draftimim*s mdc which 1 hsive on occH'
siem found carerndy useful—the PQ^lng of ubjL-cis on to filasi
before their reniDval fioai the soU. Iron ohjcccs in purdcufor are some-
times found in ho friable a cotidiiion thai they have to be waied Ot plas¬
tered (p. 171) before they caa be lifted- When that need wises* an im-
mcdiarc record is csscoUalj smd the simplest method is to place a sheet of
glass (e.g.anold photographic plate which has been donned and spon^d
wh ha lump of cotton-wool dipped in alcohol) over tiic Object and to trace
the latter in Indku Ifllt Od to the glass, Negiitirt coruact-pTinis can then,
be made from tJie wktd glass, and in due course further pwirivc rmeings
made from them* whilst the glass can be again cleaned and rc-me<L TIjc
nicthnd is both speedy and accurate,
Lastly* there h the very Lraportunt question of pottcrj'-dtauiiug-
This is in some ospeers ibe easiest task of the three, but likewise involved
careful tmlning- Accuracy is ail-importwitj and can be aided by a variety
of devices—notably graduated horizontal and verticil rules with a
Kfuduated WTO ihar can be slid op and down the vertical rule. The pot
h placed upon the fiiccd hoikonul rule against the Jked vcrUoal rule*
the adjustable vertical rule is owved up to ii, and the adjustable hori-
Kontol rule is moved down to the top of it* The pot is thus framed in a
rectangle graduated nn all sides fo inches or ccndmetics* and offBcts
am be taken from ih™ to the pot w appropriate intervals.
Pi sherd cannoii of course* be so treated. It must be wedged vrith
plosddac at the correct angle and measured ftom a vertical scale-
Great core Is nccesjsaiy to ensure the cotreenuss of the angJei by planring
the rim (if such it be) foully upside-down upon the board* or by leadng
the horiDontaliiy of the wheel-marks on a whcd-iumed sherds Fhc dia-
unetef of a fragmentarj' rim i$ ascerenned hy maiching the rim against a
chart of ctmcennic dtcles prepared for the purpose..
The standard practice now is to nendex half the pot tn RociiDii and
half in elevariniL Ali p^lf shtmld b£ drasm Hfi-ske unless occsdscly
larger the usage SKomediiie$ adopted of drawing them half-aiM saves
p.iper but incTcaws uwOCuntcyK Narnally, no pot should Ik illustraied
in a published report at le^ than oitc-quMter actual sisw and a
decorated pot (c,g, omanjcntcd firra should be reproduced
half-sixc. %^cry large pots* and thoc onlyj^ should be reduoed to less then
one-quarter by the block-maker.
St is Romeduics desirable* espcoally In hand-made poticrjv to indicate
the icxitire of the ware by hatehmg tbe ‘elevation side of the dtawing^
147
STAFJ=
This aniatry the kind indicircd in our secoodl category of

drathmansbipi md mu^t be wdl done. Above aU* both in the outline and
in hatchic^t of the drawing, care iniist be taken to avoid giving an
CTCCSsively Eneduinical aspen to a hand-made pot. Lines such as those
of base and rim should generally^ in such potteiyj be dravvo fra-hand,
nor mleA
One mere point: make Tiitrc that the draftsman appends to each of his
drawings of pottex>' and other ‘finds* a tmnscription of the appropriate
kbel.
LAHOUHEHS
Digging is a shilled etafi^ and miiny years ago 1 began a nofc on ihh
subjccT with the words Âbjure voltintoty laboiiJr\ Today, bi 1953,
vo] untaTy labour is Hobsonchoice. "ITio old^f^hioned Biid^h lahourcr
sitn'ives only in a few odd camera of the land; indeed^ so rare is he that
one h someumes indinod to daw him with the myihical Mrs. Harris.
A note, theftfofc, which before the war might have mn to some lengrR
may now be compressed inio a few par^raph&j most of which arc
applicable to labour in geoerd and are not couEned merely to the
(former) home'pnxlLici.
Let me say at once that my esperience os an employer in Grcar
Britain is b^sed on England and Wales; as an archaeohi^^ti to my great
loss, J have ncvcf attcmptcdi like the Renum legion^ to ^curb the herce
ScaT*^ On the Coniinau^ J ha^^ employed Bretons and Normans, and
in India a great variety of races, moging fmm tho tall, regal Punjabi
Miaunlnruin to die tiny^ volatile M^drasL With Arabs I have never
worked, although I have ofien cfiough watdaed them with 3 cririco] eye
upon arcliaeolo^cal excavations. *rhey do not appear xo be a vety dif¬
ferent problem from the north Indian Muslims. In facr^ 1 would ha2a.rd
a guess char, fxom Camaivon to Calcutta, the basic factoni of labour-
cootmE or, in the quaim laminofogy of die iinny, 'Man-management*
arc very much the same^ imparrial dealing, a readiness to share discom¬
fort, provision of such eomforr as may be feasible without o^rentaiion^
and above all an oocasienal gleam of demejilafy humour (fwer Mreasm),
summarize the qualities rcquiicd everywhere of the diracror by his
workmen. Something more will be said about these quaJirics, but first
Jet us cDiuido: the techniod BDcomplishmects required of The w-orfcmen
thcmsdvsp
The local tools used by the labourer will differ considerably from
MB
STAFF
place to place, and a scpanitt^ section will be dcvoicd lo tK™ (p. 15^),
But ii stands to teiôn tii*l tb^ wortamn sliall ihomu^ly imda^-nmd
their iisage^ and that hiS raind ond muscle shall be accustcHiied to them.
A drawback lo the kind of labonr that usually comes tea the £eld-aichaeo-
lo|5isi today io Great Britain is that it is o^cn unacquamted with pick
and shovel and has to be taught fiu/i£p; it cumiata of unemployed
or unemployable invalids^ garage-hands, drapers' aa^lstantSi imiversiiy
students und the hke^ 10 whom pidsa and shoals are encumbfunccs
mtlier instnimenTS. On the other hand in certain parts of Asia—
Palestine, Syria, Iraq, some districts of India—ace men whose father*
and grandfiither* have worked for British, German^ Of other archaeo¬
logical expeditions^ and who have inhmLed a useful undetstandingof the
fliatter. And indeed^ to he just, the university studentj if he is; of the
right soKt can nmally be trained without undue delay to a fair measure
of technical compcicnccj only^ he could usually be taener employed in
CDore derailed and spcdalizcd work and is largely wasted in the busk
task of digging.
*ridy cxcavadoii implies a Skilled knowledge of this basic uriik. With¬
out tidiness, all is lost. A treacb such as that ingcnuDusly ihustruted in
PI. 1 could nut safely be interpieied or recorded even by an experi-
enoed intcqircter. Tlie proposed curting must be kid out carefully by
pegged Strings at the outset, and must be dearly stnd vertically cur-^s
vertically as the nature of the material will perrmL The general tendentry
is to cue a trench witJi loo mu ch outward slope or 'batter*^ On the sides,
with the doubte result that the scratificarion is dktoned and the cutting
becomes excessively nartow as the digging proceeds downwirds. 'lliis
tendency must be watched constantly by the supervisor and conwed
immediately. If the batter be allowed tti accumulate, the subsequent
cleaning down to & venial] face will almost mevitabiy result in a mixture
of matcTiul from diêrent levels and a ooofusioD Of evideoce+ A ftirthcr
tendency Is to leave blunt angles at the cods and base of a cumng, thus
■gain obscuring the Etntiihcation and ladudng muddle. PrecLsicin must
be tlie order, wid precision and tidiness are synonymous.
Anotlier common error of the unskilled jigger is to init down
too much material ar one tune. Picking 1$ gencnlly easier than shovel-
ling, and the pick-man is Ikhlc lo show on excessive mU. The result b
tliat tlie hoot of ihc oatring is eucumbeted with loose material which b
liable to conceal sTrarificatiun and to get rrodden tnio alien strata.
Usually^ i±iC less, earth boseued Qt any oiw ame the beiiefi keep
149
$TAFP
ilic cutting dear 30 far m tlic e:Qnuo|!{!d prognsis af the wotk wLU
pcnnic.
A lAinilliiiy nf this js; avaid crowding your vvorkmcfir A Such
as tiiiEt illustrated in PL IV a implies cvcr>' son of disDEdei'+ Mtitcrial I&
locvimhly miKcd on tiic fltior nf ihc french, There is a constant risk of
'finds' bang plated in rhe wrong box or baskee, the supervisor cannot
see what is huppoiing, and chaifer amongst the men swells to on uproar.
Noise oa im extsvation generally implies inefficiency; alwi^ys with the
proviso [hut the Orioiial is 3 bom chatrerer and Is unhappy without a
considerable measure of noke abouf him^ In the East^ therefore, noise,
05 distinct from eunoentrared conversarion, is merdy local colour and
may be a symptom of happiness. 1 have stood upon an Eastern site and
marvelled at the mlHCEUaneou^ noise weLElng up aioimd me from men,
women, and childroi all busily and cffeciivcJy engaged upon their tasks.
But noise amongst British uxirlcmcn is a bad sigm Almost every gang of
them conlaius ut least one addJe-paicd gossip, who must be isolated and
extinguished at once, or the good men will rapidly deteriorate. On the
other band, during off-hours, perhaps when sheltering from rain, the
English workman will chancr merrily with the best of them, and a gang
of Welshmea will sing like an angel-choLr, iu such fashion as to make a
virrue of discomfort.
ITie word 'discomfort' raises anorher point. The direemr who looks
after the teosonablc comfort of his men, for example by the proviston of
shelter from ram or liun or by erstuifig an dfidenc waicr-supply—occa¬
sionally by a bucket of hot rc3 on a wet day—is merely doing his duty^
But a word of wifrning k ncctssaxy. An ovex-sulidmus arfention 10
^WTlfare' may easily result in diKatisfactloiL Such h the perverseness of
human nature that excess often begets on uuceusoned dcatre for more; in
Other words, breeds a new kind of discontent. A horse vriU alwnp nose as
of tight a pocket from wbieh ir has received sugar. T rxnslated bato buman
namre, this expectaiicy^ becomes a ffioiU of irtiLatioii, which is dry fuel to
■ny petty spark of discoment which may Itappen to be near* 'The dis-
enneemng result is thut ill will arises out of chaiity ^ and indeed perhaps
the word "chunty' sums up the weiakness at the hajse of the trouble.
Charity has no mevitabk Limii and is thcrefcirc susceptible to esploita-
don. In other wotds, the diaiitably minded director suddenly that
he IS being exploited, and, if remedy be not immediately and tactfully ap^
plied, ihe pnih to hell has been well and truly paved.
Oiie^s workmen, then, must be carefidly teemed, to prevent the bad
tjso
STAFF
fnim comipiinj^ ibc good; rhc fool and the HitifT be relegated
TO jobs where tlicy can mrditaie in «iliiude upon their (or their dirce^
TOf*s) folly; Uifl gai^ miBt be spaced to eiume adetûare I.^beairauin‘,
and a propet art fOf their well-being must not dcgenejaic into an over¬
anxious inlidtydc. To nil ihix must be added an nicn rudiccas on the
part of the director or supervisor to share any spedal hardshipj such OS
workiiig Ja water or mud, or in excessive heat or cold. Furtltcimorci if*
aji on very rare ocasioiw, there is some Slight elcincnt of risk, thot risk
must be sliared ar even wholly undenalreii by the senior staff. Such active
co-operation is esMtiiial and is worth more ihan an inffnrtudc of charit¬
able pellmg, And at any moment of difficulty or fatigue, the casual jest
may be the anodyne. I have seen, towards the end of the dsy, the lines of
young native basket-carriers, upon whose speed aud regularity depends
in great ineasuK the general tempo of an Eastern excavarlont falter and
chatter and play truant in spite of the despairing effcrts of a stroog-
miuded fortoian. Basket-carriers are ncMcr the most responaible lum¬
bers of the patty, and they art necessarily numerous and elusive. A jok
and the orga ntsitinn of a baskct-csHTins’ competitioo wwe the usimedi-
ate solution; the whole gimg leapt into life, and the foreman leaned upon
his stkk. îiiWruryoeBnJtp, and this applies thrOUghDui the world.
Finally, them is another 6)rm of stimulus which has bcoome habiraal
in some countries: the ahocarion of bonuses or fcaArftecs/i funivcnBl
Oriental wotdl) to workmen distingiiiahed by a special aptitude or good
fortune in tbs mcovery of ‘finds’. This system has in die past been nor¬
mal in Grew Hriiain, Egypt, and the Near East, although I understand
thot an attempt » being made lo abolish it in Irntj, ai^ under modem
economic conditions it has practically died out in Britain, So lar as 1
know, it has never been uwd in India. It is an umniiigated cupe to the
director or, rather, the deputy director! « complioiles accounting, it
add? materially to cost, and, unless administered with a rare combma-
rion of honesty and cunning, docs almost as mueh ham as gtwd. It has
been found or thought ncossaty in countries where dealers and coIIk-
rota stand menacingly at the workman’s elbow, ready to tempi with
infiared values. Indeed the problctU is mu an easy one. There can be no
doubt that iu the past much has b«n salved for science inthcEî by ihc
baJbheeih system. Even in Great Britain, objecU, ootahly cains, have
probably been saved in similar fashion from disposal at the locd pnhfic
house (though amongst Britons it is m the 'aationil sporting insdnci’
itiai the bonto is presumed to appeal). But the real answer to oU this
STAFF
is nni 'bribed honesty’j which is whsi the haksheesh system amnunts to,
but adequate supen'ision. After all, the diiector’s best guarantee of the
honcjiy of hii wotkiaeu is, in the words of the French Prehistoric
Sodcry’s Gxcivalioii Manual, 'de nc pas le$ quitter rnic mi nine'* The
hakshetsh system was in fact adopred in the iU-coalrollcd mass-escava"
dons which lune long—far too long—been chanictcristic of Eastern
archaeology, os a^subsritute for adequate Supemsiun. On a properly
aupervised excaviîoc, nearly all the diggers are cunatantly under tlic
eye of the site^pervisur, and theft would is any event be rfiffifidr
Bui I here is a little mure in the matter than that . Apart aliogetber fiont
the question of monetary gain to the workman by the sale of objects of
marketable worth, the archaeologist has to remonber that the requite-
ments of science involve a. complete change of values on the part of the
pcasaur, and the bouua-sj-arfan, wisely applied, undoubtedly helps to
impnas the new scale upon hU niioi Wlicn the workman «,
realize that what is la him a mere ftagile crock or a fragment Of carved
bone and nothing more may, to his supervisor, be an historical 'documen t*
of high itnponance, desen’ing of a moaeiaiy bnnus, he is on the way to
a menta! readjustment which may ultimately turn him into something
mure than a mere dirt-shifier. Nevertheless, my Indian expericnceshDwa
me that, by combined example and occasional conanientary, the intclli-
pni Siiperviior can achieve this end withourthe emborrassments entailed
in For enacipte. a wt)rkaun finds an old Iiotsc-shoc; he i*
told that It was made a,ooo years ago. He in turn itmarks uport its
general similarity to the horse-shoes which his unde, the lool black¬
smith, aiakes, but proceeds to observe minor differences. The brief dus-
cussion—it must nut be protracted, or all work round about ceases—
leads to other hisioriol or wchaeological points, and the man bKomes
intelligendy interested in his work. But let this discussion be of a casual
and particular nature. Don't assume that your workmen am all breath¬
lessly awaiting general instruction in archaeology. I well rEinember an
earnest antiquary inHicring an eloqucni address upon the wnrfcmen at an
KCBvanon (during the luwh-hour) on ‘prehistoric manV and ramitlg at
last, well pjea^ with his effort, to a patient member nf his audience
with the question: 'And now, my man, what intetesu _vmf most in your
work here?’—10 which the rmwy, tlowly remaving hia pipe and medi¬
tatively spitting, replied gruffly: 'The fire-o'clock whuitic.'
xn
Tools
IKADKQUATi toob mt oot 10 CKCUSc far hild work* No rcspoosibk
arcbacDLogbt will undendke an occavarion witlwutpdoquaic suff and
equipmcDt save in i imft cin«]g^cy. Tlie ci|njp4|imt falb into tvm
main cafcgprics: that of UiC dirccriog sniff, and that of the Jabaurov-
(o) EQUIPMENT Qt TKK DlllECTlN<* STAFF

t. Theodolite or idoiplihcd equivalent (ice p> 141)-
1, rlAnc-Lablc,
% RcitifOfCffd measuring tapes too feet (ar metrio equivalent) long-
4. 2-Ibol and 5-fbcK folding rîkh or their metric equivalent.
y Prunuric oil-corflpaw.
4 Drnwfi^ff-bcmrdiiH including several light boards of i-plf wood for
worit on the Hilt.
7- plumb-bobs.
8. BubhJe-lcvdj.
9. Drawing paper, inmc of it squafKl <e-g. in i-incb squiirea with
eight subdivisions)*
to. Archhectural scales,
tt. Girtd penile, and soft ertsers.
r?. Bfoad-bladed fcnivef (Wade about 7 inches kang) Hud/nr poinied
masont’ trowels.
13- Good string*
T4. tfidian ink) Pols’! and paim-brushes*
15 *Ctrcular celluloid protraemn.
16* Large cellumid icc-sqUArei*
17, T-sqiuir«.
ig^ Good diawhif-pinSi
19* l^per-clips-
aok hmalL pfly-envelDpej (for eouti« fe-)-
3i« Small rieôn labelsn
.
22 Notebooka.
23. 3-inch sndi 6-inj^ siiISh
54. Scales of various Irindi for photography, Slz.
Some of Ibe above, »iich BS ihe thnidolite, the compais, and the
majur diawiog'insrnimeBiSi will be jioiwl centrally. But laaiiy of them
mtut be 10 the poisosioii of every ihc-iupervisOT, who should keep
155
Toots
them in a tin box labelled with the cnde-iuijic of the ana for which
he ii reapoiHfble and pmfciably also wiUi his own name. This bo*
will enntain a Jnmiiriii(f’tapc. 3-fooi rule (or equivalent), plumb-boh,
bubblc-le^'cl, pencil, eraser, string, ink, pens, paiot-bnishcs, paper-dips,
envclapea, labels, nails and, not Jeast, notebook. The knife ot irowd
should accBiapiiay the supervisor everywhere, as an indispensable and
inseparable lustruiaeni. Indeed, it is altnost a badge of rank; without it,
the supet™r CM scarcely begin upon his ta^. Its uses in the detailed
cjtaniiriarinn of A Section arc almost infinite. It is used, for example, for
deaning and cbecking difficult sections, and for testing by pressure.
Tod', or sound subtle difictences in the soil. It is esseniiaj in the final
prepaiation of ohnosi every subjea for photography. It is a useful
nutrker in survey. It has a hundred uses and should be n treasured per¬
sonal possession.
G&) THE laboubehs’ equipment

These wiD vary in detail with the locality, and the following is a
gcncTaliEed list.
I. Pick].
X. pidts ur trcnchifig'tools.
3. Lujgie ihovelft.
4, Small shdvds or scoops.
5f Spade,
4, Tuif-cutrer or triminer or edging-knife.
7, Baskets or pans (in the East, for the removal of soil).
8. Wheel-bomnra (in the Wesi).
Knives or trowels.
to. Planks.
II- £jow-bar.
I a, SlDdge-hamioer (particularly for driving in fencing-posts).
Ctf these, the pick is the primaty insmiment in excavation, not merely
for the genera] loosening of the ground, but also, properly rnottyillQl by
a good workman, for carnpuratively delicate work, for which its weight
givo It a relatively cffonlcss control. The tool is easily misused, and the
pickman requires spcdal wntching and traUung. For Bcaipplf, the leti-
dency to use the broad end must often be checked—the pointed end doa
less accidental damage. And it is ftequendy dcsiiable 10 avoid luunincr-
ing cicry scrap of c^h with the pick, but reibcr m use the instrument
os a wedge with which to lever off considemblc lumps of esrth, driving
^54
TOOLS
in [he poinf well back from the oining-facc fur ibis purpose. Above ill,
except in strata known lo be of considerable i1ep±, prevent ihe pickman
from driving the pick too deeply into ihc soil. ‘Wholesale' di^g ob¬
scures the evidence. Lastly, having IcHseocd a icasonable quaniiiy of
earth, the pidunan should stand aside iind leave the spot free fur deu-
aiicc,observ3itiuD,and detailed work. A good pickinat] will with ibe
end of his pick, observing not njcidy with his c>e# bur also by slight
dilfcKnccs in the 'feel' of the canh. The pick should be kept sharp.
I'hc ttnall pick or jreiiciiing-toot is essentially the diggbg tool Of ihe
foreman or supervisor, or of b spedoUy experienced workman. The
lighmcsa of this insmimcni makes it parriCMtarly sensitive to slighi
changes of soil or even of sound—-for example in working towards a
mud-bride wall CP- 84). With the knift, it is useful for disenpgiiiB
objects from the soil. I t is the normal iostnunnu for the digging of the
control-pit CP- supervisor i» working In uiiknnwii
material. A word of warning: i laey workman (id the East) will always
try to exchange his large pick £91 OOe of the small odes, so that he may
squat and peck idly at the surface with a Riinimum of etfon. The super¬
visor will soon Icam 10 deicci the dilfercncc between the esapist and the
honest man who is ushig the small pick for a good and intcUigtui reason.
The or trimmer or edging-knife is dn essenrial instnunent.
Its feature is a sharp crescentic blade in the same plane »the handle. It
is used for the veiy impoitunt tdSk of iriramh^ the sido of cutimgs to a
dean, vertical face, without which proper exantuMtion and record are
diflkull or impossible. It should nm, however, be used fur substantive
digging, partly because it is not strong enough for the purpose but partly
also because of necessity it cuts dean through any fragile object that
comes its way. Upon its proper u-sc in suitable soils depoub 01 no small
nteasuxe the cleanliness and effectiveness of an ex^vation tu derail.
The basket or iron pim Is u-wd on Eastern sins (a) for the r^v-al of
spoil-eaiih, and (fc), when properly labelled with a tic-on label indicaung
site and strorum, for coHecting pottery in the held. The hfe of a basket
can be enendcd by rehiforcing with wi«. For ^ itmovnl of dus^ sad,
the inside can be coated with mud or lined with n rag. It should uia-
denially be a matter of routine to ensure rhai dd clod of unbroken (LjC.
unexamined) atth is ever inclutled with ihi dust.
In PI, XVII is illustrated a fairly complete set of tods from an IndioA
cxcavaiinn. The only nonble absentee are a sharp-edged spade and
shears or sossors. The objeras shown srot
I5S
TOOLS
1+ Ik and small pIckA.
3. Small pick wftb obc^vcl-cotL
4. Hond-sJwvcL
5. Sbavd.
6. Tmf-cmn*r dr triranwr (■ much worn eJCiimpJc), very uiwful for
Tnininiiig down lecrians.
7. Basket for removing wth, ufimlly carried on ite head.
8. Supcrviw's knife.
9. 2^fODt«Cft]c,
10. 4-foot pnier {Moot imd polea should be added).
IX. tjraduamd triangle^ with bubble^levels alExcd. For uiOj see p, 70.
la, 2-footnile*
ijk lOChfwjt reinforced tape.
14. Nmeboolt with aJccmatc pages squared (i-inch sqnareti subdivided
Into iîneb squAres),
15* BuBbic-Jcvel.
16. Pinmb-bob.
17. Brushes of vaiiouj sizaes and shapes. Bmcrais, hard and sofi. ahould
be added.
156
PI.ATfi XVIr
Selection of |3doI% from in lUvtcrn eMCivitkm.

