Content uploaded by Mozhgan Jayez
Author content
All content in this area was uploaded by Mozhgan Jayez on Nov 09, 2016
Content may be subject to copyright.
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016 Manuscrit reçu le 11 novembre 2015, accepté le 11 décembre 2015
A
:
C
M
vs
T
.
A
K
,
C
S
,
I
M
.
J
and
H
.
V
N
Abstract: Komishan Cave in the southeast of the Caspian Sea was selected for a small excavation in 2009 to address questions
concerning the Mesolithic industries of the region. Two AMS dates present an age between 10,628 cal. BC and 11,771 cal.BC for
the Mesolithic deposits. The chipped stone assemblage is characterised by large concentration of blade/lets, notches/denticulates,
scrapers of all forms, especially end scrapers, borers, and backed pieces. This chipped stone assemblage is obviously most similar to
those of Kamarband, Hotu and Alitepe regarding their locations, which are in the same climatic, geological and ecological conditions.
The Mesolithic chipped stone industr y of the southeast of the Caspian Sea has been previously attributed to Trialetian lithic industry.
In this article the differences in raw material access, technological organization and the subsistence between these sites are discussed;
as a conclusion based on Human Ecology the Mesolithic of the east-southeast of the Caspian Sea should be separated from Trialetian
Industr y and presented as “Caspian Mesolithic.”
Résumé : La grotte de Komishan, au sud-est de la mer Caspienne, a fait l’objet d’une petite fouille en 2009 an de répondre aux
questionnements portant sur les industries mésolithiques de la région. Deux datations AMS pour les dépôts mésolithiques donnent des
âges compris entre 10.628 et 11.771 cal. BC. L’assemblage lithique est caractérisé par une grande concentration de lames/lamelles,
d’encoches/denticulés, de grattoirs de toutes formes, en particulier de grattoirs distaux, de perçoirs et de pièces à dos. L’assemblage
lithique de Komishan est évidemment très semblable à ceux de Kamarband , Hotu et Alitepe, étant donné leur localisation dans des
conditions climatiques, géologiques et écologiques communes. L’industrie lithique mésolithique du sud-est de la mer Caspienne a
été précédemment attribuée au Trialétien. Dans cet article, les différences entre ces sites concernant l’accès aux matières premières,
l’organisation technologique et la subsistance sont discutées et, sur la base de l’écologie humaine, amènent à la conclusion que le
Mésolithique de l’est et du sud- est de la mer Caspienne doit être exclu de l’industrie trialétienne et désigné comme “Mésolithique caspien”.
Keywords: Komishan; Chipped stones; Caspian Mesolithic; Trialetian .
Mots-clés : Komishan ; Pierres taillées ; Mésolithique caspien ; Trialétien.
INTRODUCTION
Little is known concerning the lithic industries of the
northern part of the Alborz Mountains. Only few Paleolithic
lithic assemblages have ever been studied within this geo-
graphical region, and only limited numbers of Paleolithic sites
have been excavated (g.1). C.S. Coon, who has been consid-
ered by most authorities as the pioneer in the Iranian Paleolithic
archaeology, carried out the rst systematic Paleolithic exca-
vations in this region and as a result of that he managed to
recover lithic materials assigned to the Mesolithic layers from
the two caves of Hotu and Kamarband (Coon 1951 and 1952).
075-094-Jayez.indd 75 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
76 M. J
and H. V
N
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
Following in the footsteps of Coon, in search of Upper
Paleolithic occupations, Charles McBurney from the University
of Cambridge chose the two caves of Alitepe and Kiaram I
located in the southeastern part of the Caspian Sea area for
archaeological excavations. Based on the radiocarbon dating
and study of the lithic assemblages, McBurney (1964) assigned
Alitepe to the Mesolithic, and K iaramI to the Middle Paleolithic.
Perhaps the most recent Paleolithic eldwork in the southern
part of the Caspian Sea, aside from what is presented here,
would be the excavations at the open air Upper Paleolithic site of
Garm Rud2, nearby the city of Amol, which were conducted
during 2005-2008 (Berillon et al. 2007 and 2009). These eld
projects provided a signicant number of lithic materials, which
are comparable to the Upper Paleolithic industries previously
reported from the Zagros Mountains in Western Iran (Chevrier
et al. 2006).
In June 2009, the rst season of excavation at Komishan
Cave (g.2) was initiated. The entrance of this cave, which is
located a few kilometers west of the sites of Hotu, Kamarband
and Alitepe (g.1), had been blocked by sediments for hun-
Fig. 1 – Location of Komishan and other Paleolithic sites mentioned in the text.
Fig. 2 – Komishan Cave, view from north-east.
dreds of years, and was rst revealed during the 1990s as a
result of some road expansion projects1 and later, in 2003, the
1. Vahdati Nasab H., Report of the rst season of Excavation in Komishan
Cave. Unpublished report in the Iranian Centre for Archaeological
Research. Tehran, 2009 (in Persian).
075-094-Jayez.indd 76 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
A separation: Caspian Mesolithic
vs
Trialetian lithic industry
77
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
section of a pit dug by looters in front of the cave was cleaned
and the back dirt was sieved in search of cultural remains
(Mahfroozi 2003; Mashkour 2010). In order to test the strati-
graphical layers, a trench of 2x 2m was set up in 2009 at the
entrance of the cave (g.3). Due to the presence of 1.6m of
disturbed layers (the top surface of the cave was disturbed by
dozer blade), the excavation started by digging 20-30cm levels
until it reached the in situ layers in which the excavation
method was based on the arbitrary 5cm levels. In contrast to
the top 1.64m sediments of the trench, which had been severely
affected by the industrial activities, the lower layers containing
the Paleolithic materials were intact. The intact layer, which is
homogeneous both in color and composition and cultural mate-
rial, is divided into 13spits 5cm thick, here presented as lev-
elsI to XIII from top to bottom of intact sediments. The AMS
dating results obtained from the in situ layers suggest that the
intact excavated layers are located between 10,628 cal.BC and
11,771 cal.BC (Lab NºOxA-22572 and OxA-22611) and there
is no clear stratigraphic change in between the sediments
(Vahdati Nasab et al. 2011).
The faunal assemblage from Komishan is primarily a result
of anthropogenic activity and mammals (mostly gazelle), birds
and sh were exploited throughout this period.2 Our prelimi-
nary analysis of the pottery, which is exclusively found from
disturbed layers, indicates that these layers contained a mix of
Neolithic, Chalcolithic, Bronze, and Iron Age materials. The
most exclusive discovery at Komishan might be three perfo-
rated canines discovered in in situ layers (Vahdati Nasab et al.
2011). In this article we mostly concentrate on technological
aspects of the chipped stone assemblage of Komishan and its
attribution to either Trialetian Industry or Caspian Mesolithic.
The Mesolithic chipped stone industry in the southeast of
the Caspian Sea has been attributed to Trialetian Lithic
Industry by S.K.Kozłowski (1994; 1996 and 1999); different
scholars have repeated such attribution following Kozłowski
(Kozłowski and Aurenche 2005; Peregrine and Ember 2002).
With new information at Komishan in hand, we are going to
reinvestigate this attribution to nd out whether the Mesolithic
of the southeast of the Caspian Sea should be regarded as a
part of Trialetian or it denotes a separate industry.
