The excavation of Gorham’s cave and its relation to the prehistory of southern Spain
John Waechter
[page-n-21]
JOHN WAECHTER
( 1 ",I.I ~tr.)
Th~ excavation of Gorham'$ cav~ and it!>
r~lation to th~ pr~hi$to ry of $outh~rn
5pain
The excavation ot Gorhom 's cave, though not ye t completed,
has thrown additiona l light not only on the prehi story of Gibralta r
but to some ex tent on t ho t of southern Spain.
Before the work a t Gorhom's began in 1948 (I) the only si te
in Gibraltar which hod produced prehistoric implements in situ wos
the De . . il's Tower Co....e excavat ed in 1927-28 by Professor Gorrod .
This cave only contained one archaeological period, namely Mousterion, a rother rough industry made mostly of quartzite, and with
few well mode tools, but in addition the skull of 0 child of undoubted Neanderthal type . On founal e .... idence this material was doted
os lote Mousterion, largely owing to t he absence of t he larger forms
!iuch as Elephant and Rhinoceros.
Gorho m 's cave wh ich lies a t the base of the cliffs on the eastern side of Gibraltar con tains a deposit 0 littl e ove r 11 metres
t hick, composed almost en t irely of wind blown sand. So for the
excavations hove only reached a depth of under 6 m e t res below
the cove fl oor, but this has yielded m a terial of considerable interes t. The highest level contained potte ry doting from Pun ic and
Raman, with beads, scarabs, ete. Below this was 0 thin stalagmite
floor with 0 few pieces of hand mode incised pottery at t he top.
(1) J. d'A. WAECHTER: "Excavations at Gorhom'$ Cove, Gibra ltar", Paper
n." 3 reprinted from the Proc::eedings of the Prehistoric Society for 1951.
_21 _
[page-n-22]
2
JOHN WAECHTER
No clear s trotigrophical distinction was possible in th is upper loyer
os it was to 0 la rge extent disturbed. The stalagmite floor extended over the whole excavated area and was fairly uniform in thickness . It is cleor that during the formation of this fl oor, which must
have taken 0 considerable time, the cave was not only unoccupied
but little or no sand was being blown in from outside.
Undernea t h the stalagmite floor wos 0 loyer of yellow sand
wi t h traces of charcoa l but no clearly defined hearths, and below
t his 0 dark sond with abundan t hearth debris. Unfortunately the
quantity of materiol from both these layers was small, but the
broad character of the industr ies is clear. The upper yellow sand
con ta ins on, industry of small end-scrapers and small backed bla des; the bone tools are few and rather rough, but ane si ng le bevelled bone point bears a strong resemblance to t hose from Professor
Pe ri cot's excava tions at Parpall6. The lower darker laye r also contains 0 blade industry, but there is not enough material to draw 0
useful comparison with that above material, but what there was
was sim ilar in character to the material above, though there appears to be less micral ithic. These upper industries ore separated
from the Mousterian by 0 thin sterile layer.
So for four Mousterian layers have been found. though on ly the
upper has been excavated over a large area . This yielded 0 rough
Mousterion industry si milar to that from the Devil's Tower. The
greater part of it is quartz ite, though some of the better mode pieces ore made of chert or jasper: the number of fini shed implements
was small and rough util ised flakes appear to hove been the main
tool. In this upper level there was a marked tendency towards
rough blades and there were also a few simple burins.
The t hree Maus terian layers below t his have so fa r only been
fou nd in sma ll soundings, and there is insufficien t materia l for comparison with the one above, but there does not seem to be any
major d ifference between them . The founal remains from 011 the
layers was quite extensive, bu t it has not been s tudied in detail,
though t here does not appear to 'be any pochyderms.
The pasition of the cove in relation to the sea makes it poss ible
to draw some conclusions as to chronology because of the effect
of the fluctuating sea level on the formation of the cove deposi t.
It is obvious tha t the whole of the present deposit must onte ·date
the 8 metre Monasterian 11 beach of the last inter glacial, and
-
22 -
[page-n-23]
T H E EXCAVATION OF GO RHAM'S CAVE
3
tha t any la ter rise in sea level would make it im possible to reach
the cove, os well os stopping the accumula tion of the wind-blown
sand .