{Sm p- #5l J
xtn
The Potlery-Shed'
E xcavations inay be ncpcctal to produce umteridJ which re¬
quires IP be colieetedi lahellcd, stored, aod cveutUiilly packed and
tr*LDspoited from the îtc. However abumdant OT rcstriCTed this
material uwy bCj commcnjiiicaic ariBiigercetiU must be made m ensure
that it will reach its dcstmatiaii adequately Jabelied^ and m treated and
packed that it arrives in at least ns good a conditioP as that i n which it w-hs
found These ruifes are daigned as an for pnticry- or
snuUl-finds assisrants (see above, pp, 139 student-
ext^vatars who may he expected lO assist in rhe work of hdndllag the
material. It is assumed throughout this chapter that ihe poticry-^sistanC
is in stiprerf>e charge of the staff detailed for the reception, pftliimn4iry
classihcarian, and dispatch of all ^hnds^
THE ^FINUS^
'fhe rnaterial found will fall roughly into two krge classes:
(i) Poncry and bone3> which usually form the bulk of the maieml and
are treated mainly us a seiieii Of groups of material rather than on an in¬
dividual basis.
(li) Other muteriulj which is created on an mdividuil basis and may be
included^ whatever its SUEtj under the general LOdHi "stoollîuids »
ACCOM U on AT I ON
It is obvious that the ammini of stDragc-Space and working ipaoe re^
quired wiU be related directly to the type ofexca^TiTinn and its locatioiL.
For the purposes of the present description, it is proposed to take a
moderaie-sized csoivation on a site producing a fair amount of la
a country with a rainy dimaic; hut the same basic principica ipply to all
dcavatioAS^ wharcs'er tJwir locadoo and problcnu
The difceior (or deputy director) of ike cspcditiDn will have made
uTungements for the ucecMry accommodation for hundlmg and stor¬
age. The potrciy-aKsistant will, h^mti be cipcacd to utUilê these
* t gladly ackiwjwleiljc the heJp of M™. M. Aylwin Coftmij F.S.Aî

m the pfapiiixion of thji dupter.
J57
THE POTTERY-SHED
facUities as advantageously as possible, and there is commonly much

scope for his initiative in this matter. He (or she) should aim at organiz-
ing:
(a) A pottery-shed in which all the bulky material can be handled and
stored.
{b) An office in which the ‘smaU-finds’ can be recorded and stored.
If pottery or objects are to be drawn on the site, a quiet drawing-
office IS also desirable, and a small Beld-laboratory will be necessary
(p. 169). Locks should be fitted to all these rooms or sheds.
HOURS OF WORK
It is usual for the finds to be transferred from the excavations to the

pottery-shed at the end of each working-session. Whatever arrange¬
ments we made for working-sessions on the ‘dig’ itself, the pottery-shed
should be open for a reasonable time after the end of each working ses¬
sion to receive and stack ffie incoming material. As site-supervisors may
need at any time to examine the stored material or to obtain supplies or
mys, the pottery-shed should be open as far as possible throughout the
y. As the shed should not be left unsupervised at any time when it is
open, the assistant in charge should arrange to ‘stagger’ the hours of his
wor ers, or, single-handed, to arrange for a volimteer to take charge
during absence. ^
SITE AND WATER-SUPPLY
The desirable a^butes of a pottery-shed are that it shaU be spacious

an g t an equipped either with running water or within easy reach
of a water-supply. It should be as close as possible to the areas under
exarauon, and should have a clear space outside for pottery-washing
and J^g m fine weather. In wet weather, if sufficiently spacious, it can
be used for washing and drying pottery, but, if space is restricted,
the pottery-assistant should try to arrange that shelter elsewhere can be
used where volunteers can work on the material. A tent or a marquee is
M iMdeq^te subsutute for a pottery-shed. A bam, large shed, Nissen
hut, empty ^ge, or bare room serves weU. If running water is not at
Wf^h immediate vidmty, a labour-saving device, well
wonh consideration, is to run a long hose from the source of supply to
" building. If water has to be
earned from a distance, arrangements for local storage are even more
necessary.
158
THE POTTEftY-SHED
EQllJPMEMt
The mimmum finTiinirc required fur corofortaibSc wcirfcin^ in a
pottery‘Shed is one hrge trcsdc-tabLc and a bench ot s«ts- Boies, rang-
ii:^ id size from ica-chcsts to nmteh-boxes^ itre of the fint importiuiec.
Large sroragc-boics can be med boLti fur ^lOnage and for the packiog and
the cransporr of the heavier finds but, os these are nor oequirod ncwadiiys
easily or chcaplyi smaller light wooden entes and strung cardboard
boxes should also be Dollecied. Wooden fhiiE-boMs arc very useful for
storage and for the transport of lighter nuiteriabi and will carry pottery
if it w not TOO htaivy. They can often be obtained in the local market
from a ^lut-mcrchani or ftont the head poner at the goods entrance of
a laip: store. They con be slfl,cked in rhe poticrj^-shcd with their open
sides outwards to form naefe for the Temporary storage of msrerialp
Smaller cardboard boxes and tinSj of all shapes Uf Etze$i are usefid for
packing scpuiiite puls, small-finds, and carth-san3pl»r
StatiuDcry supplies retiuired are;
Paper bags of varying sizes.
Foy-envclopes.
Tic-on labels and small whlrc tags.
Paper-elips and balls of string and rwinc.
Indcx-carda and guide-cards^
Feus, mapping^peosi pendli, ehalks> and small paint-brushes.
Black waterproof Indkii, red| and wliitc ulks.
Blotting-poper-
Porrery-washing equipnetrt conswes of;

Trays of various sizes or drying imr.
Washing-bowb.
Nail-brushM, soft paiui- or poTTery^bniihcai aodold tooth-bftishci.
POTTkHy WASHING
'Finds* from tlw eXiCaviTiona arc rtcaêd dt the poffcry-shed in tmySi
baHkcis, or bags. Each reeqTtHclc will have with ii rwo labek recuidi^g
Its provenance^ OJtc loose and one tied oO to the receptacle. From the
time of its reception until it is dispatched, flH or any part of the content*
of these receptacles mu5t always have mi accompriyiiig label giving this
infomuiLion. In SO far a* ^ pmciitablo, the maferial received at the end
of cacli working-session should be worked over during the succeeding
session. If this ojuioi be done, it should be bagged temporudly and
159
THE FOTTEItV-SHED
Sf acKed in order of date received. A wet day, v^-ith extra aisiiitaiict, ttuy
provide on oppominiiy for deariog up arrears, the matcriaj being dolt
witli far as possible in the order of receipt.
The pottoy-assisrant will be espected to instruct volunieers in the
details of the handling of the inatenaJ and wiU be rcspniwihle for seeing
rhat they give it proper ireutmcnt and that it is coneccly labelled and
packed. Although die site-supervisors may have handed in their mors
important snutbfinds separarcly it will be found ihar the so-called pot¬
tery-trays often contain much that is not potreiy. On a Roman site a
tmy may well contain a f«v |jone$, oysicr-shclJs, iron slag, brick or tile,
tesserae, and perhaps fragments of pointed plaster. A preliminary «-
amiruition should be mode, thtrefore, to see whether there is iiny material
that should either nut ^ washed at all or should be washed with special
cart. In both cojes it is advisable to isolate this matcHol on sepante
^ys, of course with duplicate labels, Soft or fragUc pottciy should be
dried out thoroughly under cover before being washed. If it is still un¬
suitable for washing, ir can he brushed vsdth a soft brush and then im-
preguared with bcdaoyl in toluol id consolidate it, or, if bedaciyl is
uiwbmnahle, a to per cent, celluloid soludoa can be anhsrinned.
Each washer, starring with a tray of unwashed ponery and ocher
wis^ble material, requires a dean drying nay or mar, a bond with
wnshmg water, and suitable brushes. Tlie first step is to place one of the
two labels in die drying tray. This Is esMiuiaJ as it ensures that if the
imher IS called away and a panly finished tray is moved, the material
still retains ns jiicnriTy. Unmarked material uolamd from its label can¬
not be used as dating evidence, and the carelessness of a wasiier in this
xesi^ can destroy the work of a careful catavaior. The material is tlien
washed piece by piece and placed on the dean tray to dry. It is oot a
good thing to empty a niimbcr of sherds iiiro rhe wasliing bowl. Tliey
may not benefit fram soaking b (he water, they may sink bio the muddy
sedunent which soon coUects b the bowl and get thrown away when the
water a changed and, b any case, the washing bowj has no lubd! Hurd
whttl-ninwd ^ wdl-haked coarec pottery can b< scrubbed with a uuil-
pottery fe.g. SamJiio),
md badly hiked pottery should be washed with a soft pottcry-brush.
P^cd ponery, painted plaster. &c., should not be vrashed until the
dUôr or rtesitîupcrvitors have been oonsulted- Tlic edges of sherds
shodd be hmshedas wdlas the bsldes and cutsides. ll will save rime in
matking and sottmg trays If the maierial is placed on the drying tray b
j6o
THE POTTERT'SHSC
an urduly ituuiiier. AlJ riins, bases, flud decomed sherds should be

placed in the top half, plain sherds at the bottom ami sundry objecia
such as shells, nails tils Ak-, piled ia sepamte small heaps, l-inalty the
washer odds to the finished dîng tray the accoad label faMD ihc nnvr
empty ‘unwished’ tiay. Unless dw supply of water is lesmcied, the
water Ln the bowl should be changed ftequenUy, (Pottery waah^ in
muddy water retains a ihin coating of mud when dry ond may require to
be iewashed.)'rhe drying tray is then placed in the open or under shdtcr
.. its contenia are dry, cate being lafeoi to anchor its lahcl hntJy.
Wflshets should be encouraged to take an interest in the nuiictial they
hatwiv and in its provenance; their irbcnuity of where a parriculor pot can
be found has often proved valuable when it has been required quickly
for impection. .
Trays of dried maienul should be shown 10 the she-superviSOTS so that
they may note the enutents in their field notebooks and so that they can
reject any material tliat is not required fiirtber* Pottery that has nut
leicted odcquately to simple wasiuiig to water can at this stage be
segregated and icireaied by further washing in tcepol (a detergent) and
water, or, if caaicd in lime which ii is desired to remove, crated by soak¬
ing in a dilute solution of nitric add (looraopcr cent, solution) followed
by several rinsings in clean water. Fotteiy in add BUist be kept under
observation; Dfieu 10 minuTcs' trentmeat or until cffert'acencc stops, is
all that is required. Painted sherds should only be dipped i n the sulution,
and a comer should alwajn be tested first to bbc if tlie colour « fugitive.
POTTERYMSRKIkg
The trays sit now ready for marldng. As a genewl rule all rims, b«ts,
and decorated sherds are marked. The site-supovisors should be con¬
sulted to ssicrtaia whether they require that all sherds af s particular
group should be marked. The aim in marking is m gbfu each marked
piocc its piuvcffiince in a dearly legible but inOMBpiciMus position so
that it can be handled with safety when removed from its label. Rims and
sherds are marked on the inside, haso on the underside. For aH sherds
on which it shows clearly, bJaCk tvaierproofink » used; black and dait-
gicy sherds can be marked with white ink. Mapping-pens are used, and,
if they art kqn clan by wiping, washing, or scraping afie use, will write
more legibly and will last longer- Aiarkcis should remember that shads
may be required for display and that it should be possible to lay themout
right side up so that the markiBg does not Show. Some wares, especially
»T1 l6l
THE POTTERY-SHED
Samian, arc easily marked on the line of fracture. The exact lettering to
be used will be indicated on the labels. As a rule, it consists of a letter
or abbreviation to indicate the location of the excavations; a letter or
munber to indicate the site; the area, square, or trench involved; and
the number and perhaps the orientation of the actual layer. Convention¬
ally this last is encircled: e.g. VER/G M II fuTT) means Verulamium,
Site G, Area M II, Layer 14 North. ‘Group* material does not need any
closer identification. If, however, material has been pin-pointed three-
dimensionally (p. 69) and given a serial number in the field notebook
(a number which is conventionally included in a triangle, e.g. this
may be added. Porous pottery does not take ink readily. The area to be
marked should first be varnished and then marked over the dry varnish.
SORTING AND BAGGING
To bundle into a bag the contents of a mixed tray of material is de¬

leterious to the material itself and throws extra work on to the experts who
handle the material later. Well-washed pottery if mixed with oyster
shells comes out of the bag with a film of shell fragments; large pieces of
tile and brick may bruise or break the finer sherds; and iron slag and
nails when mixed with the pottery are a constant nuisance. Fresh dupli¬
cate labels should be written for each type of material other than pottery
in the tray, and these collections with their labels should be placed in
separate bags and relegated for further treatment. When the material is
reduced to a group of clean, marked pottery only, it is ready for bagging.
The site-su|>ervisor should be consulted as to whether special wares or
types should be subdivided and bagged separately. If so, further labels
will have to be wntten. The final group of pottery is then placed in a
strong paper-bag, or, if bulky, in a double bag, together with one of its
two labels. The other label is threaded on a suitable length of string.
Pottery bags should be treated as small parcels and should be
both along their length and their widths the string being tied
inTsingle reef-knot or a bow and not in a multiple knot. Knots are a
menace to anyone who may have to open hundreds of bags. Whole pots,
or a collection of sherds which may join to form part of a pot, be
bagged or boxed separately. The pottery bags are now ready for storage.
STORAGE OF POTTERY, ETC.
Pottery bags should be sorted into groups from each site, and from
each area or trench in that site. They should be stored so that all the bags
162
THE POTTERY-SUED
froDi any particular area or level can be [produced quickly for itiRpccriim.
Siorage-boMH should be labelled dearly with the iwrae of the cxeava-
don, the site number or letter, die treuch or area number or Icnn, and
the type of object stored lu them. If piore than one bos i* required for an
area or rmneh, boxes should be munbeted and kept together. They should
be packed firmly, but only filled level with the top so that they can be
either fitted with lids or slats or pded on top of cadi other to save space
in rran-Hport. It is usually convenienr to store all bulky material in the
poiTcry-shed, even tlwugh it may indude sherds which have b«Eii
indoed separately as 'small-findj’H The boxes used B>r non-ceraniic
material should be Clearly labelled for caeh site te.g. 'bones', ‘iron dag',
'architecture’, 'catth’samplea', ‘charcaal*, 'shells^,
PACKING
At the dose of the acesvariou-scason, the material will prababty Lave

to be packed for removal fmm the site. If it is to be senr by rail or by sea,
urrauKentents must be made to pack the boses securely, to dose them
with lids or slats and to nail on addressed label*. Wliolc pots should be
filled with wood-shavines or otlier packine-maicrial. The wedgLog
material in the pacfcinE-cases siwuld be rammed in with a wooden ram¬
mer to moke it so compact that the pots will nut work loose and break in
transit. Sawdust is not sarisfoctory as a wedging material. A list of the
contents of each case should be included in ir and a duplicate list be sent
by aidl KtKtii ibc catcavatioa^n^Jtra,
POTTERY
If pottery is lo be mended on the sire, the poticry'Shed should be equip¬
ped with boxesofsand, plasticine, and {on of cellulose dope, amyl-acetatc
and acetone^ Methods arc a matter of JanuusttatiOD rather than des¬
cription.
inuexujg 'small-Finns’
All material treated Individually os a ‘smiill-find’ wiD be received by
the small-lind recorder (p. 139) clihcr direct from the site>Supeivisor
or u baps DfiiuilEfu] sotted out fiom the ponery trays dud tiaiisfeTred
from the poticiy-shed. Whether cu: no this maTcrial is to be cleaned and
marked ca the site, it ahould always be indcred. 'Small-find' mawrial
usually includes the more dosdy datable evidence aud, in addition,
tends to be bagile and to require extra care in handling and ireatmeni.
163
THE POTTERY-SHED
The record falls into two parts: (a) the accession-registers, (b) the
card-indexes.
The most convenient system ofregistration is to keep a small accession-
book for each site (i.e. for each main subdivision of the excavation). Its
pages are ruled in columns for entering the accession number, type of
object, provenance, finder, date and amount of bonus paid (if any, see
p. 151). As each ‘find* from a site is received, it is given a serial accession-
number which is marked in red ink on its bag and/or labels. The appro¬
priate details are added: e.g. G. 49.123, Bronze brooch, MII ® Dark
Earth, John Smith, 1.8.49, 6d.
The card index is then prepared in duplicate, for arrangement on the
one hand by categories and on the other hand by sites. On the first or
‘objea’ card, the type of object is printed at the top; the accession num¬
ber is wntten in red ink on the top right-hand comer; then the pro¬
venance and any other useful details follow; and finally a sketch or small
photograph is added (if necessary on the back of the card) so that the
object can be identified if at any time it should become detached from its
label. The second or ‘site’ card is a duplicate of the first except that the
provenance of the object is printed on the top of the card and the type of
object below. Two boxes are provided to hold the cards: one for the
‘Object Index’ and one for the ‘Site Index’. The object cards are filed
in the Object Index behind guide cards for each t5rpe of object. Objects
made of the same kind of material should be filed next to each other
(e.g. Bronze, Bronze Bracelets, Bronze Brooches). The guide cards
should be arranged in alphabetical order. In the Site Index, guide cards
are made out for each area or trench (e.g. VER/G M II), and behind
these all cards belonging to that area are filled in order of successive
layers. The groups of cards for each site should be filed separately so
that, for example, all coins from Site A can be looked up easily and are
not intermingled with those from Site B. In excavations in which the
material has to be left behind, or in which a record of the finds has to be
furnished to a local authority, a single card index is inadequate. Arrange¬
ments may have to be made to compile a duplicate catalogue in addition.
If so, when photographic assistance is available on the site, one drawing
of the object may be made, and as many prints of it made as are required.
These can then be pasted on to the index cards and in the catalogue in
the appropriate places. Often, in this system, type-forms of pottery are
drawn and should be treated on a routine ‘small-find’ basis.
164
THE F0TTERy*SHKI3
IND1X1NC COTW^
CoiojK should be identified tmd fuUy Indexed in hath, ûbjcct’ and

iixdexes^ by naaou of tbcLr imporuiiKic as daic-mditaiurt. Beside ihe
record of provenance, tlic cards Sbotild contain the fuUowiag mforma-
tiont
I, The obvcHM: legend aad qT». Breaka in the Jegend are indicated by
a h illegible leitm are endoscd in squan: brackets [ ],
2h I'be luvcnie Jegend and typen
5h The mint-diaik^ If presentk
4h The coudidoa of the coin» whether it is in mint cnTidition, good
coi^tioiii wuTtij^ ooriodeds or illegibic.
5^ Jf no refcreoce book® ara avâilablci the date of the emperor^s
reign ahodd be added. If reference books are avatlahle, the min should
be looked up ID tlie$e and ivt dEnoinination 5C5ieiTius,aiiromjDttnuii
and a mote preciie dating added.
Rcfotnces for Roman coina down to A.d. 300 can at present be quoted
from Mattingly and Sydenham’s Imperial Rmam Caiman later coina
fom Cohen’s Mantmes frapph:^ sms Fempm nsmain. A specimen card
might rad a$ fcllr>ws:
DUPONDlUSfifDOMlTiAN VliR'G 49 m I
M It @ Dark JEnfili /ti, KE S 4^ i'-a' |
Obu. IMP CABS nOMlT AVO GERM COS XII [CBm PER PP]
head L,
Rgt>. MOMETA/AyO tUSTlI S C Mneti Mg. I.. Iwlilliig
«ch1o and eornuroplacn
Wcmi
M. 3B3 AJI.M
BOrJgS AND SOILS