Trialeti was originally the name of Mesolithic industries
from sites of the area of Khrami River in Georgia, which was
rst introduced by M.K.Gabunia in 1976, based on the assem-
blages from two sites: Edzani and Zurtaketi (Gabunia 1976;
2. Knapp Z
.,
Preliminary Assessment of Animal Rem ains Recovered During
Excavation in 20 09 at Komishan Cave, Mazandaran, Iran. Unpublished
Dissertation presented for Msc in Bioarchaeology. Nottingham, 2012.
Fig. 3 – Komishan Cave plan and Trench 1.
Kozłowski and Aurenche 2005: 70; Kushnareva 1997: 9;
Meshveliani et al. 2007: 49, 52). Then in another article the
lithic industry was described in detail (Bader and Tsereteli
1989). In 1994, Kozłowski for the rst time mentioned a
“Triaulian” industry with some sites attributed to this industry
among which was not Zurtaketi, one of the main sites of
Trialeti, which Gabunia had described (Kozłowski 1994: 144).
In 1996 and later in 1999, Kozłowski presented the “Mesolithic
Trialetian Industry” once again with the same sites he had
mentioned in 1994. What is important in this nal description
is, not only that Zurtaketi has been excluded from Trialetian
sites but also that the differences between the original Trialeti
described by Gabunia (1976) and the Trialeti dened by
Kozłowski (1999) have been totally ignored. The main differ-
ences between these two come directly from the extended geo-
graphical region that Kozłowski has added to the original
Trialeti area and includes the southeast of the Caspian Sea
075-094-Jayez.indd 77 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
78 M. J
and H. V
N
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
where Komishan is located along with other Mesolithic sites
such as Kamarband, Hotu and Alitepe.
It should be mentioned that with regard to the recent investi-
gations there are inconsistencies in the descriptions of Trialetian
Industry presented by Kozłowski (1999). Moreover, the extended
area and time span in which Trialetian industry has been recog-
nized by Kozłowski includes various cultural periods and eco-
logical zones. The south-southeast of the Caspian Sea is placed
in the Caucasus-Caspian province (Trialetian Industry home) in
Kozłowski’s (1999: 139) denition:
“[it] extends between the principal ridges of the Zagros and
the Eastern Taurus on the one hand and the northern slopes of the
Caucasus and Crimean Mountains on the other.”
That means the southeast of the Caspian Sea is not even in
the main area of this province and even in his description of
fauna of this province, including forest fauna accompanied by
steppe gazelle and mountain-dwelling wild sheep and goat
(Kozłowski 1999: 139), there is no mention of the marine
resources of the Caspian Sea and of the avifauna whose exis-
tence we are aware of in Mesolithic subsistence in the south-
east and east of the Caspian Sea (Masson and Sarianidi 1972;
Kohl 1984: 42; Brunet 1998).3
The chronological aspect of the Trialetian Industry is
extended through “the Upper Palaeolithic Imeretian Culture
from Georgia and the north-eastern coast of the Black Sea, as
well as from the similar, poorly researched assemblages known
for example from Turkmenistan” (Kozłowski 1999: 139); and
also we nd out that even Neolithic sites are included in the
tables describing Trialetian Industry (e.g. Ibid.: 147). It seems
that not only the geographical area of this industry is too
extended but it is also applied to so many different archaeo-
logical periods from Upper Palaeolithic to Neolithic. It is obvi-
ous that such a denition is too general. For example Hallan
Çemi, which has been listed among Trialetian sites by
Kozłowski (1999: 144-147), is a largely aceramic site in the
Taurus foothills of Eastern Turkey. Between 10,600–10,000BP
(uncal.), it was occupied year-round by Neolithic people
(Rosenberg 1994). The fact that Hallan Çemi people were sed-
entary points to a totally different culture from hunter-gather-
ers of Upper Palaeolithic and Mesolithic.
Kozłowski (1999: 144) describes the technological struc-
ture of Trialetian sites as “home”, which means that all or
almost all of the core formation and tool production processes
were performed on the site, whereas the technological struc-
ture of an assemblage is more relevant to the site function, its
3. See also footnote 2 (Knapp 2012).
position in the reduction sequence and accessibility to raw
material. It is worth mentioning differences in the technologi-
cal structure of chipped stone assemblages from Upper and
Lower Mesolithic of Kamarband, which are both Mesolithic,
but they present totally different structures due to different
functions the site had during each period (Coon 1951: 37-40).
Komishan is the most newly excavated site in the southeast
of the Caspian Sea with a Mesolithic chipped stone industry,
which can contribute in indicating the relationship between
Mesolithic sites of the region and Trialetian Industry. Here we
rst present the description of the chipped stone assemblage
and then we proceed to nd out how genuine the attribution of
Mesolithic industries of the southeast of the Caspian Sea to
Trialetian Industry could be.
LITHIC ASSEMBLAGE OF KOMISHAN
In total 16,053 lithic artifacts were recovered from excava-
tion at Komishan Cave. Of these, 5149pieces are from the dis-
turbed layers (DL) at the top of the trench, and 10,904 from the
in situ layers. The lithics recovered from the in situ layers are
divided into point plotted and non-point plotted groups; the
former are those which are recorded by their exact position in
the layer (x, y, z), and the latter are those which are from the
oatation process. These include mostly small pieces, such as
chips and debris. Every point plotted piece and also some of
the non-point plotted materials were recorded in the data base
system, including their various characteristics such as dimen-
sions and typological classication. First we present a brief
description of the chipped stones from the disturbed layer and
then we proceed to in situ assemblage.
DisturbeD Layer ChippeD stones
A total of 5149 chipped stones are recovered from an upper
disturbed layer (DL). This assemblage, though statistically
unreliable due to the disturbed context, presents an almost dif-
ferent assemblage from those of in situ layers regarding tech-
nology and tools, though the raw material is similar. The most
prominent difference between these two is pressure micro-
blade cores (‘bullet cores’), which are abundant in disturbed
layers and indicate systematic pressure removal of bladelets
(g. 4) in the Neolithic, which is not found in the in situ
Mesolithic layers. Blade/lets are detached with parallel ridges
and edges. Thirteen associated core tablets have also been
075-094-Jayez.indd 78 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
A separation: Caspian Mesolithic
vs
Trialetian lithic industry
79
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
Fig. 4 – Cores and by-product from DL layer of Komishan. 1-7, 9-11) Pressure microblade cores;
8) core tablet (drawing by M. Jayez).
075-094-Jayez.indd 79 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
80 M. J
and H. V
N
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
found from DL (g.4: 8). Other types of cores such as ake
and mixed cores are also abundant.
Technological structure of chipped stone assemblage of
DL is not much different from in situ layers (table1) and pres-
ents a very high percentage of akes and chips, though the
latter is underrepresented because of lack of otation in recov-
ering DL chipped stones. Tools are more abundant and repre-
sent a wider range of forms in DL (table2). Various scrapers
are present in the assemblage including side (g.5: 23), end
(g.5: 22, 24), round and thumbnail (g.5: 18), carinate and
convergent scrapers; they are relatively large in dimensions;
notch-denticulates are abundant and various in form (g.5:
1-6, 10, 16-17, 19-20). Burins (g.5: 9, 21) and borers are not
much different from those of in situ layers. DL tools are
mostly different from the in situ assemblage with few backed
tools (g.5: 7-8), lack of lunates, presence of drills (g.5: 11)
and geometrics (g.5: 12-15). Geometrics, though attributed
to Mesolithic industry, are also found in Neolithic sites of the
region, although with sickle gloss on the edges (Jayez personal
observation).