It was possible for Professor Zeuner of the Inst itu te of ArchaeoIcgy of London to visit the cove during the lost week of the 195 1
season and to obtain a series of full samples from all the layers
then exposed . His examinat ion of t he sam ples suggests t ha t the re
were two periods when the sea level was re la tively high, one be t~
ween t he fi rst two Mous terian layers, on one above the la tes t Mousferian layer. Un til the entire sect ion is exposed it is not possible
to draw any definite chronologica l conclusions, bu t t he whole deposi t star ted forming du ri ng the Los t Glacia t ion and tha t during this
period t here were fl uctua t ions in sea level wh ich probably accounts
for the breaks in the archaeologica l sequence
In evalua ting the resu lts so for obtained from this excavation
it is obvious that comparisons must be mode os fa r os possible with
southern Spain. So for not a great deal of work has been done in
Andalucia, but Pericot's excavations at Parpall6 and Jard6's a t
Cova Negro ore two obvious sites for comparison . There are cer tain
features in the materia l from Cava Negro, porticularly level B,
which suggests t hat the upper Mousterian layer of Gorham's Cove
and poss ibly the Devil's Tower belong to the some phase. It is not
yet poss ible to compare the lower Mousterian from Gorham's Cave
wi t h Cava Negra C, D, and E os there is insu ff icient ma te rial.
T he obvious comparison for t he la te r layers is wi th Porpoll6,
though again the amount of materia l from Gibral tar is small. The
single bevell ed bonc point is st rikingly like those from Parpa ll6,
though the Gibraltar specimen has no lines engraved on the base.
The flint implements from the lower of these levels also bear 0
strong ressemblonce to the Mogde linian of Parpall6, but the re were
no drawings or paintings, but there were seve ral fla t stones s tai ned with Haematite and in the some layer there were small pieces
of graphite.
,
Ta a ttempt to fit the Gibralta r material in the general Spanish
sequence is 0 little premature, but some ten tative sugges tions seem
poss ible. It a ppears probable that the pictu re presen ted by t he si tes
in eos lern Spain may be well extend in brood outli ne to the south
coas t, certainly from the Mousterion onwards. If this proves to be
the case then the establishmen t of 0 f ixed chronology for the Gi- 23 -
[page-n-24]
4
JOHN WAECHTER
bro lter material, provided that this mat~riol con be linked typologically with Spain, will be of the grea test importance, particularly for relating eastern and southern Spain with the Fronco-cantobr ique region.
It would not be ou t of place here to conside r t he connect ions,
if any, be tween Spain and North Africa duri ng the period covered
by the Gorham's cove deposits. In North Africa there appears to be
both 0 Mousterion and 0 Levalloise, t hough in many cases they
ore obviously mixed. This duallity appears to decreose as we proceed eastward along the African coost, the Levolloise element in creasing and the Mousterian di minishing . It was fo r t his reason
that McBurney referred to his material from Cyrenoica as Leval loiso-Mousterian and Miss Catan -Thompsan to drop the term Mousterian oltogether for Khargo and Egypt. This suggests that the
Mousterian ot North Africa may well be due to influence from
Spoin .
It is extremely unfortuna te that the excavations a t Gorham's
do not throw any light on the Aterian -Solutrian problems os no
material of th is period was found , the possible explana tion being
that the sea level was high ot this time, though the negative evidence is insu ff icient, and 0 similar explana t ion might well be evok ed for the absence of the earlier Upper Palaeolithic industries.
T he relations between the Aterian and the Solutrian are st ill a
matter of controversy, but it is not essential for there to have been
contact between t hem to explain their similarity as the basic elements of t he two industries are very dissimilar.
The development towards microlithic implemen ts in t he latest
industry at Gorhom's is interesting because i t seems to follow the
pattern for Easte rn Spain, certain ly t here is no trace of Capsian
influence either in Gibral tar or further north, and in both areas the
Upper Palaeolithic appears to become microlithic without there
being a break in t he development.
The Mousterian from Gorham's ante-dotes t he last inter glacial
and appears to survive through one change of sea level, which
sugges ts that it con ti nued into Wurm 11 , and may we ll have token
the place of the Chatelperron in this area. To attempt to fit the
later material into the time sca le is a t present unwise, but 0 Wurm
III do te for the later material, based on the last sea level cha nge,
is very tempting.