SanurwHctc in ihis boo^, and here as wcU a$ anywhere, a plea roust
be put in for a mo« sdinlarly tMUecUon of hinlogicil and bounical
inaiciml than evCD our bener excavators haw normally achieved in the
pasL For example, animal bones have, St it true, long been the tubjeCT of
special sections in archaoDlogiHil fcpottSi bur few, voy lew, of those
reporu give lu the analyrical details which, in the present state of know¬
ledge, are uigently required. What usually happais b this: a few bundles
165
THE POTTERY-SHED
of selected bones are sent off to some complaisant biologist, who sub¬
sequently reports the presence of Bos longifronsy Ovis ones Studeriy
Equus agilisy and Canis familiaris lacustrisy and receives a fulsome tribute
therefor in the pubhshed report. Let it be said at once that this sort of
thing nowadays gets us almost nowhere. Bones are documents as are pot¬
sherds and demand the same scrupulous attention both on the site, in
the small-finds shed, and in the laboratory.’Consider the information
which we may expect from them. If in sufficient quantity (as they not
infrequendy are) they can give us from phase to phase, not merely a list
of the fauna category by category, but also—what is far more important
—a hint as to the economic function of that fauna. To what extent do the
bones represent food? To what extent do they indicate sheep or catde
killed young, before the winter? Alternatively, what proportion of the
sheep or catde represented were suffidendy aged to prove organized
winter-feeding? Were sheep thus maintained more extensively than
catde, implying a dominant importance of wool? Were horses eaten?
ridden? driven? (Size and age are here amongst the diagnostic factors.)
Were pigs kept in suffident numbers to imply appredable exploitation
of marginal forest-land? Did the proportion of one category of stock to
another vary in the course of the occupation of a site? These are some
of the questions on which excavator and biologist must confer after
examination of all significant bones from a site in their stratigraphical
contexts. How often is this done? Of course the quantitative analysis of
animal-bones has in it an element of unreality, since a single skeleton is
frequendy multiplied arbitrarily in its fragments. The same objection
applies to the quantitative analysis of potsherds. Nevertheless, exer-
dsed with common sense, the counting of bones, as of sherds, may, un¬
der normally favourable conditions, be expected to yield usable statistics,
and must be attempted.
Similarly with the collection of soil-samples. The preservative capa-
dty of damp soils is truly astonishing. Seeds, leaves, plants, wing-cases
of beedes, may be incorporaj^d in a condition that renders recognition
easy. At Stanwick in 1951 we* found at the bottom of a rock-cut ditch
of the first century a.d. a layer of wet clay representing a pool of water
which stood there from the earhest days of the cutting. In it, amongst
other organic remains, was a puff-ball which the British Museum
(Natural History) had no difficulty in recognizing and naming at once
after nineteen centuries in the soil. A word of warning may be added:
some seeds may be less durable than others, and, whilst positive
166
THE POTTERY-SHED
evidence tjf on andept vegfradon is Micuifkctoi)^ oiou^h^ ju^jadve
evidence nniy it quilt ftirtlitr ihoughr and consulution.^
Finally, wliiisL OP Uifi subject of jilant-secds it may be added ihar an
impitssive number of ceteaJ-idcnrificatiDiia has been rendered iKKsihlc
by impitssions of grains on hand-made pottery. Sherds of Unte inmnaic
aignificmice Piay in this fashion achieve an amdePtai imponance which
iustifics and indeed neccssiiat-ra a carerd eatamination of es'cry stcnipj
however featureless othcrwiscj from the moment of discovery onwuids,
VAfllANT LAYOUT FOR DJtY CLIMATES
Where rdn is mj a facicr add work cian be earned out for weeks op
cad in the open, ihc tasik of the poLtery-nssistaa: can be simplified by the
use of a grid outside the potteiy-sh^.
A reasomble area adioiiilpg the pottery-shed is otrefully levdJcd and
smoothed* and a rectangular fiaoieworls jc >'arcis (bay, 20 yards) square
is niarked our by ocat lines of stones. Within the frame* subsidiary cam-
paitmcnts are formed by similar lines of stones laid down in boih direc¬
tions at r-yard intervals. These suhsidiary^ squaics art then demarcated
by fiat wooden labels driven in along the niaigins of the frmicwork and
bearing in one dkoedon the dranmioationa of the various cxcavaLed
sites (squarc-numbef or tiendi-niunbcr) and in the nr her direction the
successive numbers of the layers (i, 2, 3, &c.)r iTius every aubsidiiiry
square is identihjcd with a hori^ntal and a vertical datum on the exca-
^'aicd site. Tn the illustration (PI- XVIII), taken from the ArikainedU
exotv-arions of 1945, the line of labels across the forcgrouiid leprscots
The sire (At. Ai* A3* ^c.) and the line receding fmm the camera repfc-
aenrs ihe layer or Stratum- *
The □pcuiiioa of the grid is as follow^. As The ba5kctH::arrkr$ bring in
the labelled boskets of ponciy at intervals during the day* the poiiery-
Bsaistmt empties the conrents* with their two latoeb* earcMly on 10 ihe
appiopriuie square. In rhe illustration* he is jctualJy emptying a basket
en to the square rqmciieniii^g straniTn 8 an site A4- Tîs u imtit
drfcfuWi it must be remembered thai ar this stage the indisnduaJ sherds
ate not marked, and any careless spilling wtU irreparably rramfer a
sbeid to its WTong stratigmphical positioiL After being emptied, the
basket* po longer labeUed* tS sent back to the excavation for rc-usc.
The next stage is to inspect the sherds thus lumped and to transmit
* Far on csccikm sccoimi of the appUciticipj df pottcn-aiuljnJi, lee J. O. I>.
Clark, The Mfaoihhtt: Sûiernm e/ £luri?j« (OntbriJRtt i93*!>. FP- 3* ff-
167
TH8 POTTBIIY-SKED
aO that do not tequiic Epcda] trunneDr ti> the washosj wba in the Qlus-
tratioa ate seen sqiutang under the baayau-iiK at the back. The con-
tCDls of oFKiquart only aie vruhed by each washer or group nf washers at
any one limc, othciwue admuentre is inevitable. When dry, the washed
sherds arc transrerred froup by group to the tables where ihey are fur*
ther examined by the potrety*a$^tB4il (and as often as possible by the
dirtcior) and are marked CBicfuUy, sherd by sherd, by the market under
the pottery-assiatani's direetiOD, in accordance with the rwo JabeJs which
still, of DDuiae, accompany them. Finally, the marked sherds are bagged,
group 1^ group, with one label in the bag and rhe other ded on outside,
in accordance with the procedure alrody described.
t«£
PtAicty-gtid, ^ use in dry clEnuiei.
{Sm I*. JS?^
PLATE XVMI
XIV
The Field-Laboratoiy
archsLCQtu^cul difmLst mmt be avaiJAble to mf excovation where
fruible ur pcmhAble maietidji are likely to foaod^ nod in most
of the world this means that a smoU Md-Iaboratory Is an
iniegrul part of the outfiL li lias already been offinned (above, p. 1^44)
that ilic primary funedons of the field-chmist are lo help in iheTCinm'al
offtake objects from tbe ioD and in tbeir subseqycnT mnsportadan;
to aittSE the decay or distortion of objects on Kcpoanre; and to dean
objects, notably coins, whidi mtist be Idennfted aa the cacavaUDii pm-
cceds. He must abo, as a gidde to Jmer marmccitj keep a log-book of all
first-aid administered by him, objecr by objea:.
The chemist doe$ aoi require extravagant acuommodadon, but be
must have a pleodM supply of ™/cr. In d&ert or semi'dewrt
regions, tkb is not easy, lb rest for aaiinity, a silver nitrate solution is
recommendedH The solunon is prepared by dissolving 5 gm. of sflvcx
nitrate crystals in 500 cx. of diirriilcd watcr^ followed by loc.c. of strong
nitric addp The test is carried out thus: Two dean test-nibcs of the same
siae ore holf-Mled) one wirh pure water imm the stpedidon^s reserve
and the other with the local water. To each of these tubes are added ten
drops of the $Ilver niEnite jjalmiDn, and die tubes arc shaken^ A white
preci pitate or milkincss « produced whkli ™ie* with the amount of the
Chlorides C^hida ore invariably prcseui with Other salts) inrhc water. The
miikiness produced in the pmre water will be relatively slight.
The same test can be applied to water in which salt-imjnegnaicd
objects^ such as poneiyi have been washed, and rbe washing should be
continued until the wash-water shows no more reaction than the pure
water.
Distiikd miter is also essential, t<^ctber with the Mowing chemicals,
&C.:*
Nitric add (or hydrochloric add if nitric ts ufiobtainablc).
Accione.
Arnybaceuife.
^ Mii^? Icme Oedye, who is in charge cf the labctiiofy of the ArdiardogiMl
liutimie of the Um^tnity of Iôdoor hu kindly pidpared the Uii on p. 170^
nrnl Dr. H, J* Pkodcikiih hia been ftocifd enoyjh 10 mil ihi: dinFHer^
16^
THE FIELD-LABORATORY
Silver nitrate.
Citric acid.
Sulphuric acid.
Acetic acid.
Ammonia.
Caustic soda.
Celluloid cuttings.
Shellac.
Bedacryl 122 x.
Toluol.
Teepol or similar detergent.
Sodium sesquicarbonate.
Polyvinyl acetate.
Methylated spirit or alcohol.
Plaster of Paris.
Granulated zinc.
Graphite slab.
Copper wire.
Copper and brass rods.
Batteries or transformer.
Glass or pottery tank.
Glass or porcelain dishes and beakers.
Saucepans.
Measuring glass.
Test-tubes.
Glass bottles.
Spoons.
Penknives.
Wire brushes (steel and brass).
Nail-brushes.
Tooth-brushes.
Paint-brushes (i inch and 2 inch).
Emery paper.
Sand-bath.
Wire wool.
Soap.
Wax (to wax labels).
Small tag labels.
Reinforcing material (sacking, iron rods, and wire).
Parafiin wax.
Some source of heat.
Sandpaper.
170
tke field-laboratory
I eIo noi profKwe hereto present a manual of firsinaid: severnt books are
available.' Bur »indicaie aotoclhiflg of the required scope of this little
labofiitoiif) 1 append u few notes Oa typical operaiiona,
1, pmticuliifly ironwiOTlc, mmt be ptescived sufli-
defltly to wahjc them to travel witbotir funher dani4ge. Iron objects
(sn often be salved lonporarily by covering ihera wth aplasterenvelope,
a process which must sometunes be done before the rerooval of the ob¬
jects from the posiiion io which they arc found. After being thw-streaied,
they should be outfuUy but firraly bound to wooden splint* by luctuu:
of a clorh bandage. (Pfior to ircatment, however, they shoutd be drawn
to scale by tbc expedition's draftsman, where possible on a glass plate—
see p. T47O Subsequent tteatmeni depends upon cimunstiuiceSj but
should nonnalJy be deferred until fulJ laboratOty equipment is available.
PanfSn wax (with a high naclring-poiut) has frequently been used in¬
stead of plaster as a tempoiaty jaekci Ibr metal or other fragile objecia,
but is a last resort. Care must be taken to avoid melting the was inm
the objeas; the wos is not easy to remove completely for the subsequent
tteamanu, and, it salts are thus imprisoned, diainiegmtion Is hwtened
TBthcr dum deiayed-
2. Ccini^ if still illegible after drying and hrusbingy muw usually he
treated on the spor. The natuie of the maunenL will depend upon ihcii
metal and its condition, and the field-chtanist must bc folly tramed and
experienced in this work if he is tO do more good than harm. In paiticu-
lar, li flntil rtf fAf outui be afcerfrtfiwd tehethrr ikere it a reaf^mabti core vf
the originai wk-w/—otherwise the washing away of impmiiics wash
away the coin. (The presence ofamctalcoitiaa corroded irm object can
generally be deteeiod by its being strongly attracted by a rngoel.) In a
Qiaioriiy of instancra, where there is a considenibtc surviving core, the
electrolytic method is the safest for mcial, but this is not always ftasibhi.
It consists of the suspension of the object (coin, flee.) on a copper wire
attached to the negative pole of a baiieiy, and immeniog it in a 2I per
cent , caustic soda solution ewntaiued in a glass vessel, in which'» ™'
icer?ed a piece of graphite wired to the poflriw pole of the bancty. The
electric current passes through the objea and the solution and removs
impurities ftom the me»l. The difflculiy is that the ban«y needs
techaigitig after 24-4B hours* work, and it is iwt always possible to »us-
nin the necessary batteiy service. After tratmcol, the objects are
' e.s- II. J. Plctiderkith. The Pwrmitiim of dnhgiiifrV, ft Jindon, lOMli
.rHI wicwf Iffdiaj li£>. I (PcUii* tpp, ►
m
THE FIELD-LABORATORY
thorougWy washed in distiUed water and are then coated with bedaciyl
or polyvinyl acetate or some equivalent protective covering; bakelite
varnish will serve if nothing better is available. Gold, of course, needs no
chemical treatment, unless marred by obstinate stains or incrustations,
wWch CM be freed by immersion in a strong hydrochloric add or by
boiling in a detergent solution.
electrol)mc method is considered unsafe, copper or its alloys may
be deaned in a mixture of i part tartaric add, i part of caustic soda, and
10 parts of water. The objects should be kept in the mixture until all the
green incrustations have been dissolved away, leaving the liver-red core
behind. They are then thoroughly washed in water and finally coated
with bedac^-l or jwlyvinyl acetate. An alternative treatment is: (i) dtric
add, and pickling in 50 per cent, sulphuric add to remove any red oxide;
(ii) neutraê with ammonia or any alkali after the add treatment; (iii)
wash in distilled water; (iv) carry out the silver nitrate test with the last
wash-water (see above); (v) dry in alcohol; and (vi) coat with bedactyl
or polyvinyl acetate. Completely oxidized copper coins should, at
the outset, be left in a 10 per cent, soditim metaphosphate solution
untd free from calcareous matter. Sometimes this is suffident to reveal
the inscription; if not, they should be treated with the above-mentioned
tartrate mixture dUuted to half strength.
Silver coins and other objeas, if the metal is debased with copper (as
It not mfrequendy is in the later Roman coinage), can be cleaned with
3 per cent, sulphuric add until free from all red spots of copper oxide.
Thereafter, the coins are brushed and well washed in water. Pure silver,
if superficially corroded, can be deaned by immersion in dilute am¬
monia or dilute formic add. Or it may be wrapped in zinc sheeting and
suspended for a couple of hours in water addified with a few drops of
acetic add.
Whatever process be adopted, the director must ensure that at no stage
is the object separatedfrom its site-label (which should be waxed to avoid
defacement). The chemist must be suflSdently an archaeologist to
appreaate fully the importance of the label. See that he does not under¬
take too much work at any one time; a crowded laboratory wiU inevitably
lead to confusion of evidence, particularly in the case of coins and other
smaU objects. When a coin is immersed for deaning, its label must be
securely attached to the wire whereby it is suspended.
3. Tablets or seals of unbaked clay, such as are found on Mesopo¬
tamian and Indian sites, must be baked carefully for handling and
172
THE eiEI,D-LA90EATOHY
preservaiion. This wotk he within the scope of any citpciienced
excavator, but, all Thjnga being equal, is a rail change upon the chemisL's
Miw. The method iS Mly described and illustnied by P, Dcluugu,
'ThcTratmeul of Clay I’ablcts in the Field’, Studieiin yindtat Onoitai
cmisttliirn, no. 7 (Oiiaigo, 1933). and need not here be nspeated. It
be added, however, that tbe exposure ot ornamciit or script is
deferred unlil after the hahiiig and is most effectively developed, accord¬
ing to modera practice, with the aid of a simple forai of sand blast.
4. Ohjecti <!/w«Kifteshly dug from damp soil are liable to split, wtp,
nr be dcbiioyed altogether on drying. The taak in the Geld is to maintain
their humidity, for example by embedding them in a ihick layer of wet
sawdust, iHioss, or newspaper. Sometimes the wood caabe sffflriy dried
and the moisctirc replaced by glycerine, which is retained, by a akiii of
to pet cent, polyvinyl Boctale ot shellac in alcoLoL Wood ftoni saline
■teas should be washed in salt-fiee water ur treated with appbcitioiis Of
Liquid paper-pulp for the purpose of drawing out the salt.
An equivalent treatment is necessary for shale and Icuhcr.
The vforkiog-principle should be m keep all stich objccis damp until
they can be handed over to a fully equipped laboratory, and » avoid
unpregoatiotJ with wax or other material liable to nuke permanent
conservadoD difficult {see p. 171),
5. £tme! must be caiefitlly and lightly brushed dean, wd can be
pointed or sprayed before removal with polyvinyl aocute diluted wth
toluene or metbylatod spirits or thcllac diluted with racthylBiedi spinis
or alcohol. If the bones can be retooved but are still fragile, they
can bo soaked for 3-4 days in a tank coniaining bcdacryl or polywiiyl
acetate suitably diluted with toluol with an air-spow uiwkr ffie hd.
They should then be laid our on a wire grid, mlierwise they will stick
to everything.
17J
XV
Photography'
T he overriding difficulty of the archaeological photographer is to
induce his camera to tell the truth. That quality is as much a mat¬
ter of proper emphasis as of accumulative statement, and not a
litde of the photographer’s time and skill, both in the field and in the
studio, are devoted to the rescue of the more from the less significant.
The preparation of the subject, the selection of light and angle and lens,
the use or non-use of filters, the choice of ‘hard’, ‘soft’, or ‘medium’
printing-paper, the differential printing of portions of the same negative
ue all matters which extend the photographer’s function beyond the
limits of mere techmcal skill. In all of them the director is as busily con-
cemed as is his photographer.
No attempt is made here to interpolate a manual of photography. It is
assumed that techmcal proficiency and good equipment are alike at the
direaor’s disposal. They arc not difficult to find. On the other hand,
their efficient utilization is rare enough to be classed amongst the vir¬
tues, and a few notes upon this matter are not out of place.
No amount of mechanical skill is a substitute for the careful prepara¬
tion of the subject. Clean, sharp angles between the divergent planes of a
section, c^fuUy and emphatically cut with trowel, knife, or edging-tool,
arc essential if the section is to tell its story with the minimum of con¬
fusion. Furthermore, a spotlessly clean trench is no mere ‘eye-wash’, if
only because it gives the spectator a justifiable trust in the orderliness and
accuracy of the work. Even ±e top edges of a trench should be neatly
tnmmed and the grass cut and swept along them; a stray blade of grass
in the foreground of the picture may be overlooked by the eye but may
loom embarrassingly in the lens. Strata readily distinguishable in nature
my merge m the black-and-white of the plate and may, on occasion,
have to be emphasized by careful spraying or by additional smoothing
or even deliberate roughening, though such aids should be used only
where aU other methods (e.g. the use of a filter) fail. An example is iUus-
trat<^ on PI. XIX. Sometimes, particularly in a dry Eastern climate, a
whole section may have to be damped to bring out its texture or colour-
this ^pwr.^' enough to check the technical points in
174
PHOIOGHAPHV
laf. In oae way or aiiotlier tnusi w ihiw compensate for die absence of
phyiugmpluc colour imti] such rime is. enlour-photography becomo the
nomml cnedliirm^ 1/
Then there l5 the selection of light. FormoiiLarchicologicaJ sobjects in
the open thete Is one oprinmin momeiii daring the day, and the day it-
may have to be chosen carefully for shadow^ of half-lighi. In
the Eastj wheie the snong sun usimlLy kiJL die detail of a suhieo, moat
of my phoTOgnaphy was done in the Oeetbig moments beiwcen first light
iuid sunrise. The rime-mai^ in these dbcumstances is a matter of
minuteSj and everything must be prepared beforehand. In Epcdal ck-
cnmstiHiices rejected Euniighi tnay be employed j ihus I photogniphed
the scuJpniTCS in the Hicpimiia caves near Bombay by mcmi of sim-
light reheacd into the dark recesses by a succeswon of kript mirrors.
\\'hateveT the spcdal piobicmi it may in fflct be laid down as a general
rule ihai ihc preparaiicia of a subject ocnificgi hoiin, DcoisionaUy days*
before ihe brief sosioa witli the camm. Innuracmblc slipshod and
uninfonruTivc phatogmplis ia eicavarion-rcpcifna [PJ. XXII) prove rliar
this etabomre prepaiadon tS not tinncceasary. r
Every archaeological photograph should include a sCnlOj cither in the
form of a graduated rule of rod or in that of a human hg'^rc. Âdult
human shdctons pfovidc thcir own iralc Miith as much accuraCT as may
be expected from it photograph.) The should noran-illy be parallel
wiih The plane of the camera-plate; if the Initcr is tilted^ the gradkiati^
scale fthnuld be ODtrcspondingly liltcdT otherwise the graduations axe in
pcrepcctivc and of variable length. With very rare eroeptions, the scale
must be prcosdy parallel with the aide Of the platCi and great care
should be taken in its placing. Nothxng looks worse than a sole unin'
tcntionally out of the vtrrical or borizontoL Inridcntally the scale should
be dean and imscamd; it ia prtfcrabk to reserve a grudoated pole
spedally for photography.
Care should also be taken to ensure that ihe mlc is in the sajm plane
the mam fcatiim of the subject. It is surprising how often this obvious
precaution is overlooked, a scnle, for example, bang placed consider¬
ably nearer the camera than a srSfitU object with whicii it is supposedly
ossockted. On the oEhcr hand, the scale ihanJd not monopoliw the
attention of the spectator. A ccnml scale is, for this ct^tson* usually bad.
* tliAEi bewever cnticli lilted lldrwird tbr cafflen nuy be. It iwrf be level
borl2i.intdij * if any boruon w viwbk, nAd ihc lercUina of the camera wiib a
bubblc*lGvc]i u One of the fint aco of the pbotograpber,
m
PHOTOGRAPHY
Where the scale is a human being, as is often desirable in large subjects,