CHIPPED STONE ASSEMBLAGE FROM IN SITU
MESOLITHIC LAYER
The lithic assemblage from the in situ layer of Komishan is
studied as a whole because not only the deposit from which the
assemblage is recovered is homogeneous in characteristics but
also the assemblage itself does not reect any major inter-level
differences and is homogeneous from levelI to XIII regarding
technological and morphological characteristics (tables3 and 4).
Raw material and technological organization
In terms of the acquisition and variability of raw materials,
the Komishan lithic collection could be considered completely
local. No artifact is made of exotic raw materials, likely due to
Table 1 – Technological composition
of chipped stone assemblage of DL.
N %
Debitage 3694 71.74
Tool 645 12.41
Core and frag. 162 3.14
Debris 84 1.63
Chip 564 10.95
Total 5149 100
Table 2 – Tools composition of DL assemblage.
Tool % N Subtype N
Retouched 51.47 332
Notch-Denticulate 22.63 146
Scraper 13.02 84
Side 50
End 19
Carinate 1
Round and thumbnail 10
Other 4
Perforator 1.39 9
Borer 6
Drill 3
Burin 2.32 15
Backed 2.79 18
Geometric (without
lunate) 0.77 5
Multifunction 4.03 26
Other 1.55 10
Total 645
the exceptional proximity of high quality sources of chert. The
geological surveys of the region have shown that there is a long
thick outcrop of high quality chert, which is situated in the
Hotokash Mountains between the cities of Neka and Galugah
(less than 2 km away from the site; see Heydari 2004). The
good quality ne grain cherts also are found in the form of
nodules embedded in the limestone deposits. Both types are
homogeneous, with an opaque white cortex and a reticulate
pattern on the surface. The cores are also homogeneous but
translucent and milky white to pale yellow.
From the quantity of cores at the site, it seems that at least
some of the raw materials were brought to the site in prepared
form: cores constitute 0.28% of the lithic assemblages of in
situ layers. The average ratio of cores to akes is 1:47.68, and
the ratio of total cores to total akes is 1:49.4. Slightly more
than 2% of the assemblage (n=184) consists of tools. Flakes are
the most abundant blank form, but there are few ake tools. It
seems likely that a large number of akes are actually the by-
products of core reduction process at the site.
Only four complete cores, which are either ake or mixed
cores, were recovered from the in situ levels. These pieces have
more than 10 removal scars and less than 30% cortex. The only
true blade/bladelet core, with a prepared platform and 13
removals (g. 6: 2), was found in level VIII; yet numerous
blades/bladelets are not accompanied by their cores.
Thanks to the otation signicant amounts of chips (non
utilized akes which have both a length and width less than
20mm) were recovered. It can be assumed that chips under
075-094-Jayez.indd 80 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
A separation: Caspian Mesolithic
vs
Trialetian lithic industry
81
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
Fig. 5 – Tools from DL layer of Komishan. 1- 6, 10) Notched blade/ lets; 7-8) backed bladelets; 9, 21) burins;
11) drill; 12-15) geometrics; 16-17, 19-20) notched akes; 18, 22-24) scrapers (drawing by M. Jayez).
075-094-Jayez.indd 81 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
82 M. J
and H. V
N
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
Fig. 6 – Cores and tools from in situ layers of Komishan. 1-2, 4) Cores; 3, 5 -7) scrapers;
8-10) perforators; 11) burin (drawing by M. Jayez).
075-094-Jayez.indd 82 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
A separation: Caspian Mesolithic
vs
Trialetian lithic industry
83
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
5mm in length and width are by-products of retouching and
trimming. This group constitutes more than half of the assem-
blage compositions (table3). So the ake category consists of
akes larger than 20 mm in both width and length (8.67%),
those less than 20mm in both width and length, which were
recovered in screen meshes of 15mm (1.88%), 10mm (3.74%),
5mm (16.13%), and those passing through the 5 mm—i.e., less
than 5mm in length and width (52.75%).
The technological structure is almost similar in all exca-
vated in situ levels, except that there is a slight increase of
cores in the lower levels, which is insignicant because almost
half of them have more than 50% cortex, and more than half of
them are irregular cores made on amorphous chert nodules.
It seems that various blank forms were differentially
selected for use. Less than 50% of the akes were used as
tools. Flake tools consist of 0.67 to 4.78% of the total assem-
blages of the in situ levels (i.e., an average of 1 out of 8 to 1 out
of 3 akes was made into tools). Such a pattern might indicate
that many akes were just trimming akes or by-products of
blade/let manufacturing process. Almost the same pattern
exists in the case of bladelets. Fewer bladelets were turned into
tools than those remaining unutilized (between 1 out of 2 to 1
out of 7 of each level’s bladelets are tools). On the other hand,
blades obviously show a different pattern: in the upper levels,
blade tools are denitely dominant over non-tools (between
55.17 to 100% in levels I-VIII, with the exception of V), and
conversely less in the lower levels (30.76 to 50% in levels IX to
XII), with once again an increase in the lower-most level
(66.66% in levelXIII). Generally it seems that more blades
were turned into tools than akes and bladelets.
A large portion of blanks and tools are broken akes,
blades and bladelets. These broken pieces are mostly distal
ends, which are always more than twice the number of proxi-
mal ones and much more than medial parts. In this research
the broken pieces were once assessed as individual pieces and
another time quantied by estimating the Minimum Number
of Flakes (MNF, see Hiscock 2002). Ultimately the MNF of
each category was used as the basis for most of the statistics.
Technology
As it was mentioned, in the case of core quantity and core
reduction technology, there are major differences between dis-
turbed layers and the in situ ones. As was discussed earlier,
only a few cores were recovered from the in situ layers, most of
which are irregular ake cores followed by uni/multidirec-
tional mixed, unidirectional blade/bladelet and uni/multidirec-
tional ake cores. The presence of a few retted pieces (two
akes and one blade from AXIlevel) indicates that some lim-
ited int knapping activities were carried out at the site. Less
than 1% of the blanks have a double bulb; 14% of akes, 15%
of blades and 3.6% of bladelets show an eraillure ake on the
bulb (g.6: 3, 11). Few blade/let cores (g.6: 2) show parallel
ridges in the negative scars; the same is observed in the blade/
let edges, which indicates that pressure technique was not yet
introduced in the in situ layers. It is worth mentioning that
twisted blades are also represented by few examples in the
assemblage.
Table 3 – Technological composition of chipped stone assemblages from excavation levels.