_24_
[page-n-25]
JOHN WAECHTER
( 1 ",I.I ~tr.)
Th~ excavation of Gorham'$ cav~ and it!>
r~lation to th~ pr~hi$to ry of $outh~rn
5pain
The excavation ot Gorhom 's cave, though not ye t completed,
has thrown additiona l light not only on the prehi story of Gibralta r
but to some ex tent on t ho t of southern Spain.
Before the work a t Gorhom's began in 1948 (I) the only si te
in Gibraltar which hod produced prehistoric implements in situ wos
the De . . il's Tower Co....e excavat ed in 1927-28 by Professor Gorrod .
This cave only contained one archaeological period, namely Mousterion, a rother rough industry made mostly of quartzite, and with
few well mode tools, but in addition the skull of 0 child of undoubted Neanderthal type . On founal e .... idence this material was doted
os lote Mousterion, largely owing to t he absence of t he larger forms
!iuch as Elephant and Rhinoceros.
Gorho m 's cave wh ich lies a t the base of the cliffs on the eastern side of Gibraltar con tains a deposit 0 littl e ove r 11 metres
t hick, composed almost en t irely of wind blown sand. So for the
excavations hove only reached a depth of under 6 m e t res below
the cove fl oor, but this has yielded m a terial of considerable interes t. The highest level contained potte ry doting from Pun ic and
Raman, with beads, scarabs, ete. Below this was 0 thin stalagmite
floor with 0 few pieces of hand mode incised pottery at t he top.
(1) J. d'A. WAECHTER: "Excavations at Gorhom'$ Cove, Gibra ltar", Paper
n." 3 reprinted from the Proc::eedings of the Prehistoric Society for 1951.
_21 _
[page-n-22]
2
JOHN WAECHTER
No clear s trotigrophical distinction was possible in th is upper loyer
os it was to 0 la rge extent disturbed. The stalagmite floor extended over the whole excavated area and was fairly uniform in thickness . It is cleor that during the formation of this fl oor, which must
have taken 0 considerable time, the cave was not only unoccupied
but little or no sand was being blown in from outside.
Undernea t h the stalagmite floor wos 0 loyer of yellow sand
wi t h traces of charcoa l but no clearly defined hearths, and below
t his 0 dark sond with abundan t hearth debris. Unfortunately the
quantity of materiol from both these layers was small, but the
broad character of the industr ies is clear. The upper yellow sand
con ta ins on, industry of small end-scrapers and small backed bla des; the bone tools are few and rather rough, but ane si ng le bevelled bone point bears a strong resemblance to t hose from Professor
Pe ri cot's excava tions at Parpall6. The lower darker laye r also contains 0 blade industry, but there is not enough material to draw 0
useful comparison with that above material, but what there was
was sim ilar in character to the material above, though there appears to be less micral ithic. These upper industries ore separated
from the Mousterian by 0 thin sterile layer.
So for four Mousterian layers have been found. though on ly the
upper has been excavated over a large area . This yielded 0 rough
Mousterion industry si milar to that from the Devil's Tower. The
greater part of it is quartz ite, though some of the better mode pieces ore made of chert or jasper: the number of fini shed implements
was small and rough util ised flakes appear to hove been the main
tool. In this upper level there was a marked tendency towards
rough blades and there were also a few simple burins.
The t hree Maus terian layers below t his have so fa r only been
fou nd in sma ll soundings, and there is insufficien t materia l for comparison with the one above, but there does not seem to be any
major d ifference between them . The founal remains from 011 the
layers was quite extensive, bu t it has not been s tudied in detail,
though t here does not appear to 'be any pochyderms.
The pasition of the cove in relation to the sea makes it poss ible
to draw some conclusions as to chronology because of the effect
of the fluctuating sea level on the formation of the cove deposi t.
It is obvious tha t the whole of the present deposit must onte ·date
the 8 metre Monasterian 11 beach of the last inter glacial, and
-
22 -
[page-n-23]
T H E EXCAVATION OF GO RHAM'S CAVE
3
tha t any la ter rise in sea level would make it im possible to reach
the cove, os well os stopping the accumula tion of the wind-blown
sand .