the individual thus honoured must remember that he is a mere accessory,
just so many feet of bone and muscle. I have in front of me as I write
a monumental report on a Palestinian excavation in which a collotype
illustration nominally representing a building under excavation shows
in the centre the figure of the direaor facing the camera with a look of
smug self-satisfaction, his hat in his hand so that no feature of interest
may escape his admirers. Another, even more ludicrous, misuse of the
human scale funher East is illustrated in PI. XXI b, the sinner again be¬
ing the director himself. Two axioms of the use of the human scale are
(i) that the figure shall not occupy a disproportionately large portion of
the picture (if so, a linear scale must be substituted), and (2) that the
figure shall not look at the camera but shall be ostensibly employed in as
impersonal a manner as possible. On occasion the figure can be some¬
thing more ±an a passive scale. For example at Wroxeter, Mr. J. P.
Bushe Fox was hard put to it to express the placing of a flue in an un¬
spectacular fragment of walling, but eventually lighted upon the suc¬
cessful device of including in his snapshot a workman pouring a bucket
of water into the upper end of the flue so that the water, emerging from
the lower end, indicated the continuity of the opening through the
thickness of the wall.'
With panchromatic plates or films a colour-filter is not usually neces¬
sary. To emphasize reds and blacks, however, and to eliminate greens
and yellows, a green filter may be used with these plates; whilst a yellow
filter will produce tone-values more nearly approximating to those
observed by the naked eye. The red filter will lighten all reds and yellows,
darken all greens and blues (e.g. in the sky, thus emphasizing clouds),
and will separate red from black. Variations of exposure for panchro¬
matic plates are as follows:
With a green filter, an exposure six times as long as normal.
With a yellow filter, an exposure twice as long as normal.
With a red filter, an exposure four times as long as normal.
Orthochromatic or yellow-sensitive plates are not used with a red
yellow filter with these plates will lighten yellow up to light
orange and will darken all blues. The variant exposures are as follows:
With a green filter, an exposure nine times as long as normal.
With a yellow filter, an exposure five times as long as normal.
* Wroxeter Report 1914 (Soc. Ant. Lond. Research Report), pi. x, fig. i.
176
Pl.ATH XIK
P,uj of aKii™ thnniKh the ^.1

Imbedclcd in eanhinil sravd but cmpiiaiibed by wtrlnl Ushtinj.
iSit f lU')
PIlOTOGflArHY
Other of importanoe are the foca! Icngtli of ilie lens and ihc
Kgulaiion of the Icns-apcniuc or ‘Stop’. *I'hcsc fiicioR we related to
enct other and to the size of the camera, Remarkabk teiiilts on be
obtflinctl by v'«y small cameras, and on distani expeditions it may he
necessary to adhere to them csdusively. But 1 have no hedmiinn in
ncommendinj; the use of a fuU-pIate (Six 8 j iochoi} ojoera whcnCTcr
possible- The main loima for iliis choice has recently been condady
expressed by Miss Alisoti Fiantx as fbUotw:
Lenses are described and identified by their focal lengths- The focal
leneUi, which is fixed for each lens, is approximately the distance between
the liaii luid Xht LmaBfl on the film when ihc lcii5 is focused qd h distiuii
point. Lenses of different focal lengths may be med, within tcason, on
canienis of differeat sixes. For general use, however, a Jms of foes) length
eoual to or sfightly greater than the dtaganaJ of the picrure.area li t^stom-
ary. Therefore the lorger the camera the greater the focal length of its
mjrmd lens, The images projected cm the film by lenses of (he same local
1en(^ at a ^vcd dis^^mcc Erom (he c^tijecc are (be Mm* size nrgardlcss
the fif the nnd th£ imaBt-siz* increaKs with thfi food length.
Thcicfote Q larger camera will mice in a greater field ihati a smaller camera
equipped with the sntue Itna.'
Lenses with a short focal lengtli and a proportioMKly wide imê of

vision ore sometimes necessary, parucularly at dose range m a co. ^ ^
space, Bm such a tens esceed* the capodiy and cupeaation of the
human eye and reduces the normal visual emphasis of the êct, i.e.
flattens it; as that the effect is distortion and fidsificadon- The golden
rule is to use as long a focal length ^.e. as narrow a visual ang e) « a
subject will pennit. And since, aa explained above, a lai^ plate focdi-
tates this procedure by providing a larger and more in laive e or
any aven lens, a targe camera is preferable to a small one.
,In practice, every archacotogical camera requires rh™ aliemaove
Icitw ■ a lon£'"focu5 (nurtow Miglc) Ics^y fl incdiiJJiV-fbciia 3n
short-focus (wide angle) lens- Sotnething lihe . li-inch lens, a
lens, and a 6-foch lens, respectively, will suffice m normal usage. To
these may be added a tdcpiioto lens, not often used bur ocsssionatly of
great *
t The use of alaiê camera with a InngTocus hoi^n.
a proper uodeistanding of the use of the ^ t c tegol ^
the aperture in front of the lens. The depth ofthe field, or la other words
■ drf CA*eh- fo"-
»T¥ 177
PHOTOGRAPHY
the distance between the nearest and the farthest points of sharp forus,
decreases as the focal length of the lens increases. On the other
hand, the depth of the field increases as the aperture in the dia¬
phragm (or screen in front of the lens) decreases. Jn a subjea of any
appreciable depth, therefore, the one factor has to oe set off against the
other. A long or longish lens, desirable to ensure an undistorted per¬
spective, must be given a greater depth of focus by reducing the size of
the aperture, i.e. by ‘stopping down’. The widest stop is generally f. 6*3,
the narrowest f. 64, the usual series being 63, 8, ii, 16, 22, 32, 45, 64.
For subjeas of average depth, something midway between the two ex¬
tremes—e.g. f. 32—is commonly adequate. It must be remembered that
as the aperture narrows the length of exposure increases because less
light penetrates, the rate of increase being x 2 for each successive smal¬
ler stop. Thus if f. 22 requires an exposure of half a minute, f. 32 will
require an exposure of a minute.
One funher point in connexion with the ‘stop’. With any appreciable
reduction of the aperture, the lens must be focused on the foreground—
say, 15 feet from the camera on the average—and not on the middle dis¬
tance or background.
In the actual exposure care must be taken to avoid halation. Although
modern^ films are normally ‘backed’ or ‘dyed’ to reduce this, the diffi¬
culty is not thereby eliminated. Sky seen through trees or over a building
or rampart may produce indistinctness or fogging during a long expo¬
sure timed to bring out the detail of a dark foreground. When the lens is
within 45 degrees of the sun it must be carefully shaded by means of a
hat or book or sheet of cardboard.
To the apparatus already indicated for out-door photography one
more item should be added: a mobile tower 10-15 feet high. In the
East, where supplies and labour were relatively abundant, I had a
wooden tower run up on every major site (pi. XX). In Britain also
towers of various designs have been widely used, though here I generally
use trestles borrowed locally—a lazy man’s substitute.
So much for out-of-doors. For the indoor photography of individual
objects many of the same rules or recommendations apply but a few
others may be added. When possible, it is the convention that objects
shall be lighted from the top left, but this usage has often to be varied
in order to emphasize a design or inscription for which some other
angle is more suitable. In any case a proportion of reflected light, trans¬
mitted from a white board or a sheet of tin or a board covered with tin-
178
PHOTOGRAPHY
foil, is usually necessary to show up the shadowy side and to lift the ob¬
ject from its backgroimd. In a fully equipped studio, artificial lighting
may be used throughout, and with experience undoubtedly yields the
most reliable results.
For background I prefer black velvet save when the object itself
approximates to that colour. For dark objects, a sheet of glass raised 4 or
5 inches above a sheet of light-coloured (not quite white) paper on four
comer-blocks gives a good shadowless baclôund. Pure white paper
may produce a slight halation. When, as in the photography of coins,
it is desirable to include both sides of an objea on the same plate, by
interrupting the exposure and turning the object over, a black back-
ground is of course essential.
In photographing an objea at close range, it is important to ensure
that the scale is in or very near the frontal plane. Otherwise, if the objea
is of any considerable depth and the scale is placed on the background,
an appreciable disparity will result. In these circumstances the scale
should be raised to the required height on a thin strip of wood or plasti-
cine.
Whatever the subjea, the background and lighting should be such
that no cutting out is subsequendy necessary on the film or plate. To
have to block out an irrelevant shadow or other feature is a ronfession of
failure and, particularly if the'objea is a work of art, materially reduces
the value of the record.
Lasdy, all field-photographs must be developed immediately. Many ot
them cannot be repeated at a later date, and the director must be assured
of his results within 20 minutes of the exposures. Rough prints should
follow within 12 hours.
ADMINISTRATION AND RECORD KEEPING
The photographer’s notebook must contain all details reladng to

time, position, exposure, and filter used. Thus
Harappa_
Site E. Sect. X. South face_

1145 hrs. Strong sunlight_
12' lens. Red filter. F.22. 3 secs.

Fas/» of access to natives is essential from the outset of an excavation
and, to this end, all negatives made and approved should be numbered
179
PHOTOGRAPHY
with a serial no., orientation, the page in the site-supervisor’s notebook,

and, if possible, the position on the site and the general plan or the
drawing no. (see below). It is quite simple to do aU this in the rebate or
margin made by the dark slide, and the work should be carried out with
waterproof ink and a mapping pen. If celophane envelopes are used for
storing ne^tives, then the same details will be added to the envelope,
together with the type of printing paper used in the print produced.
Thus, if the Negative Register is ever lost or destroyed, the details would
still be available. In the same manner the keeping of a Negative Register
IS essential, columnized to receive all the foregoing details but with an
added Remarks’ column in which notes can be made of such matters as
the existence of lantern-slide negatives of the same subject, or whether
the negative h^ been used for publication, the date and reference of
publication being given.
Diagrammatic Examples
Negative bag
237. RANCHI. SITE E. SECT. X.
237
RANCHI
SITE E. SECT. X
SOUTH FACE HUMUS REMOVED
NOTEBOOK 10. P.41
DRAWING NO. I5
KODAK BROMIDE GRADE 5
10 SECS.
Negative Register
RANCHI
Slip.
Neg, Drteg, note-
serial Site Section no. hook Remarks
237 E X 10
15 Humus removed. Interim
South face p.41 Report pi. X. Lantern-slide
made.
equipment
In summary, the archaeological photographer in the field needs the

following equipment:
180
Tower for high-angle photography.
{Stt p. /7«.)
PLATE XX
PHOTOGRAPHY
1. A field camera of strong construction, full-plate size (6|x8i

inches), with rising and falling lens-panel and swing back.
At least six dark slides, each to hold two plates or films, and num¬
bered.
2. Not less than three lenses: a long-focus lens, a medium-focus lens,
and a short focus or wide angle. All should be of the best quality
and anasdgmadc, with a maximum aperture of f.6*3 (most of the
exposures will be ‘dme’). A telephoto lens may usefully be added
for occasional use.
3. A set of filters (red, green, and yellow) to fit each lens or adapted to
fit the whole series of lenses.
4. A heavy tripod, capable of raising the camera to a height of 6 feet,
with a ‘ tilting head* to enable verdcal photographs to be taken.
5. A 6-inch bubble-level.
6. A large focusing clothe
7. A small hand-camera of the Retina Contax or Rolliccad type for
speed or colour work.
8. A small box which can hold paint-brush, small scales, pins, and
plasdcine.
9. Tanks or dishes, pne for developer, one for fixing, and one for wash¬
ing. For economy, use packed developer and Hypo powder in dns.
10. Printing frame. Check-prints from negadves will have to be made
in the field, and, since condidons are likely to be cramped, the
printing can be carried out with a printing frame. Bromide paper
in three grades (soft, normal, and contrast) is required.
Note, For both ncgadvc-development and printing, a small timing

clock is an essential, together with a thermometer to ensure that all solu-
dons are at a working temperature. In Great Britain even in summer it is
necessary on occasion to bring the devclop>cr temperature up to 65 F.,
and in eastern countries it often becomes essential to bring the soludon
down to a maximum of 75^ F. by the use of ice.
181
XVI
Publication and Publicity

* A discovery dates only from the time of the record of it, and not
/A from the time of its being found in the soil.’ This classic sentence
-ôi^Pitt Rivers proclaims fairly and squarely the ultimate moral and
sdentific duty of the field-archaeologist. It may be amplified by the
famihar corollary that unrecorded excavation is the unforgivable des¬
truction of evidence j and the more complete and scientific the excava¬
tion, the greater the measure of destruction. The less persistent methods
of an older age may yet leave vital evidence for recovery: for example,
in Lydney Park, Gloucestershire, a prehistoric and Romano-British site
dug very extensively by an antiquary in 1805 retained all the evidence
necessary to reconstruct its character and chronology by re-excavation in
1929. But today a site comprehensively excavated by modem methods
may be practically gutted of its evidence, and must be wrinen off. The
emphasis is on the word ‘written’.
The first task in the compilation of an excavation-report is adequate
illustration. In this matter there is little to add, in principle, to Sir
Flinders Petrie s assertion half a century ago that ‘nowadays the main
Struve of a book on any descriptive science is its plates’. In faa,
Petrie s own standards of illustration, genius though he was, generally
fell far short of those of his older contemporary, Pitt Rivers, and would
not be accepted by the average British excavator of today. Nevertheless,
his chapter on publication* stands as a fair general statement of the
theory of the business.
In the matter of literary content it may be said that there are two pre-
luninary problans to consider: those of substance and of form. First,
how much detail shall a report contain? The answer is necessarily con¬
ditioned in some degree by circumstance. Quantitatively, Pitt Rivers
digging a reluctant ditch in Cranbome Chase and Petrie digging a
^wded cemetery in Egypt are up against very different propositions.
Let us glance at their reactions.
Pm Rivers approaches the problem with a formidable combination of
acute sense and Viaorian probity.
* Methods and Aims in Archaeology^
182
PUBLICATION AND PUBLICITY
The record of an excavator [he says] takes about five times as long as the
actual digging_In my fourth volume... everything has been recorded,
however small and however common. . . . Everything has been drawn,
down to the most minute fragment of pottery that had a pattern on it.
Common things; are of more importance than particular things, because
they are more prevalent. 1 have always remembered a remark of Professor
Huxley’s in one of his addresses. ‘The word “impor^ce”,’ he said,
‘ought to be struck out of scientific dictionaries; that which is im;»rtant
is that which is persistent.’ Conunon things vary in form, as the idea of
them passes from place to place, and the date of them and of the places
in which they are foimd may sometimes be determined by gradual
variations of form. There is no knowing what may hereafter be found to
be most interesting. Things apt to be overlook^ nwy afterwards turn out
to be of the greatest value in tracing the distribution of forms. This will
be admitted when it is recognized that distribution is a necessary prelude
to generalization. I regret to find in endeavouring to trace the distribution
of panems, that archaeological societies illustrate fewer things than for¬
merly. It is thought, perhaps, that when a form Im become coôn, there
is no use repeating or even recording it. This is a great mistake in my
opinion_The illustrations need not be elaborate, but sufficient to trace
the transitions of forms.
The General goes on to describe his modus operandi.
The compilation of a work of so much detail necessitates the employ¬
ment of clerks. I make it a rule that nothing in the letter-press should be
issued that is not in my own writing, and of course I am responsible for
the whole. But the calculation of the numerous and tedious mdices; the
compilation of reUc tables; the photographs; the identification, measur^
ment and restoration of the skulls, bones and pottery; the the
contouring; careful labelling and correction of proofs; the of
the plates; . . . requires the assistance of at least three men of different
qualifications. Living in my house [he adds with a deUghtful Victoi^
smugness] they must necessarily be men of good character as well as
energy. Those who have left me have generally obtamed more lucmnve
employments_As a rule I have been well served by my clerks.
Thus Pitt Rivers. Petrie, overwhelmed by the mass of finds from his
Egyptian sites, seeks refuge, less in men of good charaaer, than tn basic
index-series or corpora, to which newly found material can be briefly
and simply related.
The practical utility of such a carpus is found at once when excavating.
* Excavations in Cranbome Chase, iv (1898), 27—28.