Level Debitage Tool Debris Chip Core and Fragment Total
%n%n%n%n%n
I (C.1) 52.94 36 14.71 10 14.71 10 17.65 12 0.00 0 68
I12.83 63 5.50 27 19.76 97 61.91 304 0.00 0 491
II 12.84 14 3.67 48.26 973.39 80 1.83 2 109
III 9.34 59 2.53 16 11.23 71 76.90 486 0.00 0 632
IV 8.92 161 1.16 21 10.42 188 79.45 1434 0.06 1 1805
V18.80 75 1.50 610.53 42 69.17 276 0.00 0 399
VI 13.86 46 3.01 10 7.83 26 75.00 249 0.30 1 332
VII 14.45 99 2.48 17 5.11 35 77.96 534 0.00 0 685
VIII 19.07 143 3.60 27 3.07 23 73.60 552 0.67 5 750
IX 15.54 39 3.98 10 3.98 10 75.30 189 1.20 3 251
X6.72 37 1.63 94.54 25 86.93 479 0.18 1 551
XI 11.53 49 3.06 13 3.29 14 81.65 347 0.47 2 425
XII 5.20 39 0.93 74.13 31 89.73 673 0.00 0 750
XIII 7.31 43 1.19 715.99 94 74.32 437 1.19 7 588
Total 11.52 903 2.35 184 8.61 675 77.23 6052 0.28 22 7836
075-094-Jayez.indd 83 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
84 M. J
and H. V
N
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
Tools and retouched pieces
Tools in the chipped stone assemblage of Komishan are
heterogeneous. The main component of the tools consists of a
predominance of backed blades and bladelets, many in the
form of lunates (table4; g.7: 1-11, 13-34), retouched pieces,
and notch/denticulates (g.8). Some backed pieces have a dou-
ble edge back with a pointed tip (g.7: 32-33).
Scrapers are present in the assemblage in a variety of forms
(gs.6: 3, 5-7; 8: 3, 18) and there are few borers (g.6: 8-10).
Other tool types such as burins (g.6: 11), geometrics other
than lunates (g.7: 12), a heavy duty scraper, and a probable
point exist in relatively low quantities (table4).
The overall characteristics of the Komishan tool kit consist
in an absence of symmetric points, and an abundance of
numerous backed and retouched blades. There is no large vari-
ation in tool types between the in situ layer, except for total
decrease in the number of tools from level IX through XIII
(table4). Such a small uctuation should not be assumed to be
a marked change in the cultural phases, but is more likely
reective of minor changes in the activities taking place during
the occupations at the site.
As previously discussed, the most distinguishing feature of
the Komishan lithic tools is the abundance of backed pieces
(blades, bladelets, and their segments). There are no major dif-
ferences between simple backed pieces with one edge blunted
via abrupt retouch and lunates, which have a geometric shape
in the form of a crescent (the backing forms the curve and the
opposite edge is almost a straight line). Both are usually trian-
gular in cross section (with their back forming the shortest
side). These pieces are present in all levels; as is the case in the
Mesolithic of Kamarband Cave, they are mostly made on
broad blanks and the edge opposite the back is sometimes
nely retouched. Common forms of large geometrics are
absent in the assemblage from in situ layers of Komishan
assemblage, so the lunates are the only geometrics (and a sin-
gle trapezoid from AIX; g.7: 12).
Notch/denticulate pieces are one of the most diverse cate-
gories. Notches are made either with a single blow or multiple
removals of small akes from one or both edges of akes,
blades and bladelets. Notches are mostly made on large akes,
but there are also blades with multiple notches on two edges.
Denticulates are mostly formed by deep adjacent single blows
on both ake and blade edges. Bladelets are rarely turned into
notch/denticulates (g.8).
Various types of scrapers are represented in the assem-
blage, including side, end, and atypical thumbnail scrapers
(g.6: 3, 5-7). Side and end scrapers account for more or less
the same percentage of the tool assemblage (table 4). The
combination of side and end scrapers with notches simultane-
ously on the same piece occurs in some examples (g.8: 3, 18;
although they are categorized as ‘multifunctional tools’, they
are counted in the scraper analysis; see table 5). Generally
speaking, the Komishan scrapers are made on large blanks
with direct continuous retouch, but inverse and alternate
retouch also is present. Thumbnail scrapers are few and are
different from ordinary thumbnail scrapers in that they have
alternating semi-abrupt retouch, which leads to an almost
blunt edge around the pieces (g.6: 6).
Despite the presence of some perforated pendants made of
Table 4 – Tools composition in Komishan assemblage.
Level Side
scraper End scraper Borer Notch-
Denticulate Retouched Multi -
function Backed Lunate Other Total
I (C.1) 1 0 0 0 2 2 3 1 1 10
I 10159271127
II0010002014
III10023064016
IV12024082221
V0000301206
VI01032040010
VII03025124017
VIII 2 2 0 2 10 0 3 7 1 27
IX12013021010
X0002421009
XI00017040113
XII1003102007
XIII0002400017
Total 8 10 2 25 57 7 45 22 8 184
% 4.34 5.43 1.08 13.58 30.97 3.8 24.45 11.95 4.34
075-094-Jayez.indd 84 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
A separation: Caspian Mesolithic
vs
Trialetian lithic industry
85
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
Fig. 7 – Blade/let tools from in situ layers of Komishan. 1-11, 13-31, 34) Backed/lunate; 12) geometric;
32-33) double edge backed; 35) retouched blade (drawing by M. Jayez).
075-094-Jayez.indd 85 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
86 M. J
and H. V
N
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
Fig. 8 – Tools from in situ layers of Komishan. 1-2, 4-17, 19-20) Notch /denticulates; 3, 18) Scraper and notch (drawing by M. Jayez).
075-094-Jayez.indd 86 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
A separation: Caspian Mesolithic
vs
Trialetian lithic industry
87
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
animal teeth, only a few perforators were found in the lithic
assemblage. All perforators are exclusively made on akes
using dorsal retouch (g.6: 8-10). Burins are also very rare and
are of the simplest kind, made by a single or double blow on
the lateral side of the blank (g.6: 11).
Metric analysis
The metric analysis of the Komishan lithic assemblages
indicate the presence of some patterns concerning the tools
and debitage. In almost every dimension, tools are larger in all
types of blanks (g.9).
The same pattern exists in the case of blade/let width
(g.10). In this category no tool is made on a blank under 6
mm in width, and in widths larger than 17mm there are many
more tools than blanks. There seems to be a trend to produce
broader blades in order to use or turn them into tools. In the
case of bladelets, tools are more elongated (length to width
ratio = 3.59) than simple bladelets (2.09). Altogether, the blades
in the Komishan assemblage are broad and more like elon-
gated akes than slender parallel edged pieces.
The dimension analysis of the backed category indicates that
most of the backed pieces and lunates are 6 to 13mm in width
(mostly between 8 and 9mm), a somewhat standard pattern.
Fig. 9 – Tool s vs Debitage dimensions in Komishan (mm).
Fig. 10 – Blade/ bladelet width in tool and debitage in Komishan.
075-094-Jayez.indd 87 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
88 M. J
and H. V
N
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
DISCUSSION
The lithic assemblage of Komishan is mostly representa-
tive of base camps. At the same time it seems that almost the
entire chaîne opératoire took place at the site (except prelimi-
nary core preparation). The Komishan assemblage is charac-
terized by a large quantity of akes probably indicating
knapping activities at the site and a slightly smaller frequency
of blade/lets, as well as a moderate percentage of tools and a
small number of cores.