It was possible for Professor Zeuner of the Inst itu te of ArchaeoIcgy of London to visit the cove during the lost week of the 195 1
season and to obtain a series of full samples from all the layers
then exposed . His examinat ion of t he sam ples suggests t ha t the re
were two periods when the sea level was re la tively high, one be t~
ween t he fi rst two Mous terian layers, on one above the la tes t Mousferian layer. Un til the entire sect ion is exposed it is not possible
to draw any definite chronologica l conclusions, bu t t he whole deposi t star ted forming du ri ng the Los t Glacia t ion and tha t during this
period t here were fl uctua t ions in sea level wh ich probably accounts
for the breaks in the archaeologica l sequence
In evalua ting the resu lts so for obtained from this excavation
it is obvious that comparisons must be mode os fa r os possible with
southern Spain. So for not a great deal of work has been done in
Andalucia, but Pericot's excavations at Parpall6 and Jard6's a t
Cova Negro ore two obvious sites for comparison . There are cer tain
features in the materia l from Cava Negro, porticularly level B,
which suggests t hat the upper Mousterian layer of Gorham's Cove
and poss ibly the Devil's Tower belong to the some phase. It is not
yet poss ible to compare the lower Mousterian from Gorham's Cave
wi t h Cava Negra C, D, and E os there is insu ff icient ma te rial.
T he obvious comparison for t he la te r layers is wi th Porpoll6,
though again the amount of materia l from Gibral tar is small. The
single bevell ed bonc point is st rikingly like those from Parpa ll6,
though the Gibraltar specimen has no lines engraved on the base.
The flint implements from the lower of these levels also bear 0
strong ressemblonce to the Mogde linian of Parpall6, but the re were
no drawings or paintings, but there were seve ral fla t stones s tai ned with Haematite and in the some layer there were small pieces
of graphite.
,
Ta a ttempt to fit the Gibralta r material in the general Spanish
sequence is 0 little premature, but some ten tative sugges tions seem
poss ible. It a ppears probable that the pictu re presen ted by t he si tes
in eos lern Spain may be well extend in brood outli ne to the south
coas t, certainly from the Mousterion onwards. If this proves to be
the case then the establishmen t of 0 f ixed chronology for the Gi- 23 -
[page-n-24]
4
JOHN WAECHTER
bro lter material, provided that this mat~riol con be linked typologically with Spain, will be of the grea test importance, particularly for relating eastern and southern Spain with the Fronco-cantobr ique region.
It would not be ou t of place here to conside r t he connect ions,
if any, be tween Spain and North Africa duri ng the period covered
by the Gorham's cove deposits. In North Africa there appears to be
both 0 Mousterion and 0 Levalloise, t hough in many cases they
ore obviously mixed. This duallity appears to decreose as we proceed eastward along the African coost, the Levolloise element in creasing and the Mousterian di minishing . It was fo r t his reason
that McBurney referred to his material from Cyrenoica as Leval loiso-Mousterian and Miss Catan -Thompsan to drop the term Mousterian oltogether for Khargo and Egypt. This suggests that the
Mousterian ot North Africa may well be due to influence from
Spoin .
It is extremely unfortuna te that the excavations a t Gorham's
do not throw any light on the Aterian -Solutrian problems os no
material of th is period was found , the possible explana tion being
that the sea level was high ot this time, though the negative evidence is insu ff icient, and 0 similar explana t ion might well be evok ed for the absence of the earlier Upper Palaeolithic industries.
T he relations between the Aterian and the Solutrian are st ill a
matter of controversy, but it is not essential for there to have been
contact between t hem to explain their similarity as the basic elements of t he two industries are very dissimilar.
The development towards microlithic implemen ts in t he latest
industry at Gorhom's is interesting because i t seems to follow the
pattern for Easte rn Spain, certain ly t here is no trace of Capsian
influence either in Gibral tar or further north, and in both areas the
Upper Palaeolithic appears to become microlithic without there
being a break in t he development.
The Mousterian from Gorham's ante-dotes t he last inter glacial
and appears to survive through one change of sea level, which
sugges ts that it con ti nued into Wurm 11 , and may we ll have token
the place of the Chatelperron in this area. To attempt to fit the
later material into the time sca le is a t present unwise, but 0 Wurm
III do te for the later material, based on the last sea level cha nge,
is very tempting.
_24_
[page-n-25]