183
Formerly it was needful to keep dozens of broken specimens, which were

of no value except for the fact of being found along with other vases.
Now the excavator merely needs to look over the corpus of plates, and
writes down on the plan of the tomb, say B 23, P 35 b, C 15, F 72, thus
the whole record is made, and not a single piece need be kept unless it is
a good specimen.
Be it repeated that these two different methods arise from different

local problems, and for neither can universal validity be claimed. There
cannot be the slightest doubt that the Pitt Rivers system is the ideal aim,
but it is only feasible where the material is limited in quantity, or where
it does not lend itself to corpus classification: that is, where it is too frag¬
mentary or variable for easy systematization. It cannot be disputed that
in the more evolved industries, with which Petrie mostly dealt, a useful
corpus'-sysittn can and should be evolved. An obvious example of a cera¬
mic susceptible of corpus treatment is Roman terra sigillata or Samian
ware, which is, or shortly should be, nearly as capable of a detailed
standard notation as are Roman coins. These are now becoming a re¬
latively simple matter. Thus in a report of mine some years ago Air.
B. H. St. J. O’Neil, later Chief Inspeaor of Ancient Monuments, was
able to list the Roman coins with abbreviated references largely to the
then new standard works of Mattingly and Sydenham, and thereby
compressed his reasoned catalogue of 1,668 coins into eleven pages.
This achievement may be contrasted with the allocation of eighty pages
to 1,000 coins required in the Wroxeter reports in the same series, issued
before the Mattingly-Sydenham corpus was published. In other words,
publication without the corpus occupied twelve times as much space as
publication with the corpus. The advantages of a scholarly corpus or
yardstick need no further emphasis in such cases, and the extension of
the corpus-sysiem is certainly no less urgent now than it was in Petrie’s
day. But always with the proviso that it presents a very serious danger: it
lends itself to loose usage and to the overlcxiking of those subtle varia¬
tions of form, the imponance of which Pin Rivers rightly emphasized.
Generally speaking, only an evolved and largely mechanized industry
offers suitable material for a corpus. In practice, particularly in a mal¬
leable material such as pottery (unless made in moulds), it will always
have to be extensively supplemented by individual illustration.
Whilst on the subject of the illustration of pottery, it may be recalled
as a matter of passing interest that 1952 marked the centenary of the
systematic sectional illustration of pottery in this country. The method,
184
now standard, whereby a pot is represented wholly or partly in section,

although used occasionally by Samuel Lysons early in the nineteenth
century, was first formulated by (Sir) John Evans in his report on Roman
remains found at Box Moor, Herts.’, published in Archaeologia in 1852.
He there shows four Samian vessels in section, and is at pains to des¬
cribe his modus operandi. His description is a curious legacy from a
ppinctafcingr age, and the description is worth quoting.
As the taking an accurate section of vessels such as those delineated

fhe writes] at first sight presents some litde difficulty, it may not be
altogether useless to record the process by which these secuons were
obtained. The piece of Samian ware after being slighUy greased was
plunged into fine sand, in a direction perpendicular to the a^s of the
vessel, to vrithin about a quarter of an inch of the centre, and ^ smfacc
of the sand was then made level. A thin mixture of plaster of Pans was
next poured upon the sand until its upper surface vras level with the
centre of the vessel. When this was set, the fragment of Samian and Ac
plaster were removed from the sand, and the plaster broken into a suffi¬
cient number of pieces to set the fragment free, and the^ pieces bemg
reunited with their upper side downwards on a piece of paper, ^ve a
section from which the form of the vessel could be accurately traced.
All this is a little reminiscent of the burning-down of the swmeherd s
cottage to procure roast pig; but in substance the procedure was a (ordi¬
nal one, and is not the least of the many gifts for which our disaplme is
indebted to the great Evans clan.
I turn now to the second of our preliminary problems, that of the
form which the excavation-report should assume. Here thê is more
room than usual for idiosyncracy. Certain desiderata can, howler, be
premised. The report is intendedprimarUyfor thespea^t-reacto, but
even the specialist may be accredited with a measure of human fi^^.
He is not above the appreciation of conciseness, clan^, and rwdy
accessibility-three virtues which are less common than ^
in such Uteraturc. The Royal Society, m a pamphlet issued °
‘The Preparation of Scientific Papers’, has clear wortk upon this matter.
‘Most [scientific] journals’, it remarks with an i^ble ^ymasm, pre¬
fer papers written for the moderate specialist, that is to say, an auffior
shoSirwrite, not for the half-dozen people in the
terested in his line of work, but for the hundred or so who my in¬
terested in some aspect of it if the paper is weU written. After aU, an
■ General Notes on the Preparation of Scientific Papers (London, the Roy
Society, 1950), p. 3*
185
excavation-report is a scientific newspaper, with news-paragraphs, lead¬

ing-articles, stock-market, hatch-match-and-despatch, and even ‘small
wants’. Let the writer of it study unashamedly the higher forms of
journalism, and neither he nor his clientele will lose thereby. And just as,
in a newspaper, one does not want or expect to read solidly from front
to back in order to discover the sahent news and views of the day, so in
a well-balanced excavation-report the student may properly expect to
discover something of the wood without a prolonged, tedious, and exas¬
perating hunt amongst the trees. Of course the trees must be there, or
the wood would not exist. But, alas, how many reports are merely
jungle!—a-biblia, fit only to be added to Elia’s catalogue of books
which are no books,
A year or two ago, a joint-meeting of three sections of the British
^sodation for the Advancement of Sdence discussed ‘The presenta¬
tion of technical information’, and the primary speaker. Professor R. O.
Kapp, had much that was pertinent to say. Today, he observed, talk and
paper were among the more important of the tools with which the sden-
tist and the engineer had to work. Time was just as predous when read¬
ing a report as it was when usmg an instrument, and the worker could
not afford to wait while a verbose author was developing an argument at
unnecessary length. If one could justify training for research one could
justify training in the an of exposition. Let sdentists attempt deliber¬
ately and systematically to raise the standards of exposition in all its
aspects, and try to perfect and teach it. Until that was done, sdence
would continue to be hampered by the bad work of poor expositors.
Professor Kapp was followed and remforced by the representative of a
distinguished firm of pubhshers, who remarked that in ten years, of over
600 artides of a sdentific kind read and checked for publication, only
five authors had sent in manuscripts in such a form that they could be
forwarded to the printer without dday or query. He might well have
added an old dictum of J. M. Barrie’s: ‘The Man of Sdence appears to
be the only man who has something to say, just now—and the only man
who does not know how to say it.’ It is but fair, however, to add that our
leactog sdentists are today aware of this shortcoming. At the Royal
Sodety’s Sdentific Information Conference in 1948 considerable con¬
cern was expressed about the quahty of sdentific papers presented for
publication, and die pamphlet (already referred to, p. 185) on ‘The
Preparation of Sdentific Papers’ was a result. Nor are the disease and
its remedies confined to the sdentific world. I commend to all and
186
sundry two excellent booklets recently issued as directives to the Civd

Service in an attempt to purge official English. They are entitled Plain
Words and are from the experienced pen of Sir Ernest Gowers.* They
are admirable reading and should be in the hand of every archaeologist.
Yet another word may be said upon this matter, for it is near to the
heart of the business. I would appeal for a little more artistry in our
scholarship. A generation ago, G. M. Trevelyan was inveighing apinst
the dullness of historians, and we archaeologists are certainly m no
better case.
The idea that histories [Trevelyan said] which are delightful to read
must be the work of superficial temperaments, and that a crabbed style
betokens a deep thinker or conscientious worker, is the reverse of truth.
What is easy to read has been difficult to write. The labour of wntmg
and rewriting, correcting and recorrecting, is the due exacted by eve^
good book from its author, even if he knows from the beginning ejactly
what he wants to say. A limpid style is invariably the result of tod labour,
and the easUy flowing connection of sentence with sentence and paragraph
with paragraph has always been won by the sweat of the brow.
That is, of course, as true to the archaeologist as to the historian. Sweat¬
ing with the pen is no less important than sweating with the spade. And
Trevelyan cannot be bettered when he stresses the literary funcuon of
the historian,
the exposition of the results of science and imagination in a form ttot
attract and educate our feUow countrymen. . . . I wish [he adds] to by
greater stress than modem historians are willing to do l»th on the d^-
culty and also on the importance of planning and writing a powerM
narrative. . . . Arrangement, composition and style are not as ea^y
acquired as the art of typewriting. Literature never hdpsjmy ^
task until, to obtain her services, he is willing to be her faithful apprenuce.
Writing is not, therefore, a secondary but one of the primary tasks of the
historian.^
Of this excessively neglected matter of Uterary style more wiU be ^d
later in the chapter. MeanwhUe, it is opportune at this pomt to consider
the form and planning of an excavation-report. Plam and effecuve wnt-
ing impUes a plain and effective structure, and even second-rate expns-
sion may sometimes pass if the basic arrangement is sound and simple.
■ Sit E. Gowers, Plain Words; A Gtdde to the Use of English (H.M. Sutionery
Office, 1948), and ABC oj Plain Words (same, 1951)-
» Clio a Muse and other Essays (London, I9*3)> P- 34-
^ Ibid., p. 31.
187
a part of his archaeological training, to visit a blockmaker’s premises and

to see the processes in operation. Pity the poor blockmaker! His task at
the best is no easy one.
A few other elementary exhortations may be added. Let us see to it
that our maps, plans, diagrams, photographs state their purpose lucidly
and concisely, without excessive margins and other waste space, illegible
or untidy figures and lettering, or other irrelevant litter; and let them
be printed on the right kind of paper. ‘Of course’, you will remark; but
in self-justification I reproduce here two illustrations from a standard
archaeological publication as cautionary examples (PI. XXI A and b).
Let us in our line-drawings ensure clean lines, with proper differential
emphasis, and simple lettering properly incorporated into the ‘picture’
with a just sense of balance. And though the artistic appeal of a scientific
drawing is of secondary importance, aesthedc quality is not a negligible
faaor. It is at least as important as is good literary form in the text. It
can help to attract and hold the eye, and so tighten the liaison between
author and reader—an eminently desirable objective. In an archaeo¬
logical photograph, a clear-cut, well-swept subject will tell its story with
least effort to the spectator and will, incidentally, carry additional con¬
viction as evidence of dehberation (contrast PI. XXII). It is a sound
general rule that tmtidy work is muddled work. All that I have said in
the ^lier part of this chapter about the need for clarity of literary style
applies to the all-important matter of illustration.
So much for general principle. In practice, as we all well know, archaeo¬
logical illustrations are normally reproduced in one of three ways:
(a) by half-tone block, (b) by line-block, or (c) by lithograph (direct and
offset). Collotype and equivalent processes are sometimes used, particu¬
larly in foreign publications, with results which may be superficially
attractive, even sumptuous; but detail is liable to be lost in inky shadows,
and the restfft cannot compete with good half-tone. Colour-processes
are here omitted, although they have manifest advantages over black-
and-white and will in due course become the normal medium. Indeed,
they would doubtless already be the standard method, but for the mess
which the world has made of its economics during the past dozen years.
But it is time to pass on to another aspea of the subject: to turn from

the problems attendant upon scientific publication to a related matter
which IS at the same time easier and more difiScult: to the not unimpor¬
tant question of publicity and vulgarization.
190
Once more we may begin with Pitt Rivers. Speaking in 1897 he

laments the inadequacy (at that period) of archaeological publicity.
If ever a time should come [he observes] when our illustrated news¬
papers take to recording interesting and sensible things, a new era will
have arrived in the usefulness of these journals. The supply, of course,
must eqiial the demand, but the demand shows what intensely stupid
people we are. People bowing to one another appears to form the staple
of those productions, as if it were not bad enough for those who are
compeUed actuaUy to take part in such functions. Field sports are no
doubt thing*; to be encouraged, but can it be necessary to have a picture
of a man nmning after a ball on every page of every illustrated journal
in this country? Let us hope for evolution in this as in all other things.
It is gratifying to obserVe that, since 1897, evolution has not been idle in
this matter. Within the last thirty years, our leading newspaper has seen
fit to acquire exclusive* rights in the primary publication of an Egyptian
tomb. Another daily newspaper has financed the excavation of a Roman
amphitheatre in Wales. Yet another has dug for prelmtory in the bed of
the Thames at Brentford, under arc-lamps at midnight. A well-known
illustrated weekly is frequendy first in the field in the announcement of
archaeological discovery. Pulvis et umbra are news-items, and the excava¬
tor is perpetually harassed by amiable young men and women with
probing pencils and cameras. In such circumstances, whether he vrills
it or no, it pays the excavator to give his courteous attention. A swiy
in any case in the making, and, if it is to have any sort of authenticity, its
fabricator must be led gently up the right sort of path. The press is not
always accurate and does not always emphasize those aspects of an
excavation which are scientifically the most important; but sympatheuc
help is the best corrective of these failings, and may be regarded as a
scientific no less than a social duty on the part of ±e modem archaeo¬
Long ago, G. M. Trevelyan remarked that ‘if histonans neglect to

educate the public, if they fail to interest it intelligently in the past, then
all their historical learning is valueless except in so far as it educate
themselves’.* Recendy, Mrs. Jacquctta Hawkes has been urging much
the same thing.
This [she remarks] is the cenniry of the common man. Just m m the
18th and 19th centuries archaeology was adding to the art coupons,
the architecture, interior decoration and fiimiturc of the wealthy an
> Clio, &c., p. 18.
aristocratic, so it seems that in the 20th we must take deliberate pains to
make it add something to the life of a democratic society. Our subjea has
social responsibilities and opportunities which it can fulfil through school
education, through museums and books and through all the instruments
of what is often rather disagreeably called ‘mass communications*—the
press; broadcasting, films and now television. If archaeology is to make
its contribution to contemporary life and not risk sooner or later being
jettisoned by society, all its followers, even the narrowest specialists,
should not be too proud to take part in its diffusion. I would go further
and say that we should not forget the problems of popular diffusion in
planning our research.*
Indeed, at the present time the public is prepared, nay eager, to meet
the scientist more than half-way. It is now up to the scientist to contri¬
bute his share. To do him justice he is not unaware of this duty. There
may yet linger in remote cloisters a few pedants of the old school who
will have none of this vulgarization^ but their mortality-rate is happily
The modem scientist increasingly recognizes the public as his
paraers. He spreads the empyrean at their feet; he serves up to them the
p^dples of evolution in simple language, sometimes spiced a little
with a topical tendenciousness. A monumental work entitled A Study of
History is a best-seller in ten volumes. Prehistoric Britain and Prehistoric
India are popular Pelican books. A scholarly account of the Pyramids of
Egypt finds thousands of readers. Of Digging up the Past a quarter of a
milhon copies have, I belie\"e, been sold. And so on. In this welter there
IS no secret refuge for the field-archaeologist. If he would hide, he must
hide in the limelight.
All this is as it should be. A present-day excavation must provide for
the Gmeral Public as a routine activity. Indirectly at least, that public
is paying for a good deal of archaeology, through the national or rate-
aided museums, through the Ministry of Works, through various Royal
Commissions, through the umversities, through archaeological schools
ovei^s in Italy, Greece, Turkey, Iraq, and Jerusalem, through the
British Council, and in a variety of other ways. In spite of taxation, it
still contributes also a little directly and personally as a testimony of its
practical interest. The popes and princes of the Renaissance have in
faa b^n replaced by the British taxpayer. However embarrassing its
attentions may be on occasion, the public is now in one way and another
our patron, and must be cultivated and suitably rewarded.
X ^ Prehistory*, in London and Middlesex Arch. Soc, Trans., N.S.,
192
PLATE XXI
B. ‘Copper Jar no. 2^^'
How no. .o illustrate! Two examples, with und«lines. from well-known publication..
{See pp. 176 oni /90 )
On all my major excavations I have accordingly made some special
provision for this incidental partner to our work. I may again quote
from one of my reports on a site in Dorset.
Under conditions of imobtrusive discipline, the general public were
deliberately encouraged to visit the site. Notices directed the visitor’s
approach from the nearest main road. He was told (by notices) where to
park his car and where to apply for information. Throughout the excava¬
tions it was the duty of an official guide-lecturer either to explain the work
to visitors or to organize reliefs of student-lecturers who, for regulated
periods, undertook this task, which, incidentally, provided for the
students in question an admirable training in clear thinking and siniple
exposition. The public was not charged for these services, but was invited
to contribute to the cost of the work—a system which is in practice both
more democratic and more productive than a fixed tariff. And, finally,
a well-stocked postcard-stall is as popular as it is profitable. Picturc-
postcards of the site can be produced [or rather, could before 19391
a cost of litde more than a half-penny each and will sell readily at tw^
pence each. Interim reports of the work, produced at fourpence each, will
seU at one shilling each. [Approximately 64,000 postcards and 16,000
interim reports were sold at the site in question.]... In such midtifarious
ways can the present-day pubUc be drawn to contribute directly or
indirectly to archaeological research.
I would particularly stress the value to the archaeologist himself of
speaking to and writing for the General PubUc. It is not difficult to be a
spedaUst, to write fairly inteUigibly for two or th^ fcUow-spedahsK, to
produce ‘a preparation of opium distilled by a minority for a minonty’.
I know a distinguished archaeologist who claims that he wntes for five
people; most of us are less ambitious. And as spedahsts we tend to
velop a sort of professional jargon which is a deterrent to a wider
audience and ultimately a handicap to the spedahst himsdf. I have
already in this chapter quoted from the proceedings of the British Asso¬
ciation, and I am now reminded of a relevant presidential address at an
earUer meeting of the same eminent body. At to meeting attention was
drawn to the plague of pedantic verbiage which had infested nicmem
sdence, and a plea was made for simpUfication and dassificawn. That
plea was a timely one; it might fittingly have been extended frona pro¬
fessional sdence to professional sport, to the dnemato^ph mdustry
and to professional joumaUsm in general. The danger of all to )argon,
• R. le Galliennc, Prou Fancits (1895). P- 8»> »pe«>d°8 Poetry. but singu-
lariy appropriate to much archaeology I
o
M70 193
at any rate in science, is not merely that it alienates the ordinary educated
man but that it is a boomerang liable to fly back and knock the sense out
of its users. In the words of the infallible Quiller-Couch, ‘If your
language is Jargon, your inteUea, if not your whole charaaer, wiU
almost certainly correspond. Where your mind should go straight, it will
dodge: the diflSculties it should approach with a fair front and grip with a
firm hand it wiU seek to evade or circumvent.’ I have been turning over
the pages of an exceUent journal which makes it its business to present
the results of scientific archaeology to the general public, and my eye
has fallen upon three articles by three of the most eminent archaeologists
o the day. On one page I am caught up in the hyphenated tongue-
twister ‘leaf-shaped-sword-culture-complex’; on another I am invited
to consider ‘the diagnostic value of negative lynchets’; on a third I am
informed that cenain place-names ‘were left by the equestrian inhuma-
tors who brought in the later Hallstatt culture’. (One almost expects to
turn the page and find a reference to ‘tram-riding cremators’.) An excel¬
lent friend of mine, in an attempt to distinguish between the significant
M the accidental aspects of megalithic tombs, has recendy brought to
binh two hideous monstrosities, the twins genomorph and phenomorph.
I pray m all friendliness that they may be short-Uved. Yet another of my
friends has attempted to make the fairly simple phrase ‘historical ap¬
proach easy for us by defining it as ‘the endeavour to achieve a con¬
ceptual mtegradon of individual phenomena in terms of specified time
and space’. For a less academic parallel to this sort of thing I need look
no further than the newspaper which lies beside me as I write: on its
ont is a quotation from a diplomatic manifesto in which the
Umted States reafl5rms ‘its stand against unilateral cancellation of con¬
tractual relationships and actions of a confiscatory nature’—lovely
phrase! Well, well; so one might go on; I have in fact quoted relatively
inntôus examples of a widespread and malignant disease—far worse
might rradily be found. Admittedly, an advancing science is from time
to tunc faced with a genuine and rational need for new nomenclature,
n«v p *®g- let the need be met with restraint and circumspection.
One word IS not necessarily more economical than two, whatever may be
said elsewhere (m other contexts) to the contrary. On the other hand, one
word will suffice to describe the scientific argot of which I am speaking;
Md that word is Hokum, an alternative to Quiller-Couch’s ‘Jargon’.
Let us purge our wnting and our thinking of Hokum. It is an infection to
which the over-educated and the under-educated are alike prone. G. M.
194
Trevelyan once lamented a comparable failing amongst historians. ‘The
substitution of a pseudo-sdentific for a literary atmosphere in historical
drdes’, he wrote, ‘has not only done much to divorce history from the
outside public, but has diminished its humanizing power over its own
devotees in school and university.’*
A household remedy for this disease is not far to seek. It is to be found
in deliberate and periodical vulgarization. My advice to the aspiring
archaeologist is. Go and explain your ideas, young man, to the Much-
binding-on-the-Marsh Antiquarian Sodety and Field Qub. Shun all
that is comprised in ‘gobbledygook’, that wonderful American carpet¬
bag for wordy and woolly pomposity;* use language intelligible to the
local bird-scarer. Then at last you will begin to understand yourself
if you have aught to say. Let us not scorn the profanum vulgus. In faa
some of our greatest archaeologists and anthropologists have needed no
reminder in this matter. The dassic Romamzation of Roman Bntainy by
Haverfield, J. G. Frazer’s Golden Bough (until it became an encydo-
paedia), and more recendy Prehistoric Britain^ by Jacquetta and
Christopher Hawkes, and Archaeology and Soaety by Grahame Qark,
are outstanding examples of the scholarly approach to a wide public,
whilst Sir Leonard Woolley is a famous adept at the art. But perhaps for
the masterpieces of popular writing in the sdentific field as a whole we
have still to look to men like J. H. Jeans or J. B. S. Haldane or Julian
Huxley, or ultimately to Charles Darwin himself, whose strong, simple
prose thrust down all barriers between minds and men. There was no
Horauan sniff about Darwin.
Terse, vigorous, direct prose is ultimately a matter of the sensibility
of the writer, but the smallest seed of it can be cultivated. I know cap¬
able archaeologists who are yet so myopic as scarcely to read a book
outside their tiny ‘subject’. And I find that Mr. St. John Ervine has
observed the same phenomenon: ‘Archaeologists’, he remarks, are odd
fish addiaed to periods, and unwilling to take interest in anything
outside their own speciality.’ They have ceased, or have not begun, to
cultivate their garden; at best, they are concerned only with a cabbage-
patch. What can they really comprehend of works and days, or how ex¬
press their comprehensions? It is a truism that words create thoughts,
only less than thoughts words. And the creation of words is no unskilled
* Clioy p. 25.
* On gobbledygook see E. Gowers, ABC of Plain Words (H,M. Suuonery
Office, 1951), p. 57.
195
job. ‘You will not get there*, says the infallible ‘Q*, ‘by hammering away
on your own untutored impulse. You must first be your own reader,
chiselling out the thought definitely for yourself; and, after that, must
carve out the intaglio yet more sharply and neatly, if you would impress
its image accurately upon the wax of other men’s minds. We found*, he
adds, that even for Men of Science this neat clean carving of words was a
very necessary accomplishment.* Yes, words are uncommonly important
things, even for Men of Science. And in the end, the vulgus is no bad
judge. It is the duty of the archaeologist, as of the scientist, to reach and
impress the public, and to mould his words in the common clay of its
forthright understanding.
POSTSCRIPT TO CHAPTER XVI
With reference to pp. 189-90 above, the following notes are added on the
three principal processes of pictorial reproduction. By way of preface it
may be observed that line blocks, when not exceeding the dimensions of
the text-page, are printed with the text and are normally numbered as
text-figureSi with Arabic numerals. Half-tone blocks and lithographs are
usually printed separately from the text, on higher-grade paper, and are
normally numbered as plates^ with Roman numerals. Line blocks of
larger dimensions than the text-page are also treated as plates.
(a) HALF-TONE BLOCKS
Wash-drawings, photographs, paintings, &c., are ‘continuous tone*