The exact interpretation of tool functions is not possible at
the moment (although planned use-wear studies may provide
data). With respect to L.R.Binford’s models of hunter-gatherer
strategies (Binford 1980), it is not easy to specify the place of
Komishan because our picture of site distributions in the region
is so fragmentary, partly due to the environmental conditions
of the region under which Komishan is located. But the diverse
tool assemblage of Komishan suggests a wide range of activi-
ties, as a high value of diversity in artifacts is more indicative
of a base camp than a special-purpose site (Andrefsky 1998:
204-206). This is supported by the evenness index of assem-
blage (0.73), which indicates that all tool types are almost
equally present in the assemblage of in situ levels.
Scrapers at Komishan are relatively large and the side
scrapers are larger than the end scrapers (table5). Both catego-
ries are made on broad blanks (length to width ratio for side
scrapers is 1.35 and for end scrapers is 1.38). Denticulates are
rather big and side-scrapers sometimes have denticulate
retouch on the edges (g. 8: 3).
Geometrics other than lunates are almost completely
absent. There are two explanations for this: rst, as observed at
Kamarband (Coon 1951), the Mesolithic industry in the region
consists of two phases and Komishan is in the second phase
(Late Mesolithic), which lacks trapezoids and triangles; and,
second, regarding the function of the geometrics as compound
tools (Neeley and Barton 1994) or arrow components
(Olszewski 1993), lack of them in the assemblage could be the
result of activities which have taken place.
The chipped stone assemblage of Komishan is obviously
most similar to those of Kamarband, Hotu and Alitepe regard-
ing their locations (g.1), which are in the same climatic, geo-
logical and ecological conditions. All these sites share the
accessibility of high quality raw material and their chipped
stone assemblages are made of chert (Heydari 2004).
Unfortunately it is not possible to compare the technological
structure of Alitepe chipped stone assemblage with the other
ones, because McBurney (1968) failed to present the amount of
technological categories, i.e. cores, debris, tools and blanks. In
Hotu (Dupree 1952), there are 84.3% blanks, 11.1% tools, 4.6%
cores and core fragments; Dupree has not mentioned debris at
all. The same is observed in Kamarband statistics. It is neces-
sary to note that the quantity of chipped stone assemblages
from Hotu and Kamarban comes from an extended excavation,
while in Komishan we had just a 2x 2m trench. If we compare
the density of chipped stones in these sites, we nd out that the
amount of chipped stones from those three sites is denitely
less than what is expected.
In Table6 we have tried to estimate the minimum volume
of excavated soil based on the dimensions of the trenches pre-
sented in Kamarband, Hotu and Alitepe excavation reports;
dividing the total number of artifacts from the sites to the
minimum volume of excavated soil in each of them could give
us a density scale—unfortunately we cannot estimate the total
number of chipped stones in Alitepe due to the lack of statis-
tics in McBurney’s 1968 report. In this table we have excluded
the chips from Komishan statistics, because in the time when
Kamarband and Hotu were under excavation, there was no
awareness of otation techniques and since the chips are
mostly recognized in otation, we could expect that the total
number of chipped stones from these two sites does not
include chips. The table clearly indicates that although we
have excluded the chips, and excavation volumes are minimal,
still the density of chipped stones in Komishan is more than
Table 5 – Dimensions of side scrapers and end scrapers (mm).
Side S. End S.
Max. L 71.05 58.4
Min. L 39.27 27.71
Max. W 62.49 44.14
Min. W 24.3 17.88
Average L 53.47 45.56
Deviation 11.95 9.73
Average W 42.54 30.25
Deviation 12.36 9.57
Table 6 – Lithic density per cubic meter of excavation in Mesolithic
in the sites of southeast of the Caspian Sea.
Site Excavated
soil (m3)
Chipped stones
(without chips)
Density of chipped
stones per m3
Komishan ~1.5 2632 1754.5
Kamarband ~189 17163 90.8
Hotu ~24 2874 119.7
Alitepe ~516 ? ?
075-094-Jayez.indd 88 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
A separation: Caspian Mesolithic
vs
Trialetian lithic industry
89
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
10 times more than in Kamarband and Hotu. We might not
be wrong if we assume that C.S.Coon (1951) has not collected
all the chipped stones from Kamarband and Hotu, or at
least he did not include all of the chipped stones in the statis-
tics he presented. This should be noticed in all statistical
comparisons.
Table 7 shows the technological structure of the sites. In
this table we have excluded the debris and chips from Komishan
statistics, because they had not been taken into account in the
assemblages from Kamarband and Hotu. Alitepe report
(McBurney 1968) provides no statistics in this regard. It is
clear in Table7 that the technological structure of Komishan is
mostly similar to the Upper Mesolithic of Kamarband.
It is almost impossible to compare many aspects of tech-
nological and typological characteristics of the chipped stone
assemblage of Komishan with those of Hotu, Kamarband and
Alitepe due to lack of information and because the archaeolo-
gists who described those assemblages have used different
terms in their classications (especially regarding tools). For
instance, in describing the tools from Kamarband (Coon
1951), side scrapers have not been segregated and notches are
included within scrapers category; there is no clear denition
for ‘knives’ and those which are presented as ‘beak’ are not
common in recent classications; in the descriptions of Hotu
assemblage (Dupree 1952), knives and side scrapers are in the
same category and there is no clear denition as to what he
has called “points”; in the description of Alitepe assemblage
(McBurney 1968), there is no side scraper and notch/denticu-
late category in tools classication, whereas our revisionary
classication of those parts of Alitepe chipped stones which
are kept at the National Museum of Tehran indicated that
there are numerous scrapers and notch/denticulate tools in
this assemblage.4 It is clear that there is no possibility for com-
paring the percentage of tools among Caspian Mesolithic
sites.
Putting aside the percentages of tools, there are some simi-
larities and differences between the total characteristics of tool
typology among Caspian Mesolithic sites. Table8 shows the
most signicant similarities between Komishan and the Upper
Mesolithic of Kamarband. Most remarkable differences from
Komishan assemblage are observed in comparison to the
Lower Mesolithic of Kamarband. It is also clear that Hotu is
closer to Komishan than to Alitepe. As we noticed earlier, the
Upper Mesolithic of Kamarband has also the most similar
assemblage to Komishan regarding technological structure. By
comparison, Hotu seems to have a position between the Upper
4. Jayez
M
.,
Chipped Stone Assemblage of Alitepe (in the repertoire of
National Museum of Iran. Unpublished report. National Museum of Iran,
Department of Palaeolithic studies, 2011.
Table 7 – Technological composition of chipped stone assemblages from Mesolithic sites in the southeast of the Caspian Sea.
Sites Cores and Frags. Blanks Tools Sum
n%n%n%
Komishan 22 1.98 903 81.42 184 16.59 1109
Alitepe ? ???4232 ??
Hotu 132 4.6 2422 84.27 320 11.13 2874
Kamarband (11-17) 5 0.3 1397 86.02 222 13.66 1624
Kamarband (18-21) 16 1.32 111 8 92.86 70 5.81 1204
Kamarband (22-27) 419 3.26 12275 95.58 148 1.15 12842
Table 8 – Comparison of lithic assemblages from Mesolithic sites in the southeast of the Caspian Sea.
Sites Large End
Scraper
Denticulate
Side Scraper
Lack of
Burin Borer Lack of
Point
Large
Backed
Lack of Large
Geometric
Centripetal
Core ΣSource
Komishan 1 1 1 1 1 1 1 1 8 Vahdati Nasab et al. 2011
Vahdati Nasab and Jayez 2011
Hotu 1 1 1 1 1 0 1 1 7 Dupree 1952
Kamarband
(11-20) 1 1 1 1 1 1 1 1 8
Coon 1951
Kamarband
(24-28) 1 1 1 1 1 0 0 0 5
Alitepe 1 1 1 1 1 1 0 0 6 McBurney 1968
Jayez 2011
075-094-Jayez.indd 89 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
90 M. J
and H. V
N
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
Mesolithic of Kamarband and Alitepe on the one hand and the
Lower Mesolithic of Kamarband on the other hand.
As we mentioned above, the Mesolithic chipped stone
assemblage of the southeast of the Caspian Sea has been
attributed to Trialeti (Kozłowski 1999; but we should note that
there are some scholars who did not use such a terminology,
e.g. see Coolidge 2005). But in the descriptions of Gabunia
regarding the characteristics of Trialeti there are some differ-
ences observed to what Kozłowski has described. For example
Kozłowski (1999: 46) states that “the presence of a single
arrowhead in Edzani is a complete surprise”, whereas in the
description by Gabunia of Trialeti the presence of arrowheads
is essential. The characteristics of Trialeti which Gabunia
(1976) describes, based on Edzani and Zurtaketi chipped stone
assemblages, include abundance of slender asymmetric trian-
gles, backed bladelets and presence of tang arrowheads.
Moreover, in Trialetian Industry which is attributed to the
Middle and Upper Mesolithic of Georgia, burins are more
numerous than scrapers (Meshveliani et al. 2007: 49, 52), a
characteristic which is not observed in any chipped stone
assemblage at least from the sites in the southeast of the
Caspian Sea that we have already described and also the sites
in the east of the Caspian Sea, such as Djebel and DamDam
Cheshmeh (Conolly and Schlanger 2010).
Caspian MesoLithiC or triaLetian?
The reconstruction of the cultural sequence of Trialetian
Industry was based on the assumption that “cultural develop-
ment was everywhere more or less similar” (Kozłowski 1999:
140). On the contrary, as we shall present, such assumptions
are not adjustable to concepts of Cultural Ecology, which could
help us in discussing the differences between Caspian
Mesolithic and Trialeti.
Environment is the most important variable in explaining
the differences in Cultural Ecology. Two basic divisions of the
biotic environment, which are most important in our discus-
sions, are Ecozones and Ecotones.5 An ecotone is usually a
more productive place than either of the individual ecozones
because species of both zones intermingle within it. Even in
cases where there is less diversity, an ecotone is a good place
5. An environmental zone, or ecozone, is a geographic area dened by fairly
specic biotic communities. Ecozones are most commonly dened based
on dominant plant communities because they are easy to recognize and
map, animals are harder; an ecotone is the geographic intersection of—as
well as the transition between—ecozones. Examples of ecotones include
estuaries, shorelines, and areas where forests and grasslands meet.
for an organism to be located as access to both ecozones is
easier (Sutton and Anderson 2010: 36-39).
As we have mentioned, the original Trialetian Industry
emerged in the area of the Khrami River in Georgia (ecotone
between river and steppe). Chokh, which is an important site in
the Kozłowski (1999) denition of Trialetian Industy, is located
in the mountainous area of Dagestan (mountain ecozone; g.1),
whereas the main Mesolithic sites of the Caspian Sea
(Kamarband, Hotu, Alitepe, Komishan) are all located in a tran-
sitional ecotone between the Caspian Sea and Alborz Mountains,
which are more productive regarding accessibility to various
sources of faunal, botanical, avifauna and marine food.
In the denitions of freshwater environments, still water
(lakes, ponds and marshes) is considerably more diverse regard-
ing resourses, even more than moving waters like rivers (Sutton
and Anderson 2010: 40). Among the sites, which Kozłowzki
attributes to Trialetian Industry, Caspian sites are located in a
totally different ecological zone compared to Georgia and
Dagestan sites. Even if we exclude the differences in food sources,
perhaps the most important aspect in the comparisons is the dif-
ference in access to raw material in these different regions.
The importance of raw material availability and its effect
on the organization of technology has been referred in many
investigations (Andrefsky 1994; Binford 1979; Sievert and
Wise 2001), but apparently at the time when Kozłowski was
dening Trialetian Industry the information about raw mate-
rial was only available for Hallan Çemi and Chokh. In the for-
mer site the dominant raw material (up to 60%) was obsidian
imported from Bingol, some 100km away. In Chokh a non-
local gray int predominates over two local varieties
(Kozłowski 1999: 144). In Trialetian Industry from Georgia
also the raw material is mostly exotic obsidian (Kuftin and
Field 1946: 358; Gabunia and Tsereteli 1977; Kushnareva 1997:
9). On the contrary, raw material used in Mesolithic sites in the
southeast of the Caspian Sea is locally available and high qual-
ity chert. This is not the only major difference between these
two groups of sites; we have already mentioned that there are
no tang arrowheads among the tools of Caspian sites and also
the percentage of burins is lower than scrapers in these sites,
which is not similar in original Trialetian sites. Here we dis-
cuss how the difference in raw material access and technologi-
cal organization, and in a much larger scale the subsistence
between these sites, is important in their attribution.
Much of the ecological work relating to humans was cen-
tered on diet and subsistence. Subsistence is not simply a list of
foods but a complex system that includes resources, technol-
ogy, social and political organizations, settlement patterns, and
all of the other aspects of making a living. The technological
075-094-Jayez.indd 90 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
A separation: Caspian Mesolithic
vs
Trialetian lithic industry
91
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
system is composed of the material, mechanical, physical, and
chemical instruments, together with the techniques of their
use, by means of which man, as an animal species, is articu-
lated with his natural habitat. Here we nd the tools of produc-
tion, the means of subsistence, the materials of shelter, the
instruments of offense and defense. It is mostly through tech-
nology, rather than biology, that humans have adapted to virtu-
ally every ecosystem on earth. Technology is the result of
need, available materials, innovation, and inuence from other
cultures. If one of these conditions changes, the technology
will also change. Technology is a part of culture core in Human
Ecological point of view (g. 11), because it has the most
important role in adapting to various environments.
Social systems are in a very real sense secondary and sub-
sidiary to technological systems. In fact a social system may be
Fig. 11 – Diagrammatic representation of culture core
concept (Richerson et al. 1995: 6).
dened realistically as the organized effort of human beings in
the use of the instruments of subsistence, offense and defense,
and protection. A social system is a function of a technological
system. The technology is the independent variable, the social
system the dependent variable. Social systems are therefore
determined by systems of technology; as the latter change, so
do the former. The importance of technology is reected in
J.Steward’s key idea: Technology is the “window” through
which people look at their environment. Our adaptations are
mainly technological, and how we interact with any given
environment depends rst of all on the tools we bring to that
environment (White 1949: 364-365; Steward and Shimkin
1961: 481-482; Richerson et al. 1995: 4-5, 36; Sutton and
Anderson 2010: 91-101).