subjects. For purposes of reproduction, such pictures have to be broken
up into half tones (or broken tones), so that the final result is a printing-
block made up of a series of dots, varying in size with the relative light
and shade of the copy. This necessity arises because half-tone blocks are
printed in exaedy the same manner as type. A printing-roller charged
with printers ink passes over the surface of the block and by a ‘kiss*
touch imparts ink to whatever it contacts.
In order to produce these dots for inking, the process includes the re¬
photographing of the original through a ruled screen interposed in front
of the photographic plate. The screen consists of lines ruled on opdeal
glass to form square apertures with opaque dividing lines. These break up
the continuity of the subject into dots of varying sizes, the high lights
being represented by exceedingly fine dots, the middle tones by larger
dots, and the deeper tones by stiU larger dots. From the negative so
obtained a copper plate is made: this is done by coating the copper with
bichromated glue and printing it in contact with the negative: the inter-
196
Overloaded photographs of unprepared subjects.
pp. 17S and 190.)
PLATE XXII
I
i
sections between the dots are eaten away chemically, thus lea\^g the dots
in relief. This copper plate is then mounted on wood and printed just as
type is printed in a letterpress machine.
The size of the dots is controlled to meet various conditions of printing.
Coarse newspaper-printing calls for a screen-ruling with approximately
6o lines per square inch, ruled at right angles to another 6o lines, gi\^g
intersections of 60x60 = 3.600 dots per square inch. For magê-
printing and such like, a screen-ruling of too to 120 is mostly used; for
commercial work, 120 to 133 lines per square mch; while for
printing, such as that of scientific subjects prmted on high-grade art
paper, the screen-ruling is generally 133 to 150.
A printer may spoU a good block by bad inking-by the use of tw much
or too little ink, or by uneven inking. Only an experienced and staU^
printer will get the best out of a block, and it is sometimes preferable to
the blocks printed by the blockmaker himself, if he has the necessary
trained staff. Again, a block may be spoilt by the use of
At its best, a half-tone block can be very good, but a second-rate block or
AeTecond-rate printing of a good block will very seriously mar any pubh-
cation, however brilliant its other matter may be. Careful scrutmy m
proof-stage is essential. . . ., 1 u
^ There are certain points in which the author must help the blockmaker.
Fi^t, gtêd photoiaphic prints are best for repletion: matt pimtt,
‘a^ ies’,Ld toned prints should be avoided. Good glossy bUck-and-
having regard to the scale of reductton desired. Remember
197
the text and half-tone illustrations are inserted, usually on special paper,
as plates (not figures). See above.
ip) LINE BLOCKS
^ reproduced by the line-block process-

ê- I photographed by the blockmaker in a special
of^Sh^, f?' “ transferred on to a zinc plate, the ‘background*
^lirffOOsuhtn" hath, leaving the black lines in
rehef for subsequent inking and printing. The zinc plate is then nailed or
fine hib h recessed portions. The
line block, being printed in exactly the same fashion as letterpress can
SeTe peÔ^^.^" simultaneously with the latter, prodded’
onO^h^^’ “ reduction gives satisfactory results, but

one-half or one-quaner (rarely smaller) may have to be used. In the case
4X ^ drawing 12x6 inches will come out at
4x2 mches and will therefore actually be only one-ninth of the area of the
OSom ““ ^ 3pecif,^n^?of
thi memrfilo'iO! T W* drawing, shall know

toe extent of proposed reduction, and shall so be enabled to regulate the
^ckness of his lines, the size of his lenering, and his general style accord-
to the a« of •*'* ^ “ manufacture and damage

pnntmg. Lanes drawn excessively close together will tend to
from one Ime to the other, fomung an untidy and unexpressive blot This
js p^cuWiy happ,„ esS^toe c^-
shdlê^toW S d •* individual lines of the hatching

siXlv to^ nl^ o a T And nothing is more un!
S fi 1^ diagram than lettering which cannot be easily read
o"^Z“Xir;r“'r'^
In sele^TcT,!' ^ sm<dl.
A maximum size for a line block mav k,. Tou
h toi'!J^r*Io lithography, andli
198
is less costly than a large line block. Small and simple corrections can be
made on a line block but are very imdesirable.
(c) LITHOGRAPHS
The old method of lithography—printing from drawings on plane-

surface stone—has given way to modem processes of photo-lithography
in which the map or plan is reproduced first as a photographic negative
and then printed therefrom on to a thin sheet of zinc. By chemical pro¬
cesses the image is made to attract ink and the clean portions of the zinc
to repel ink. Such a plate is held taut on a machine printing-cylinder. Ink-
rollers charged with litho-ink and damper-rollers charged with water
alternately pass over the entire surface of the plate. The inked image is
pressed on to another cylinder around which a sheet of special rubber is
stretched and therefrom on to the paper, hence the term ‘offset’ or ‘litho-
offset’ or ‘photo-litho-offset’.
The advantage of the ‘offset’ process is that fine line-work can be
printed by the resilient ‘kiss’ touch of a pliable rubber sheet so that all
imdue pressure is avoided, and the inked lines are pressed gently into ê
texture of the paper without any image being visible on the revewe side
of the sheet. Corrections, even minor corrections, should be avoided so
far as is possible. . . j u
A lithograph thus requires careful separate pnntmg and cannot be
struck off with the text. ^
Lithographic plates may normally be made up to 40 X 30 mches.
199
XVII
What are we digging up, and why?

*No amount of technical knowledge can replace the
comprehension of the humanities or the study of
history and philosophy.* wins ton s. Churchill,
reported in Kebenhavns Universitets Promotionsfest
T he
den Jookt, j^$o (Copenhagen, 1951).
previous chapters have touched upon the history of archaeo-

logical excavation, upon the search for an absolute chronology as
the ultimate basis for the ordering and interrelating of our data,
upon the stradgraphical method as a contributory procedure, upon the
need for long-term planning if we are to secure the systematic advance of
knowledge, and latterly upon the very vital questions of publication and
publicity. A little has been said also of the actual technique of digging,
of recording in the field, of laboratory-work, and of staff. But there is one
overriding aspea of our task which cannot be passed by in a concluding
chapter; and that aspea may best be expressed by the question, ‘What
does it all amount to?* What are we trying to do in this rather complex
fashion, and how far can we hope to succeed? Any answa to this ques¬
tion is inevitably subjective and prejudiced, but ±ere is no great harm
in a little honest prejudice. It may at least stimulate the wiser judgement
of those happy critics who are devoid of bias.
Archaeology is primarily a faa-finding discipline. It has indeed been
stated by an American writer that ‘Archaeology per re is no more than
a method and a stt of specialized techniques for the gathering of cul¬
tural information. The archaeologist, as archaeologist, is really nothing
but a techmdan.*' I have no hesitation in denounring that extreme view
as nonsense. A lepidopterist is a great deal more than a butterfly-catcher,
and an archaeologist who is not more than a potsherd-catcher is un¬
worthy of his logos. He is primarily a faa-finda, but his facts arc the
material records of human achievement; he is also, by that token, a
humanist, and his secondary task is that of revivifying or humanizing
his materials with a controlled imagination that inevitably partakes of
the qualities of art and even of philosophy.
But these are after all mere words. What in faa docs this thing,
* W. W, Taylor, as dted, p, 43.

200
WHAT ARE WE DIGGING UP, AND WHY?
Archaeology, really amount to? The question is one which, in our

evaluation of evidence, can never be far from our minds. No matter that
in a majority of instances we cannot truly know {it is the privilege of the
intellênt to ask questions, and the claim of the unintelligent to have all
tfie^nswers.^The state of philosophic doubt is not the least enviable of
human conditions. Definition does not necessarily clarify.
Here at the outset we are confronted with elements of conflict. We
have on the one hand the technician or, if you will, the ‘scientist’, busy
wth inches and analyses and smudges in the soil; busy with the anatomy
of history or prehistory. On the other hand we have the humanist busy
with its vital interpretation. There is a widespread fashion today to in¬
cline towards the former trend, to regard archaeology as a natural
science and to discount the intervention of ‘motive’ and ‘free will’.
‘Motives’, says Professor Gordon Childe, ‘are in fact hardly capable of
genuine historical study.’ This trend is a useful reaction from the
romanticism of a past century. But it can easily be carried too far. I have
remarked elsewhere upon a tendency to devolve archaeology into a sort
of dehydrated humanism, to miunmify the past, to transform our pre¬
decessors into ‘battle-axe folk’ or ‘beaker folk’, imdl, by an instinctive
and forgivable reaction, we begin almost to personify battle-axes or
beakers with a sort of hungry latter-day animism. Such phraseology
represents a tendency that can only be deplored. It nms deeper than its
exponents sometimes realize. However broadly we use the words, man
is in some sense the casket of a soul as well as five-shiUings-worth of
chemicals. And the soul or sensibility or mind—^whatever we choose to
call it—îs beyond the reach of finite intelligence, since the mind ob¬
viously cannot encompass itself. Within the far-off limit of ultimate
causes, the geologist can encompass or objectify the rocks with which
he deals; not so the humanist with the intellect. There in the last resort
all is subjective, it cannot be otherwise. Archaeology increasingly and
very properly adapts and adopts the methods of natural science and
unblushingly seeks its aid. It is not on that account itself a science in the
class-room meaning of the term. At the best it is a very inexact science.
But perhaps for that reason its demands upon the constructive imagina¬
tion are more immediately insistent than are those of some of the more
self-explanatory class-room sciences. For that very reason, be it re¬
peated, the archaeologist is a great deal more than a rather sujjcrior
laboratory-assistant. He is also something of an artist. O. G. S. Crawford
was near the mark when he affirmed that ‘Archaeology is an art which
201
employs a scientific technique’. Or, as the Oxford historian, Sir Llewellyn

Woodward, remarks, ‘Historical understanding is more than a series of
detective tricks. It requires a mind already attuned to the scale of human
action and practised in the subtlest use of language to express the depths
and heights.’ That is well said. The historian, and with him I group the
archaeologist, must have a spark of the intuitive comprehension which
inspires the painter or the poet. ‘The highest reach of science’, pro¬
claimed Matthew Arnold, ‘is, one may say, a faculty of divination, akin
to the highest power exercised in poetry.’ All great historians have some¬
thing of this4>ower, and owe their greatness no less to it than to their
scholarship^They make the past live because they are themselves alive and
can integrate their reasoned facts with the illogicalities of life. Other¬
wise they were mere cataloguers, adding dust to dust and ashes to ashes.
As archaeologists, then, we are at the same time coUeaors and inter¬
preters. The obvious next question is, What do we collect and seek to
interpret? The question lands us at the outset in a minor quandary from
which cscap)c is urgent. Throughout these chapters the term archaeo¬
logy has been used in the widest possible sense, including equally the
study of eolithic choppers and of Viaorian gas-lamps. Others are, I am
afraid, sometimes less catholic in their usage. The French appear to
have evolved a hierarchical distinction between Varchiologie and la pri-
histoire that is subtle enough to escape the average foreigner, but we
have our British counterpart. From time to time one hears the terms
‘archaeologist’ and ‘antiquary’, or even that hideous and unnecessary
pseudo-noun ‘antiquarian’, used with a sense of divergence significantly
akin to that of ‘sheep and goats’, or ‘chalk and cheese’. The antiquary,
it seems, is the more genteel of the two; he sits in a chair and uses a quiz¬
zing-glass, or in moments of supreme afflatus crashes upon his knees and
rubs a brass. The archaeologist, on the other hand, wears corduroy
shorts, strides about on draughty landscapes with a shovel and an odorous
pipe, and is liable to be an undergraduate. To these divergent types, might
be added a third, the anthropologist, vaguely interested in flagrantly un-
British ‘natives’. Of course all this dichotomy or trichotomy is nonsense;
but there does lurk behind it a nucleus of actuality of a not wholly desir¬
able kind. The common tendency to discriminate archaeologists as pre¬
historians and antiquaries as medievalists docs good to nobody. If any¬
thing, it attempts on the one hand to rob j)rchistory of a little of the
humanity that comes more easily to the Middle Ages; and on the other
hand to deprive medieval studies excessively of the cold and calculating
202
objectivity that is attributed to the prehistorian. Recently, after training

the young members of the staff of one of our Historical Monuments
Commissions on a typical prehistoric site, I was glad to see them pro¬
ceed with the excavation of a medieval site by the identical technique,
with fruitful results ranging in period from the eleventh to the seven¬
teenth centuries. And yet how rarely has that simple and obvious
procedure been attempted! Let it be agreed that the two words ‘archaeo¬
logist’ and ‘antiquary’ shall in future be exactly synonymous, rooted in a
common discipline and striving by the same or closely similar methods
to the same end.
But what, when all is said and done, is that end? We have just agreed
(I hoi>e) to work as brothers, but what of the nature of our task? Let us
for a moment consider its range, with specific examples. In a single
year we have had the (former) Disney Professor of Archaeology at
Cambridge busily and successfully exploring the Upper Palaeolithic in
France and Dr. Leakey sweating after his pleistocene industries through
tropical Africa. Professor Grahame Qark in Yorkshire was extracting
with consummate skill the remarkable relics of as squalid a huddle of
marsh-ridden food-gatherers as the imagination could well encompass.
Sir Cyril Fox, of science and imagination all compaa, was, as recalled
in a previous chapter (p. 3), reconstructing the mumbo-jumbo of
miserable Bronze Age barrow-burials on the ultimate fringes of the
ancient world. The writer was digging into the first foothold of wretched
Iron Age immigrants on the English south coast and classifying their
incompetent dog-biscuit potsherds. Mrs. Stuart Piggott, with an in¬
spiring fortitude, was ransacking the vacant windswept vestiges of first-
century hill-forts in the Scottish Lowlands, buoyed by the thought that
it is better to travel hopefully than to arrive. Professor Ian Richmond,
on Hadrian’s Wall, was expending the profoundest learning and an in¬
comparable astuteness upon the uncovering of a medidne-man’s go-
down. The list might easily be extended. But whither is it getting us?
One may sometimes find oneself in two minds about it all. Is all in¬
formation worth while, or are we justified in selecting and grading our
material on a basis of priorities? And, if so, what priorities?
This question might easily lead us into a prolonged discussion of a
semi-philosophical nature. The temptation may be resisted without
complete evasion of the issue. At the outset it is sufficiently clear that no
single or simple answer is adequate. Much depends upon the angle of
approach. Man is an animal with a biological tree. He is physically an
203
unspccialized animal, but with a specialized brain which enables him to

elaborate and amplify his physique by means of an increasing range of
artifacts or artificial limbs. He thus becomes the author of industries,
which accumulate into cultures: until ultimately he is sufficiently
equipped to live in large communities, or in other words becomes
civilized. In a large community the constant clash or collaboration of
brains leads, under balanced conditions, to a great civilization; under
unbalanced conditions, to a proportionately great human catastrophe.
The pjeriod of balance may be brief, as in fifth-century Greece, or long
as in D5mastic Egypt. The poriod of unbalance may be one of oblitera¬
tion and oblivion, or it may be one of patching and reshaping, of constant
if unequal striving towards a new goal. Such sequences and possibilities
are sufficiently familiar and need not be particularized, if only because
Mr. Arnold Toynbee has displayed them to a wide public in terms of
his own stimulating philosophy of history.
This elaborate process manifesdy offers to the student more than one
line of approach. In faa, modem archaeological research displays a
recurrent duality. The normal approach of archaeology in Great Britain
has in the past been along the lines of the classical tradition. The Grand
Tour and the standard classical education have never, until recent years,
been far fmm the mind of the British archaeologist; and names such as
those of Stukely or Sir John Evans or even Pitt Rivers hiTn«;glf do not
out-weigh the average truth of that statement. Greece and Rome were
the mecca of the British student, provincialized a little in latter years by
the great influence of Haverfield. For example, the Yates Chair of
Archaeology in the University of London, in spite of its generalized
designation, has always been interpreted as a chair of classical archaeo¬
logy. And long ago we find the young Arthur Evans inveighing against
the same bias at Oxford. Evans in his twenties was urged to apply for
an archaeological studentship which had just been established in the
University. His immediate reaction was: ‘One feels that what is wânted
is a student of “Classical Archaeology” and that anyone who wasn’t
would probably have scant justice done him at Oxford.’ He went on to
protest that ‘the great characteristic of modem Archaeological progress
has been the revelations as to periods and races of men about which
history is silent.... Oxford, however, seems to have set itself to ignore
cv'ery branch of Archaeology out of its own classical beat.’* Evans’s
* Joan Evans, Time ond Chance! the story of Arthur Evans and his Forebears
(London, I943)» PP- 221-2.
204
youthful protest was in 1879. Four years later his mind was turning once
more towards Oxford, and Dr. Joan Evans has recorded an entertaining
and significant correspondence between him and Freeman which further
defines the contemporary attitude towards archaeology in these islands.
There is going to be established a Professorship of Archaeology, [wrote
Evans] and I have been strongly advised to stand for it. I do not think
I shall, unless I see any real prospect of getting it: and to say the truth
I see very little. To begin with, it is to be called the Professorship of
‘Classical* Archaeology, and... to confine a Professorship of Archaeology
to classical times seems to me as reasonable as to create a Chair of Insular
Geography’ or ‘Mesozoic Geology*. . . . Europe, except of a favoured
period and a very limited area (for I take it that neither Gaul, Britain or
Illyricum were ever ‘classical* in Jowett’s sense) is to be rigorously
excluded!
Freeman’s reply to Evans was,
I think you should stand, if only for a protest. ... Of course they will
have some narrow Balliol fool, suspending all sound learning at the end
of his hooked nose, to represent self-satisfied ignorance against you, but
I would go in just to tell them a thing or two.*
The upshot was that Arthur Evans went his own way and discovered im-
mortahty in Crete. Not indeed until 1926, when the Abercromby Chair
was established in Edinburgh, was there a real professorship of pre¬
historic archaeology in Great Britain^ and, if we ignore the purely per¬
sonal and honorific appointment eventually extended to Arthur Evans
(in 1909) as ‘Extraordinary Professor of Prehistoric Archaeology’, only
within the last few years has Oxford so shaken free from the old Jowett
tradition as to follow suit with its own substantive chair of European
archaeology. This is remarkable, but such is the fact.
In Scandinavia, on the other hand, the situadon has been exactly the
reverse. There not merely prehistory in general but Scanc^vian pre¬
history in pardcular, to which much other European prehistoty is very
greatly indebted, held the field from the beginmng of the nineteenth
century onwards, and it is only in recent years that classical archaeology
has achieved a recognized academic status; for example, by the founda¬
tion of classical chairs at Lund and Uppsala about 1910. In Scandinavia
the undisguised parish pump has been respectable for a century and a
half; in Western Europe as a whole it has, until recent years, been scarcely
tolerated save under a classical pavihon.
* Time and Chance, pp. 261-2.
205
Today a great reaction is in progress. Swedes have become classicists