What we are trying to clear is that the distinctions between
Caspian Mesolithic and Trialeti are not just differences in
chipped stone industries and we need to consider the differ-
ences between subsistence patterns and how hunter-gatherers
adapted to their environment via technology. It is clear that
regarding complete different raw material exploitation, trans-
port costs of raw material in the Caspian Sea Mesolithic indus-
try is at the lowest level, because the high quality raw material
is easily accessed, whereas in the sites in Georgia and Dagestan
the raw material should be imported sometimes from 100km
away. On the other hand, Caspian Mesolithic hunter-gatherers
needed a technology adapted to exploitation of marine and
mountain resources, while Trialetian people were engaged in a
totally different ecozone. In Table9 we have summarized the
total differences observed in Mesolithic assemblages. They
also have different cultural phases; there is continuity from
Upper Palaeolithic to Mesolithic in Trialeti of Georgia,
whereas Upper Paleolithic layers are not found in any of the
Caspian Mesolithic sites. Each of these complexes has their
Table 9 – Comparison of Caspian Mesolithic with Trialeti Mesolithic of Dagestan and Georgia.
Southeast of
Caspian Sea Dagestan Trialeti (Georgia)
Environment Ecotone (contacting zone between sea,
plain and mountains) Ecozone (mountains) Ecozone
(riparian steppe)
Raw material Local high quality Behshahr chert Local-exotic
(low quality local chert + exotic chert)
Local-exotic
(low quality local
chert + exotic obsidian)
Chipped stones
Absence of tanged arrowheads
Lack of burins
Moderate quantity of scrapers
Presence of Chokh points
Presence of tang arrowheads
Abundance of burins
Less end scrapers
Archaeological
sequence
Mesolithic
Neolithic
Mesolithic
Neolithic
Upper Palaeolithic
Mesolithic
Neolithic
075-094-Jayez.indd 91 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
92 M. J
and H. V
N
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
own characteristics. Although some similarities might be
found between them, it doesn’t mean that there is such an
extensive homogeneous industry covering all sorts of environ-
ments from Central Georgia to Western Turkmenia.
CONCLUSIONS
It seems that the sites in the southeast of the Caspian Sea
are totally different in their industry from other Trialetian
sites. It is not so unrealistic if we assume that the Mesolithic of
the east-southeast of the Caspian Sea should be separated from
Trialetian Industry. We suggest “Caspian Mesolithic” to be
used in future investigations to denote specically the sites
such as Kamarband, Komishan, Hotu, Alitepe, Dam Dam
Cheshme, Djebel and other Mesolithic sites in the region. Such
a separation does not mean that there was no connection
between Caspian Mesolithic people and other Mesolithic peo-
ple from other regions; because Mesolithic contact, exchange,
and interaction took place at scales ranging from individuals to
local bands and larger regional populations, and were cata-
lyzed by a complex interplay of social and economic needs.
What we observe archaeologically are the results of complex
decisions made by Mesolithic hunter-gatherers about how they
interacted with their landscape, how they organized them-
selves socially, and how they engaged and negotiated with
their neighbors. The archaeological proofs of the differing
contacts, exchanges, and interactions at these multiple scales
vary. These signatures are embedded in the structure of the
archaeological record at the site and larger regional levels, and
are expressed in sources, kinds, styles, and contexts of specic
artifacts. No hunter-gatherer group can be understood in isola-
tion. Any hunter-gatherer group must be seen as part of a larger
regional population of social groups, which are linked and
integrated in multiple ways. This means that, even if Caspian
Mesolithic had a distinct industry from Trialetian Industry,
Caspian Mesolithic people might have been in contact or inter-
action with Trialetian people.
aCknowLeDgMents
We thank journal editor and anonymous reviewers for their con-
structive comments. We are indebted to Ali Mahfroozi, the director
of the Gohar Tepe archaeological project and deputy of research at
the Cultural Heritage and Tourism Organization in Mazandaran at
the time of the eld project for his generous support in logistics and
allocating fund to conduct the eld mission. We also thank Mehdi
Mousavi Koohpar and Hasan Fazeli, directors of the Iranian Cen-
tre for Archaeological Research at the time of the eld excavation,
who issued the excavation permits that allowed us to do the work.
We would like to thank Gilles Berillon for his help with the French
abstract. We are grateful to Esmaeel Jafarain for his tremendous sup-
port. Special thanks to the excavation and lab crew: H.I.Moghaddam,
M.Z.Khalili, R.Mojtabaee, M.Abolfathi, Z.Lotfalikhani, M.Miri
and H.Abbasi.
Mozhgan J
ayez
Hamed
V
ahdati
N
asab
Tarbiat Modares University
Jalal Ale Ahmad Highway
P.O. Box: 14115-111 – Tehran – I
ran
m.jayez@modares.ac.ir
vahdati@modares.ac.ir
BIBLIOGRAPHY
A
W
.
J
.
1994 Raw Material Availability and the Organization of Technology.
American Antiquity 59,1: 21-34.
1998 Lithics. Macroscopic Approaches to Analysis. Cambridge:
Cambridge University Press (Cambridge Manuals in
Archaeology).
B
N
.
O
.
and
T
L
.
1989 Mezolit Kavkaza (The Mesolithic of the Caucasus). In:
K
L
.
V
.
(ed.), Mezolit SSSR (The Mesolithic of the
USS R): 93-129. Moscow: Nauka.
B
G
.,
A
K
A
.,
A
P
.,
B
J.
-
J
.,
C
B
.,
Z
V
.,
A
N
.,
B
M
.,
E
C
H
.
and
N
S
.
2007 Discovery of new open-air Paleolithic loca lities in Central Alborz,
Northern Iran. Journal of Human Evolution 52,4: 380-387.
B
G
.,
A
K
A
.,
C
B
.,
A
P
.,
Z
V
.,
B
M
.,
N
S
.,
E
C
H
.,
R
R
F
.,
B
J
.-
J
.,
A
P
.
and
V
C
.
2009 Late Pleistocene Prehistory in Central Alborz: Preliminary
Results of the French and Iranian Paleoanthropological
Programme 2006 on the Excavation of Garm Roud 2 (Amol,
Mazandaran). In: O
M
.,
B
F. and J
J. (eds.),
Iran Paleolithic: 101-108. Oxford (BAR Int. Ser. 1968).
B
L
.
R
.
1980 Willow Smoke and Dogs’ Tails: Hunter-Gatherer Settlement
Systems and Archaeological Site Formation. American
Antiquity 45,1: 4 -20.
1979 Organization and Formation Processes: Looking at Curated
Technologies. Journal o f Anthropological Research 35,3: 255-273.
075-094-Jayez.indd 92 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
A separation: Caspian Mesolithic
vs
Trialetian lithic industry
93
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
B
F
.
1998 La néolithisation en Asie centrale : un état de la question.
Paléorient 24,2 : 27-48.
C
B
.,
B
G
.,
A
K
A
.,
A
P
.,
B
J
.-
J
.
et
Z
V
.
2006 Moghanak, Otchounak, Garm Roud 2 : nouveaux assemblages
paléolithiques dans le Nord de l’Iran. Premières caractérisa-
tions typo-technologiques et attributions chrono-culturelles.
Paléorient 32,2 : 59-79.
C
J
.
and
S
N
.
2010 Stone Tools from the Dam Dam Cheshme Rockshelters.
In: H
D
.
R
.
(ed.), Origins of Agriculture in Western
Central Asia: An Environmental-Archaeological Study: 203-
205. Philadelphia: University of Pennsylvania Museum of
Archaeology and Anthropology.