and are digging up the Roman Forum; Britishers, careless of their
classics, are digging up mesolithic forest-scavengers in the mists of their
own countrîde. Indeed at the present time, save in the dilute form of
Romano-British archaeology and apan from some notable work in
North Africa by the British School at Rome, classical archaeology is here
largely under an eclipse, whilst prehistory flourishes as never before.
Even the modem Romans themselves have brazenly uncovered the post-
holes of some disreputable prehistoric huts within a few yards of the
respeaable House of Livia on the Palatine. Old values are being
shuffled and new ones are emerging. What is their relative significance?
For example, is it really worth our while to expand money and talent
upon an entirely empty stone hut on a Welsh mountain, whilst great
sites on the anerial routes of human development remain scientifically
untouched? True, the one is nearby and costs a few pounds, the other
may be thousands of miles away and will cost thousands of pounds; but
such opportunist factors are not for the moment at issue. The question is
rather one of the principle upon which we are going to work, of our
sense of proportion.
Once more, it is, one may suppose, all a matter of our line of approach.
Much depends upon whether our starting-point is that of the biologist
or that of the humanist. It has been argued by the biologist that human
insututions are governed by the same kind of natural laws as is the de-
\clopment of the human body; that humanity and human institutions
may properly be studied in much the same way as pigeons or earth¬
worms. It is the developmental process itself that focuses the inquiring
mmd. Achievement, good or bad, is an element of processional change
Md IS of no special intrinsic interest; it is a register of trial and error in
the stumbl^ progress of evolution. Indeed the words ‘good’ and ‘bad’
^ subjective misnomers. Who are we to distinguish between them?
They are arbitrary grades in our yardstick of change; we might as weU
say that a fish is ‘good’ and a jellyfish ‘bad’,
Adnunedly the biological attitude is here stated crudely and perhaps
unfairly. But Oswald Spengler, who has done not a little to lend this
atutude the combined prestige of philosophy and history, states the
matter m no very different style, and Professor Henri Frankfort has
recently commemed upon him in terms essentially parallel with my
own. Spengler, remarks Frankfort,
actuaUy calls civilizations ‘living beings of the highest order’, and he
206
undertakes to state with precision which phenomena characterize each

stage in their life-cycle. For him, an imperialistic and socialistic order
follow a traditional and hierarchical society; expanding technique and
trade follow greatness in art, music and literature as certainly as the
dispersal of the seeds follows the maturing of a plant which will
never flower again. But to take the biological metaphor literally,
to grant in this manner reality to an image, is not morphology but
mythology; and it is belief, not knowledge, which induces Spengler to
deny the freedom of the spirit and the impredictability of human be¬
haviour.*
I too am of those, then, who, with all proper respea to Spengler and
his kind, are not over-readily tempted to equate the development of
human institutions with the normal processes of organic evolution, to
Darwinize human ‘progress*. As Julian Huxley long ago observed,
‘Numerous writers—largely because purely biological are simpler than
human phenomena—have been obsessed with the idea that the study of
biology as such will teach us principles which can be applied directly and
wholesale to human problems.** The tendency is doubtless in part a
reaction from the Book of Genesis, but needs a more serious apologia
than that. Organic evolution and social evolution are not, in the present
or any forseeable stage of research, equivalent processes. Doubtless
a remote eye, with a comprehension beyond that of the mere homuncidusy
would integrate organic and intellectual development and sec them both
as facets of the same crystal, cogs of the same machine. But to the close-
up, myopic view with which we must content ourselves, the differences
between the two processes arc more significant than the resemblances.
Professor Gordon Childe has recently had wise words to say on this
matter. Speaking of the nature of cultural changes, in faa of‘progress’,
he remarks:
Inventions can be transmitted from one sodety to another [by diffu¬
sion]. But that is just what is impossible in organic evolution. By no
possible means can one species transmit to another the mutation which
has proved beneficial, even though both inhabit the same region. All that
can happen is that natural selection gradually eliminates the species that
lacks the mutation. It is, I suggest, the operation of diffusion more than
anything else that distinguishes social from organic evolution and ex-
* H. Frankfort, The Birth of Civilization in the Near East (London, 1951),

pp. 18-19.
* Essays of a Biologist (London, 1926), p. 75.
207
WHAT ARE WE DIGGING UPj AND WHY?
plains the curvatures of the lines in any graphic representation of the

process.®
No doubt the old postulate of universal and identifiable and even pre¬
dictable stages in human progress was the wishful thinking of an age
which, both in the moral and in the scientific sphere, was strenuously set¬
ting its facts in order. Only, in the scientific sphere at any rate, its facts
were, of course, inadequate. They consdnited a sort of half-knowledge
which, as is the way of half-knowledge, has usurped an authority out
of all ratio with its intrinsic worth. It is a paradox that in an age when on
the one hand the rights and prerogatives of man have been assencd as
never before, on the other hand the scientific urge towards system and
sequence has tended to put him into a queue or * crocodile’. Must we
queue up for everything, even for our humanity? Are we not all, even
om scientists, getting excessively queue-minded? I am indeed at one
with G. M. Trevelyan when he says that, 'even if cause and effect could
be discovered with accuracy, they still would not be the most interesting
part of human afiaks. It is not man’s evolution but his attainment that
is the great lesson of the past and the highest theme of history.’^ There
speaks one with a surviving belief in that archaic phrase, the Nobility of
Man. We need not dose our eyes to Man-the-Jelly-fish or Man-the-
Whole-time-Food-gaihercr in order to belietê in Man-with-Time-to-
think-between-Meals, in Civilized Man, but the last is, surely, of
ovemding unportance. OviHzation has been defined as 'the aggregation
of large populations in cities; the differentiation within these of primary
producers {fishers, farmers, &c.), fiill-time specialist artisans, merchants,
offidals, priests, and rulers; an effective concentration of economic and
political power; the use of conventional symbols for recording and
transmitting information (wndng), and equally conventional standard
of weights and measures of time and space lading to some mathematical
and calendrical sdence.’^ What a ripeness of human mind and effort all
that implies! Man in his most fully expressive phase. As long ago as
1852 the president of the Archaeological Institute could pronounce with
some show of reason that in archaeology
it can no longer be assumed that the obscurest periods are the most
wo^y of investigation. Those, on the contrary, should be preferred
which are richest in the materials intrinsically deserving of study; that
* Social Eiâlutian (London, 1951), p. 170.
* Clht &c., p. 12.
* Childc, op. dt., p. i6t.
208
is, in the visible development of the human intellect, the display of

personal character, the creative activity of the arts, the variety of the
social relations, and the analogies or contrasts which these may present
to life amongst ourselves.*
True, a modem supporter of that view lays himself open to an easy

charge of atavism, and of choosing as his subject the kind of Man who
has had most to say, has expressed it most amply in material things, and
is therefore the most susceptible subject for the archaeologist. But I
think not. After those words had been written, I heard Sir Llewellyn
Woodward, whom I have already quoted, saying much the same thing,
though in better shape. He was speaking of certain scholars who had, he
said, regained for history a place among the Muses, affirming that they
did so * ultimately because they set a high value upon the dignity of
man’. T repeat this term deliberately,’ he added, ‘because one of the
signs of disintegration in our own culture is an imwillingness to con¬
sider that man has dignity, that his acts may be noble. Once this con¬
ception of nobility is lost, history becomes nothing more than a rag-bag,
a pawn-broker’s catalogue, or at best a psychiatrist’s case book.’* To
this it may be countered that ‘nobility’ is not of necessity absent from
the savage, and that for example a Chatelperront point or a Solutrian
lance-head may represent an achievement which partakes of the quality
of intellectual nobility. But I am not prepared to admit the Noble Savage
within my general definition of the term, for the simple reason that he is
a savage, suffering from a savage’s restriaed vision, tangental reasoning
and lack of opportunity. I have in mind something far more complex and
comprehensive; something in fact which implies the background of
civilization or some approximation to it; where the intelligence has been
subjeaed to the widest possible range of stimuli and where its firuits
have been most widely, quickly, and intelligently shared.
There may be less hesitation in emphasizing this approach, since
there is, it seems, a likelihood that it may receive less than its due
patronage from the archaeological student of the future. It seems prob¬
able that the humanistic approach to the study of antiquity may from
now on give way increasingly to what has above been crudely dubbed the
biological approach. That change is due in part to the fault of the human¬
ists themselves. Until the citizens of Minos and of the Indus valley
* E. Oldfield in Arch,Joum, ix (1852), 3.

* Froc, of the British Academy^ xxxvi (1950), 112.
6570 209 P
arrived wth their unread scripts to complicate the scene, a normal

approach to the great civilizations of the ancient world—those of Greece,
Rome, the Nile, Palestine, the land of the Twin Rivers—was through
the study of language. Archaeology suffered from a linguistic pre¬
occupation, very necessary and valuable in itself but liable to obscure
more material and equally valuable aspeas. For it would appear to be
a truism that the preoccupying study of language is rarely in practice
compatible in one and the same individual with the scientific, analytic
study of phenomena as they present themselves in the earth. The linguist
is, in my experience, a different sort of man from the excavator or the
methodical student of cultural evidence in the wider sense. He is liable
to be Litde Johnnie Head-in-Air, and we all know what happened to
him when he came to a hole in the ground. The consequence of all this
has been to isolate certain fields of study, notably the classical civiliza¬
tions, as preserves of a particular class of humanists, who have tempered
their linguistic and historical training with art-criticism but have rarely
got down to earth and studied the stuff of which their chosen civiliza¬
tions are composed, or out of which their civilizations have grown. For
them, Pitt Rivers has lived in vain. And now their own days appear to be
numbered, and with a perverseness which I do not attempt to conceal I
begin to lament their passing. Today the study of Greek is very definitely
on the decline and even Latin is under a cloud. We are in a period of
speedy transition. The most eminent of our materialistic archaeologists
today can, and does, read the odes of Pindar after dinner, but he is an
elderly gentleman and it may be doubted whether there are more of his
kind. The time has come to ask. What is replacing the traditional dis¬
ciplines? A great variety of skills and techniques, but comparatively
little that can educate and stimulate the humanistic imagination. For
this vital quahty is, it would seem, only in part innate. It is liable in most
of us to be a tender growth and needs careful nourishment. It needs the
sort of nourishment that reflective literature of the highest quality, with
an historical colouring, can best supply. It needs something equivalent to
a classical education, and it is difficult to say what the suitable equivalents
are. Whatever they be, they are urgendy necessary if we are now to save
archaeology from an oven^ helmingly biological bias.
In the past few moments the argument may seem to have turned a
circle. It began with regret for the stranglehold of the linguisdc tradidon
on certain fields of archaeological study, and ended with regret for the
decline of that tradidon. In actual fact I am once more regretting the
210
prospea of archaeology passing wholly into the hands of the biologist

and the technician, just as much as I have been regretting the older and
now doomed monopoly of the linguist and historian. There is today a
genuine risk of a new severence between humanism and science in these
matters, just at the time when a closer integration of the two is feasible
and necessary. The co-operation of biologists and geologists and botanists
and physicists in our research is a welcome portent; there is more cause
for doubt when we find our research passing increasingly to those
whose main education has been in these fields of natural science. Man,
we may be forgiven for recalling, is something a good deal more to us
than an ingredient in the chemistry of the cosmos; and a course of
poetry or philosophy may properly be regarded as no less needful for the
young archaeologist—or the old one for that matter—than a course of
pot-making or pollen-analysis.
This part of the present chapter may be closed, then, with the truism
that Man is not only the author or veWcle of a culture-trend, he is also
a personality. The struggle towards civilization has been the struggle
towards the fuller development and the more ample expression of that
personahty. The corollary is that a fair proportion of our effort should
be expended upon the exploration of sites and regions which are likely
to reflect the major achievements of civilization. At present we are not
doing that. Half a century ago the situation was difierent. In 1908
Hadrian Allcroft could write: ‘Characteristically the English, who have
done so much for the Hittite, the Aiinoan, and the Egyptian, have as yet
scarcely concerned themselves to apply the same methods to the secrets
of their own soil.’ Today the position is reversed. Today we are per¬
haps excessively content with those little domestic secrets: year êr
year we devote in our island an abounding skill and enthusiasm to the
huts or graves of the uttermost rejects of the ancient world. True, we
add thereby some particle to the sum-total of human knowledge. But
what does it all really amount to? How much does it really matter} What
of the great world beyond? I was standing not long ago 5,000 miles
away from here, in the steppe of Turkestan, upon a tumult of mighty
mounds whither age after age came men from China, from the Mediter¬
ranean, from Ind, to exchange their goods and fertilize their ideas, to
express the ancient world socially and aesthetically in the most complex
polity then known to man. Could we but transfer a tithe of our scholar¬
ship for a few seasons to this great workshop of civilization—or indeed
to many others of the same high potentiality—what should we not gain
211
WHAT ARE WE DIGGING UFj AND WHY?
in OUT knowledge of human achievement? What could matter to us more

than that?
This final chapter, therefore, is primarily an appeal for a higher
measure of concentration, on the part of British archaeologists, upon
the riper achievements of Man as a social animal. It sometimes seems to
me that we are liable to start the race in fine form, but to run out half¬
way down the course. Let us for a change try to come in over the last
jump. There is no sort of doubt that we in this United Kingdom can
supply an initial field-training of a quality unsurpassed in the world.
Here, in Britain, is our training-ground; but where is our ultimate ful¬
filment? The great adventure still awaits us in a shrinking and increas¬
ingly regimented world; and surely adventurers were ever a British
export?
The word ‘adventure^ may be allowed to evoke a tangentai sentence
or two about a theme which is near my heart. The term ‘adventure’ is,
of ojurse, a relative one. Some, like the late Andrew Lang, are content
to find ‘Adventures among Books’, Others have to go to the poles or
climb Mount Everest to find them. Many so-called adventures are
entirely spurious. The other day I was reading a preposterous account
by a professional adventurer of a journey through the Khyber Pass to
Kabul, such as is made daily by countless greengrocers’ vans. It was
described as a journey ‘through the Pass of Death to the City of Brood-
mg Suspicion’. But, nonsense apart, I am firmly of the opinion that our
young men and women will lose nothing by a little real adventuring, and
their search as archaeologists for the footsteps of civilized man in Asian
or Afiican tracks will add an incidental stimulus of a kind which is
otherwise increasingly hard to find. It is now nearly thirty y^rs since
John Buchan wrote The Last Secrets^ heralding the end of discovery in
a urorld which now had nothing more to hide from m. Nevertheless,
until a very few years ago the young Britain was still enticed into the
paths of adventure by the worldly prospects of governing a tract of Asia
many times the size of England, or of trafficking with tribesmen on the
roof of the >vorld. Now, in 195 numbers of these avenues are dosed
by political change. Much of the potential adventure overseas has sud¬
denly gone out of our life, and, with it, that sharpening and shaping of the
chaiaaer which is a by-product of it, I am speaking soberly, not as a
romanticiat, Romance is merely adventure remembered in tranquillity,
devoid of the ills and anxieties, fleas, fevers, thirst, and toothache, which
are liable to be the more instant experience, I am commending first-hand
212
Reconsiruaion of an Early Iron Age homestead at Little Woodbury, Wiltshire.
iSe< p. 2t4.)
PLATE XXIII
WHAT A“RTE WE UFj AKD WHV?
adventure as a necessary ntedidne to die ctwracter of the

ardineology of tl« )dnd which I have been waamg before the w
fraught with the right sen of advtnmrc. If ever the iWdent, ^
recesses of his northern lauvcrtity, foci somnhmg of that '
«un tliat weariness of the north {aUtfatigm du mirJ) rfwhich Madame
de Slid wrote, with Vf'itickelmflnn in mind, 1« him iadtijfe « for ft u ' e
S^d dreTrl epporturiry lo do so. h will serve ulUtmtelj; « enhance
St P«per namSm. and meanwhile will haihoui him mutably,« ^
Milri^hummiist, in the lands of the south and the '
h^nanities and science and dvilnatiOB were ^cniMlvia brought
birth He will gaihcr no harm and possibly irmch good, ^
fTreatiiiiis to touch lighdy on the huge problem of *c Pinnate ^
tialiiy of our archucologital evidence. Whiihcr, ai
us? Wlwt, after oU, mattcis most in this dLsapbac of emrs? U wc
,o select oW ohiceti« viith a little more thought for essen^
fteccmly, in an jifti:r-dinner disciL^sioa on relative valnca iC was y
»od=*, -h.., ^ p™l™

ibc Trinity/ TteCnnvcrration thcnlap^d. I have snra
afrnh whai the value of tmr archaeological TOdeocc really ^ _ ^
our hits and pieces amonni to? Listen 10 the graw words which Dr,
John Donne Itteicd cn this subjcci three and a half centimes ago ■
The Mhes of an Osh in the Chimney are no epitaph of Hat
i_ VI k tarrt ihdir WM i it tclla nv: noT what flocks it ibcUcr*?*
loo; it say* nothing, it disringuirfie* As ™n

dust oTTwrercli whom thou wouldest not, of a pnoc* whom
Chimb, _ ^ (D picnuiincc. this i* the

?Srici?thls Uthe noble timii, sr«l this the yecimanly. «hii the Plebeian
bran?
to, ton .tonxO ll.« . P« m™ “f

™bbi.h-top. ArJ to » ptotoW
113
WHAT ARE WE DIGGING UP^ AND WHY?
to award the palm to the uokoowA Suim^nuii vrha wds buHcd at Ur with
sbr^-three hclmeted soldiers^ KtouEcs» and goM-gErkndcd
two cbonoU and six bidlcckSj ot to iht NBÂnnu: in a JDuiH:lo[h who
^ns$ nailed up on Golgotha between two [liievcs? I merdy ask ihc qu«-
doDj hut Comot help feeling that^ were aiebuieulogy uIodc the arbiter^ rbe
answer would iiot be in doubt. Give us hclmeEa and gold garlands; evety
time; bread and drcuscS give us* provided that the bread is csrboni7£d
and the circuses well-fttmished wiili good solid bronze and marble ^
Hui Ici us ar tcasit in our gratitude for these things, remember th^ mijifl-
ing values that earmoi be appraised in incht* or soil-samples or smudges
in the cartlu
M this is not vmy cncDuraging. The archaeologist may fmd the tub
but altogether miss Diodes. I Ic may answer with boiamcol prodsion
Erowning's quesdon, "What porridge had John Keats?" without a pass-
log rtcugnition of the author of Hndyf/imh He must accept these risks,
consoling liimsclf with the reftcciion that no single approach to human
aCcomptLshmenL can be other than partiaJ and chancy. The literary
historian who overlooks art and enfremanship and environment may
Jose os much as the Biehaeologist who can produce for us a harp without
it$ music or a tub without its philo^pher. Let us therefore count our
blessings* We cannot fully read the brngnage of the Minoansj but ilieir
palaces and ftesoCKS* their w^rts and jewels, are them^dves a piao-
graphic language that tells us not a little of ihcir way of living^ mud
hints, however ^ngudy, at thdi way of thinking. We must be content
TO do what w'e can with the material vouchsafed m us, in fuE con-
^dousness of its Inoompleteaess.
Such is The uneven foundation on whidi the archaeologist i$ expected
tn artempr a vital leconstmction of tnonâ paai achievement- And m this
term "mponJitTUcdon* is included literal, three-dimensional re-creation:
the PaliDc of Mtnos as rebutlt by Sk Arthur Evans, Little Woodbury
as rc-mated by Mn. Jacquetm Hawte and her oolleagaes {FI. XXlIIJj
the Celtic chiijim vt Kt on its wheels again by Sir Cyril Fox* Whilst
lauding such recDnstTuaion, however, wc may be well aware of its
dangers* In the task of recoEtsimetion it may be ûapccacd that we arc
safest when wc set aside the great dvilmdom oTuniiquiry and conhne
onrwlves os anthiupologists to the less evolved Lummuninoi, to folk
whose anxious cxhtcnci: was concerned mainly with preservatinn from
hunger, gods, and neighbours. Whilst we con never hope to follow in
detail the COnvolutiOJUror that imnmiic organism the ûnruiored
214
WttAT ARE WE niRRlNG UT, AND WHY?
mind' and will lumbk ioeritahly into niimy gaps in iis rnisoidpg, fll
least it* mcandcriijgs will He generally wilhui the honjon of our coin*
pieheiwlBu. But wiiCR, in the security and amplitude nf dty*life(ciYiliM-
tion), men acquired leisure lo think between meifci, they began also to
escape iis midlcttually, UJitess they were fully and intelligibly liKraie-
Thus, it ia not very difficult for the erchaeobgiiit to itconsmici the
- humanity of Little Woodbury but it i$ near the limit of his powm to
revitaliac Mohiefl]<Hlaro. for all iis Hstonisbing completeness and in
admirable plutabing. Moheniu-dato remaiiis an isuluicd and petnfjcd
complex of another world, a dead city of the mlecheldiai or oliens whi^
unimidligible words arc not oompcnsaicd for by any adequate pictorial
art. The difference lies not merely in the relaiivc dimensiona of the two
problems but, above all, in their vridely diOetUig quality.
Be it repoared, all this i$ not s’cty encouraging. But the t«k of
rEKorntruCtlon whether ihree-dinwiKioniil or two-dimensional, is one
froni which the aichaccilogist must not shrinlt. It Is in a way the crown
of his work. And it is surprising and tcassuring to find how tnueb good
cDDStiucUve material can iu fact be extracted from a rubbish-pit—or lie
implicit, for rhat matter, jn the tale of a tub.
That Is the end of il»e present argument, but a postsîpt may Ik
added. The writer lepicseuts the end of an active generation. From hta
motnentary vantage-point, on the one liand he looks back upon the
patli w hich he and his colleagues have tried W mark out during the past
ihiftj years, and on the other hand peers hopefidiy into the mists of the
fulme. Of the past thirty years, it may be averred that wt have ^d to
devote a disprnportionatfi share to the invention and daboration of basic
techniques. That process vrilt, of txiurse, cotirinue in the future, but we
can perhaps claim—although it is rash to prophecy—to have reached a
point ftom which IKbnical improvement will be inddenwl and miiidy
ia detail. A good deal of the rough pioneering has been done; h rnoains
10 exploit and develop. A second retfospecrive observation ia that we
hate perforce devoted a great deal of our time to tlie sysidnatiTaiioo of
eulturea. This was inevitable. Tliiity years ago we knew all too little
where wc were, all too Utde of the extent and duectiOP of our matctial.
We have had to s« laboriously about the preparation ofa grammar of our
suhiom. WMk amplifying that grammar, it is for the future to use u
consmiciivcly and signiGcanUy. To change the mcraphor, we have, »
was rematked in Chaplet X, b«n preparing time-tables; let us now
have some tram*. Cultural catalogues are all very well, so far ns they go.
Its
WHAT ARE Wt DlGClHG UP, AND WHV?
Hwi they do iu>i, of themselves, go vety far. They arc q mcnjs lo od cdeL
An admitted need of the present day is the incthcidlail espbiraiioin of the
jodat unit on a more expansive sÂlC than lias been normal Ln the past.
The phase of the sandagem isolated trial ~pit, however sldlfiilly executed,
is now in large measure past. Wtat '*'"0 nc^ now is huneontaJ excavation
on an extensive scale. os tike, as an acamplc, an Aoglo-Saicon
ccmcicry. It i$ not enough lo know its genenl period and character.
Wc now require more predae aud ample i^rmation. 1 low many graves
dues it contain in its emiicty? What range of tinie does it cover? What
social grades does it indicate? What popiilatinn-nnit docs it represeot?
To answer these questions, we need (he carefiil unOOtiaing ofa
ctmetety, not merely a few grav es here and there. Arid, in qppraisinE
Signiheance, a suggesrion may be ofTcred if only as a ;fu d’erpn'r. Let
some enterprising invicsdgaior make an aeiuarid and chiunoinigical nir-
of the reasonably modem gravestones of some existing nan-indus*
trial village in England or Soudatid, and eoinpare iis grsve-group.s with
the equivalent periodic census-reiums for that village. What ratio does
the one bear to the other? A series of such czperimentSj used with pro¬
per care and a good deal of reservation, might introdnee 3 new dement
of iwm-objectiviiy into the papulalion-problcins which are of such
vital impononce to our undcTstandiug of past sodetic*. IndcHl, if I were
asked ro name one problem mom than another which demands invesii-
gation during the next thirty years, whether here or abroad, I should
say, ‘the problem of numbers'. It is not an inappropriate exhottatmu in
a ceusiB-yEir, and within a few mocLlhs ofPmfesior Max iVlallawan's
great discovery of the census-record of Assyrian Kimrud, to uige that
the people of the post be numbeted. No li|^t task fgr the arehacoloî,
but an csscoiial one if we are to transfono dty bona into something
apptoaclimg live social hisioiy. I.ei us—and by û$’ I incan 'yon nf (he
new ip)ciatjQn‘--get down to this task with steadfastness and dctci^
tnimiicm. Lang ago, ot Troy, at Silchester, at Mohenfo-daro, at Glaston¬
bury, wc learned to hack open the earth and to broadcast its trcuiirea.
Later, we Jeamed by careful small-scale dksectiou how to otplore
unalyiicaUy in deprh. We now need liilucaje thiw-dimtntional exenva-
Tinn, combining the merits of both merhodsj the total eicavatWQ of
cmnpleic settlcmenQ; the provision of reliable data for esdmating the
densiiy and sodfll structure nf popidatioii. Up to date, very linie of the
requistic work has bcsi undertaken and its distribution is very uneven.
Indeed, neatly the whole sodal problon lies before us, and we are now
ar6
WHAT ARE WH OTIjOI NQ AND WHT ?
At Jasl Gt to taddt iL I aivy the n^cw' gencraii™ its great oppommlty, as

aevei; bcfofCj to dig up ppoplc rather than mere things^ aDd to enahk U$>
in the fulness of eJmc, to view the past and the present as a singlCj 00J>-
Lmuous and not always tmsucccsafiiJ hitdc berween Man and his Kntiron-
ment and^ above aU, between Mm and hirnfelfn
And from the privilege of the printed page one GnaJ challenge may be
ibtown to the young arehaodugisL In yora past^p Vietor Hugo reminded
us diat we ait all mdci sentcuue of death.
We have on interval [added Walter Pater] and then our place knows iu
no mcire^ h, »» Our one change lies In expanding that tnxervaJi in getting u
mmy pulsations as possible into the given time. Great paa&lona may give
us this quickened Ecmc of life, eeno&y and sorrow of love^ the various
forma of enihusiaatic activJtyi disinterested or cthcrwiscj which come
natiinilly to many of us. Only be sure it is pasoion^—diai it docs yield
you this ^nit of a quickened^ muIdpUed eoiucfousoess.
Those are fine words of Waiter Puter'Sj finely thought. They may here
serve to remind us, once and for allj that the sdendst, the arehaeologist.
Is no mere clerk in a coujning-house^ no mere drofyanao m a drawing-
oflBce, Fassiou, enthusiasm^ call it what you will—'vitality", the over¬
worked ilm vita!f will do—that is tiie basic quality which our disdptjne
eimets ftODi US- And If you who are cmeiiiig upon It have no vital urge
Within, turn^ I begof you^ to some less human and more finite avocation
than the study of moukiud. There ore enough already of the houiic^
paimers who ape the aitist.
SELECT BIBLIOGRAPHY
R. J. C. Atkihsdh, Fitld Arcfatoiogy (London, 1945).

R, M> KnffVON, Btgim^ hi Afdmvfosy (Londfln, r953L).
GRAHAm CuiiR, Archaeotvgy and Stxitty (LoDdon, 1939).
O. G. S. OtAWFt^, Afshaei>ltgy m tkt Field (lAindan, 1953).
F, E, Eeuser, Datnt/s the FOif (£ad cd., London, 195a).
A. Luukg (ediior), La Dice>uverte du fatsi (Paris, 1951).
H. L PLENDatmru, The Fmcrvatian tf Anti^tut (LondDa, 1934J.
Glto E. Dameel, a Hundred Yetirf of Atdtaadogy (London, 1930)-
Index
Adventure, 212 f. Coffins, 81, 98,103.
Afghanistan, 89. Coity urns, 108.
Alaja HQyttk, 27. Control-pits, 66.
Allcroft, Hadri^, 211. Cookson, M. B., 174.
Amaravati, 121. Corpora, 183 f.
Anderson, Dr. E. C., 34. Cotton, Mrs. M. Aylwin, 157.
Animal bones, 165 f. Crawford, O. G. S., 189, 201, 218.
Arikamedu, 56, 120. Curie, Dr. James, 189.
Arnold, Matthew, 202.
Arnold, Dr. R. J., 34. Darwin, Charles, 195.
Arretine ponery, 43. D^ Geer, 28.
Dendrochronology, 31 f.
Bactra, 89. Deputy director of an excavation,
Baksheesh y 151 f. 137 ff.
Barrie, J. M., 132, 186. Diospolis Parva, 37.
Barrows, 7, 42, 81, 93. Director of an excavation, 131 ff.
Beech Bonom dyke, 116. Donne, Dr. John, 213.
Bersu, Dr. Gerhard, 18, 59, 128. Douglass, Dr. A. E., 31.
Bethel, 18. Draftsman, 144 ff.
Bhubaneswar, 125. Dunstable Downs, 94.
Birdoswald, 47. Dura-Europos, 87.
Blackburn, Dr. Kathleen, 47.
Brahmagiri, 50, 121, 123 flf. Earthquakes, 27 f.
Braidwood, R. J., 36. Egyptian calendar, 26.
Brickwork, 38. Emerson, Ralph Waldo, 50, 56.
British Association, 186, 193. Ervine, St. John, 195.
British Museum, 6. Evans, Sir Arthur, 27, 204 f., 214.
Browning, Robert, 214. Evans, Dr. Joan, 204, 205.
Buchan, John, 212. Evans, Sir John, 185, 204.
Buddhist stupas, 99. Evolution, II, 24.
Burgh Castle, 79.
Burials, 3, 7, 42, 68, 81, 93 ff. Fergusson, James, 123.
Bushe-Fox, J. P., 20, 80. Field-chemist, 144.
Field-laboratory, 169.
Caerleon amphitheatre, 75 ff. Fluorine test, 33.
‘Carbon 14’, 34 fif. Foucher, A., 89.
Cemeteries, 68, 103. Fox, Sir Cyril, 3,94,189, 203, 214.
Chagar Ba^, 27. Frankfort, Professor Henri, 206.
Chandravalli, 29, 121, 125. Frantz, Alison, 177.
Chanhu>daro, 56. Frazer, Sir James, 195.
Charsada, 89, 90. French Prehistoric Society, 15.
Childe, Professor V. Gordon, 201,
207. GaUey HiU skull, 33.
Chronology, 23 ff. Gedye, lone, 169.
Churchill, Sir Winston S., 200, Gergovia, 9.
Cists, 98. Gezer, 46, 49.
Clark, Professor Grahame, 20, 36, 80, ‘Ghost* walls, 78.
189, 195, 203. Gibbon, EdwaM, 126.
219
Glaiiotibiiry 117* Maiden Catcle, D^raerj £3.
G<»wm, Bk Btncftj 1^7- MiUmnn^ Pirafea^^^r Alu, iii!.
OrAplu^ 139. Manieiw iidmice^ 36.
GrmiwdU Quidd, 7^ MirahAli, Sir John, 89^ 139.
Orimn^ P+| Mailci, J3I.
Muonr^'p 3S.
Hidrun'i Wall, 47. Dean, 94.
ProTmKf J. fl. S., 193- MciQUihi>^ 3fS.
HirappS; 2Qt II, 6Bj fii, 871 91, iVlinoan CivUbiajiodx x 14.,
Hav7iiicict» Pfv^tor H. J.j 195^ 204. Moho£t|o-dafOj 51, fly, lay^ 315^
Hawfccsi F^rcswr C. F. iSp. Moulc paveRt£tii4r Ultln^ of, tji IT.
Hiiwkcii Mr»r Jicquccn, 40^ [91, 195^ Mouldinpp ^fl.
ti4^ .Mud-briek, 33, 84^
HiuCT I'TWlgp W9. Myiarep 48, i3i.
Hiwp VictWj 317*
Kmdinr, Juliim. t95j lay. Xripolcoii ip
HyUcmb^ Statc^ B, 121. HApoleon III, 4k 9.
Hywd Pd*> 50, hfewbould. Alba Theodomi v,
Nrw^tefliJk 1S9.
lllujcin[j0Dp eBIj iBgf.j 19^ C
Novlorii Bk isaaC) 40.
Indsdng caina, 1S5. Oakk^p Dr. Kcnncjhi 33i 34 ^
Indoclisg ^lmaIl-fi£kd&^ T O’Ncllp n; H. SiTh J*, 184^
Indii^ I soil. Oi^, prehiitOTic diy* toB.
lodii, Anchieotaâl Survey of^
l^ndui ViUcy OvIUaudatip 24, 5 j PackinE ponnyp 163.
57. Fjddjcanp 24, 29^89^ 137.
Ifluv fiibylcD, 79. pBlacoliEliic CEdniTcij aB.
Pucip Walieff 317,
I94h FnrlOp Sir W. Al. mniiertp 14,39, 37^
Jarmeij 36, jZp B4,86p9ip 183 ff.
Jejuup Sir J, IL^ 195. FhiHiptp C. W., Si fF.j iQi.
Jeffcnoflij Thi:ri£Lftf^ 4T f. Pbofopaphyp 143^ 174 ff.
Jeridko^ 27p Pifgon, Atri. Siuan, 303.
j€fD£iy Sir William^ rT« PiggQnp Profeuor SnmfTp 137^ 1S9,
PlbdEnm ikuU, 34.
Kf nr'» Ctveroj 53^ PtcdaTi 310.
Kcnyom, Miu K, M,* 79, Pitt Riven, v, 4, 11 fT.^ 30, 59, zoo,
KhAf^job^ 30^ 101 j iiSp iSiLp 189, 191.
KzkouAs 37i Pkndcrldclip Dr. II. J., 111, z£9.
KdldcwA^^^ R_, 79p IMicn, 35.
Kmuispiir, PqndidiciTy^ 43, 120.
Y^’ncry'-uiiicinZf 140 ff.
Pb«ery-dnmiiBk 147 f., 183.
Libel*. 55.159.
Labourer^ 148 PiMl^-ahcdk 157
Publication. iSi ff,
L«nî AuJrcWi 114*
Piihlidey, 19$^.
Len^jey* Dr. 203,
l^kivdi 4. Quadrant metliiW rtf wsflfttiaaf 94 C
Ubbr« Dr. W* F.p, 34* 37, Qullln'^^ûriik Sir ArthuTk 194. 196.
IkWf Cc^numwiHit Gctnuqi 4^ 20.
JJnIc WoodbEvj fsnmtMij, 13$, 314, Hu b~hiiiir3p 37.
LjndiKy Fflrtt Ranvm ille^ 182^ Hccordingk 51, €8, 184.
INDEX
Richborough, 8o. ‘Tactics* of excavation, 114 ff.

Richmond, Professor I. A., 20, 47, Tarsus, 27.
203. Taxila, 29, 68, 88.
River-gravels, 32. Taylor, Captain Meadows, 8, 18, 123.
Roman Britain, 22. Tel Brak, 27.
Roman coins, 117 (Beech Bottom), Tell el Ajjfil, 16.
120 (India), 125 (India), 184 (re¬ Tell el Amam, 27.
cording). Tell en-Nasbeh, 37.
Roof-trusses, 38. Tellsy 88, 91.
Rouletted ware from India, 121 fif. Tepe Gawra, 27.
Roy, General, 4. Timberwork, 80.
Royal Cômmission on Historical Tools, 153 ff.
Monuments (England), 146. Town-sites, 86 ff.
Royal Society’s Scientific Information Toynbee, Arnold, 204.
Conference, 186. Trenches, 68.
Trevelyan, G. M., 187,191, I95j 208.
Scarborough, Iron Age site at, 29. Trial-trenches, 63, 86.
Schaeffer, Dr. Claude, 27. Troy, 27, 43-
Schliemann, 9, 43. Typology, 37.
Seamer, 36, 80.
Sialk, 16. Ugarit, 27.
Silchester, 127. Ur, 109.
Simpson, F. G., 30, 47. Ussher, Archbishop, 25.
Sisupalgarh, 125.
Site-supervisors, 138 ff.
Van Giffcn, Dr., 18, 80, 99.
Skendleby long-barrow, 81 ff., 101 f.
Varved clays, 28.
Smith, William (‘Strata’), 41.
Vegetation, 47.
Sodcty of Antiquaries of Scotland, 6. •
Verulamium, 78, 79, 115 ff.
SoUs, 166.
Virginia, 42.
Solar radiation, 32.
Sondagesy 63, 86.
Spengler, Oswald, 206. Wheathampstead oppiduniy 118.
‘Square’ layout, 63, 64 ff. Wheeler, Mrs. T. V., iii.
Stflel, Msidamc de, 213. Whitman, Walt, 56.
Staff of an excavation, 130 ff. Winkelbury camp, 14.
Stanwick sword and scabbard, iii. Wisconsin glaciation, 36.
Stoffel, Colonel, 9, 64. Woodward, Sir Llewellyn, 202, 209.
Storage-pits, 45. Woolley, Sir Leonard, 103, 105, 109,
Straight-joints, 79. 195.
‘Strategy* of excavation, 114 ff. Wor barrow, 14, 100.

Stratification, 29, 40 ff.
Structures, excavation of, 72 ff. Xerxes, i.
Stukely, 204.
Sumner, Heywood, 146.
Ysceifiog, 94.
Surveying, 142 ff.
Sutton-Hoo ship-burial, 82 ff., 106 ff.
Swanscombe skull, 33. Zeuner, Professor F. E., 31, 213.
221
nuKrm »
QldPVt IKETAZK
Al Jim
CrKiVUSITY ntui
DXFfikb
Pit
CTTATIP RATIT
WKtU
70 im
t»Gt¥lUITV
•I f

Archaeology From The Earth

Uploaded by

Copyright:

Available Formats

Archaeology From The Earth

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Archaeology From The Earth

Uploaded by

Copyright:

Available Formats

GOVERNMENT OF INDIA

ARCH/EOLOGIOAL survey of INDIA

SIR MORTIMER WHEELER

tmr Ji» iht Unitt

-L,\ J ,v vi_ AKCHAEOLOCIGAL

JV. A. Chacft; csEOavarinii in the Eist^ Jyjy

y. Layout of an arei-cKwstioii (Ta3tilii-Bhir Mniind, 1944) *5

VI r. A. Section pegged for thrce-fibaicnsional repjniii^

X, Tracrci of huidle-comtmainn in the nKlithic Iwng-buTow

!tj. Trace! of the Sutton Hs>;> *hipr Sulfoik- Phoioiraph* British

*11. Section through the mud-bfick defence* "f g<

V, THE LAYOUT OF AN EXCAVATION ««

YI. THE EXCAVATION OP A STRUCTURE T*

VII. ON DIRGINC TOWN-SITES

IX. WATCn-MARSRS’ JOBS

X. TACTICS AND STRATEGY IM

'XIIL THE POTTERY-SHED 'J7

XIV. THE FIELD-LABORATORY

XV, PaOTOGRAPHY , _ *74

IVI. PUBLICATION AND PUBLICITY ■*

XVII. WHAT ARE WE DIGGING UP, AND WHY? 200

SELECT BIBLIOGRAPHY »*®

that I am indulging in temiiibceuee, not laying down laws. For thirty

Eprefer to regard it as an art, or even as a philosophy, we must still afi&rm

across itic ages* whether the time-inLetvui be 500 or 500^000 ymn, h

therefore a disproportionate amount of archaeology; and, unless the

The woodcuts (PI. i) speak for themselves.

Ev»viikN:i of 1 bvrnw, 1K44

number of The megiiliThic tombs charafterisiic of central and soutbetn

IL may be added that the Colonel's rej^enred enthusiasm was shared

Gcncral^L^c^F^ by & ^kurpming ^orici of diuocts sooceeilcd Lo flit

It will, perhaps, be thought by nunc [be rcmarl^] that I have recorded

In practice, Pitt Rivm's method was to record every objM bi such

* W. M. Flinders Petrie, Methods and Aims in Archaeology^ p. 29.

FtC. 4- JkoicKi ihivy^ part of the site Bethel j PsiJcitiiic* 133^

I that B3 feechtly as 1940 the Survey cmii publish ia itKsniimcntfll furm

in tSw, ihc ricnHsm Lifiut Conomisswn, thougb oft™ tcduuQiIIy bdow

Fig. 6. ScctiMi part of HaroppA, Fakistmij

Fig, 7. Seetton ihraugh part of a ttni««r n Huappa> FaXiiton, t«o

with a thick tiKUMlft of uadilfcimtiated ^dl Wrttnnal ^

rf,taok IX. ™ the n.M that ia«

sucotssion flnd ibc poiiiiliddn of a caavIndflE OJlnire-sequmCC arc dot

^ <1 F. A. Schaeffer, Scrai^pAtV o>mparU rr eAriJunlcieiV dp rAn* ui^idnntak

been nude lo crnivcrt iHs relalrte into an absoluie lime-iwalc. *TiniC'

lUidcdyLng die Roeimii sigool-iiiatiuii, Mr. F. G, SiiBp&oiij mostlj

ihc limt represented by thui acaunnlaiiuii ul ifi.- ]l = lOO yens. Ta

important to know what the gcochronologists want for their alchemy,

quintitie£ of tlicm, virlth special ‘boostin^^' supetadded^ will m fact yield

meihod of to Iwvc fknimhcd at aboui the a(xq>Uible dflte of

AJl paraSleli in form to a pot-type tuive b«n tmeed, including Middle

(Egypt Ejcplonitiijn Fund, 1901)^ pp^ 4 - -

p. li'ald p'l^! (Han-ard, 19,4).

rr t!** “"*• “S« aw not enough, (Mrs

In more plOSAjc seose^ this letm scniiification ]ms aineady appared

■ A Land CLotidcnip t^i), p. a6.

^ ^ Q^' scientific skill fell upan LnTtrtlLc ]

In praeuocj^e idcutificaClOO and corr^daiinn of the strata or layers )

table, therefore, of a related human phase. But it is rarely, in all con¬

IhnJ.' Secti°" illus^ting (A-B) an apparent succession of strata (3-7)

!tj. Trace! of the Sutton Hs>;> hipr Sulfoik- Phoioiraph British

11. Section through the mud-bfick defence "f g<