C
J
.
2005 Southern Turkmenistan in the Neolithic. A petrographic case
study. Oxford (BAR Int. Ser. 1423).
C
C
.
S
.
1951 Cave Explorations in Iran, 1949. Philadelphia: The University
Museum, University of Pennsylvania.
1952 Excavations in Hotu, Iran, 1951: A Preliminary Report.
Proceedings of the America n Philosophical Society 96: 2 31-
249.
D
L
.
B
.
1952 The Pleistocene artifacts of Hotu Cave, Iran. Proceedings of the
American Philosophical Society 96,3: 250-257.
G
M
.
K
.
1976 Trialetis mezolituri kultura (The Trialetian Mesolithic Culture).
Tbilisi: Metsniereba.
G
M
.
K
.
and
T
L
.
D
.
1977 Mezolit Gruzii (The Mesolithic of Georgia). Moskow:
Izdatelstvo Nauka (in Russian).
H
S
.
2004 Stone Raw Material Sources in Iran, Some Case Studies.
In: S
T
.,
S
R
.
und V
A
.
(H r sg .),
Persiens Antike Pracht. Bergbau – Handwerk – Archäologie
(Katalog der Ausstellung des Bochum: Deustsches Bergbau-
Museums Bochum von 28. November 2004 bis 29. Mai 20 05):
124-129. Bochum: Deutsches Bergbau-Museum.
H
P
.
2002 Quantifying the Size of Artefact Assemblages. Journal of
Archaeological Science 29: 251-258.
K
P
.
L
.
1984 Central Asia . Palaeolithic Beginnings to the Iron Age. Paris:
ERC (Synthèse 14).
K
S
.
K
.
1994 Chipped Neolithic Industries at the Eastern Wing of the
Fertile Crescent (Synthesis Contribution). In: G
H
.
G
.
and
K
S
.
K
.
(eds.), Neolithic Chipped Stone Industries of
the Fertile Crescent: 143-172. Berlin: ex Oriente (SENEPSE 1).
1996 The Trialetian ‘Mesolithic’ Industry of the Caucasus,
Trans caucasia, Eastern Anatolia, and the Iranian Plateau.
In:
G
H.G. and
K
S.K. (eds.), Neolithic
Chipped Stone Industries of the Fertile Crescent, and their
Contemporaries in Adjacent Regions: 161-170. Berlin: ex
Oriente (SENEPSE 3).
1999 The Easter n Wing of the Fertile Crescent: Late Prehistor y of
Greater Mesopotamian Lithic Industries. Oxford (BAR Int. Ser.
760).
K
S
.
K
.
and
A
O
.
2005 Territories, Boundaries and Cultures in the Neolithic Near
East. Oxford (BAR Int . Ser. 1362).
K
B
.
A
.
and
F
H
.
1946 Prehistoric Culture Sequence in Transcaucasia. Southwestern
Journal of Anthropology 2,3: 340-360.
K
K
.
K
.
1997 The Souther n Caucasus in Prehistory: Stages of Cultural and
Socioeconomic Development from the Eighth to the Second
Millennium B.C. Philadelphia: The University of Pennsylvania
Press (Universit y Museum Monograph 99) (Translated by
H.N.Michael).
M
A
.
2003 A Preliminary Report on Archaeological Surveys and
Excavations in East of Mazandaran Province. Archaeological
Reports 2: 263-278 (in Persian).
M
M
.
2010 Faunal Remains f rom the Epipalaeolith ic Site of Komisha n Cave
and its Dating, Preliminary Results. Iranian Archaeology 1:
32-37.
M
V
.
M
.
and
S
V
.
I
.
1972 Central Asia . Turkmenia before the Achaemenids. London:
Thames & Hudson (Ancient peoples and places 79) (Translated
by R.Tringham).
M
B
C
.
B
.
N
.
1964 Preliminary Report on Stone Age Reconnaissance in North-
Eastern Iran. Proceedings of The Prehistoric Society 30: 382-
399.
1968 The Cave of Ali Tappeh and the Epipalaeolithic in N.E. Iran.
Proceedings of The Prehistoric Societ y 34: 385-413.
M
T
.,
B
-
O
G
.,
B
-
Y
O
.,
B
-
C
A
.,
B
E
.,
J
N
.,
K
I
.
and
M
Z
.
2007 Mesolithic Hunters at Kotias Klde, Western Georgia: Preli-
minary Results. Paléorient 33,2: 47-58.
N
M
.
P
.
and
B
M
.
C
.
1994 A New Approach to Interpreting late Pleistocene Microlith
Industries in Southwest Asia. Antiquity 68,259: 275-288.
O
D
.
I
.
1993 Zarzian Microliths from Warwasi Rockshelter, Iran: Scalene
Triangules as Arrow Components. In: P
G.,
B
H.M. and M
P
.
(eds.), Hunting and Animal
Exploitation in the Later Palaeolithic and Mesolithic of
Eurasia: 199-205. Washington, D.C.: Americ an Anthropologica l
Association.
P
P
.
and
E
M
.
2002 Encyclopedia of Prehistory, Volume 8: South and Southwest
Asia. New York: Kluwer Academic/Plenum Publishers.
075-094-Jayez.indd 93 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
94 M. J
and H. V
N
Paléorient, vol. 42.1, p. 75-94 © CNRS ÉDITIONS 2016
R
P
.
J.
,
M
M
.
B
.
and
V
B
.
1995 Principles of Human Ecology. Pearson Custom Publishing.
[http://www.des.ucdavis.edu/faculty/Richerson/Books
Online/101text.htm [accessed 4 Nov. 2015]
R
M
.
1994 A Preliminary Description of the Lithic Industry from Hallan
Çemi. In: G
H
.
G
.
and K
S
.
K
.
(eds.), Neolithic
Chipped Stone Industries of the Fertile Crescent: 223-238.
Berlin: ex Oriente (SENEPSE 1).
S
A
.
K
.
and
W
K
.
2001 A Generalized Technology for a Specialized Economy: Archaic
Period Chipped Stone at Kilometer4, Peru. In: A
W.
Jr. (ed.), Lithic Debitage. Context, Form and Meaning: 80-105.
Salt Lake City: The University of Utah Press.
S
J
.
and
S
D
.
B
.
1961 Some Mechanisms of Sociocultural Evolution. Daedalus 90,3:
477- 497.
S
M
.
Q
.
and
A
E
.
N
.
2010 Introduction to Cultural Ecology. Lanham: Altamira Press.
V
N
H
.
and
J
M
.
2011 Techno-Typological Analysis of Komishan Lithic Assemblage.
Archaeology and Histor y 50: 56-78 (in Persian).
V
N
H
.,
J
M
.,
H
N
A
.,
K
N
F
.,
I
H
.
and
M
A
.
2011 Preliminary Report of Excavation in Komishan Cave,
Mazandaran, Iran. Antiquity 85,328, Project Gallery. [Online:
www.antiquity.ac.uk/projgall/nasab328]
W
L
.
1949 The Science of Culture: A Study of Man and Civilization. New
York: Grove Press.
075-094-Jayez.indd 94 06/06/16 14:36
CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART • CNRS ÉDITIONS - TIRÉS À PART
