Salt mining tools and techniques from Duzdaği (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.

Abstract

The emergence of mining reflects profound changes in the organization of late prehistoric societies. In terms of lithic and ore mining, salt is a highly strategic mineral resource which was exploited for its dietary and healing properties. The exceptional prehistoric salt mine of Duzdagi (Nakhchivan, Azerbaijan) is located in a high salt-bearing region centered on the Aras River in the Lesser Caucasus. Field survey, spatial recording of archaeological materials and use-wear analysis of macrolithic tools have allowed us to outline the mining operations and techniques there from the Chalcolithic period onwards. We have identified workshops for the manufacture and repair of tools, as well as salt extraction and processing areas.

Full text

Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=yjfa20
Download by: [Caroline Hamon] Date: 20 July 2016, At: 01:04
Journal of Field Archaeology
ISSN: 0093-4690 (Print) 2042-4582 (Online) Journal homepage: http://www.tandfonline.com/loi/yjfa20
Salt mining tools and techniques from Duzdaği
(Nakhchivan, Azerbaijan) in the 5th to 3rd
millennium B.C.
Caroline Hamon
To cite this article: Caroline Hamon (2016) Salt mining tools and techniques from Duzdaği
(Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C., Journal of Field Archaeology, 41:4,
510-528, DOI: 10.1080/00934690.2016.1201615
To link to this article: http://dx.doi.org/10.1080/00934690.2016.1201615
View supplementary material
Published online: 14 Jul 2016.
Submit your article to this journal
Article views: 2
View related articles
View Crossmark data

Salt mining tools and techniques from
Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th
to 3rd millennium B.C.
Caroline Hamon
CNRS, France
The emergence of mining reflects profound changes in the organization of late prehistoric societies. In terms
of lithic and ore mining, salt is a highly strategic mineral resource which was exploited for its dietary and
healing properties. The exceptional prehistoric salt mine of Duzdag
˘i (Nakhchivan, Azerbaijan) is located
in a high salt-bearing region centered on the Aras River in the Lesser Caucasus. Field survey, spatial
recording of archaeological materials and use-wear analysis of macrolithic tools have allowed us to outline
the mining operations and techniques there from the Chalcolithic period onwards. We have identified
workshops for the manufacture and repair of tools, as well as salt extraction and processing areas.
Keywords: salt production, mining techniques, Chalcolithic, Bronze Age, Caucasus, use-wear analysis
Introduction
The emergence of mining mirrors profound changes
affecting the organization of late prehistoric societies.
With regard to the procurement of ore resources,
different levels were established within competitive
contexts of access to the most valuable and the most
strategic resources. These levels relate to dietary, tech-
nical and symbolic aspects of the resources. Between
the 6th and 5th millennia B.C., mining emerged on a
relatively large scale on every continent and involved
a large range of raw materials. Flint, quartzite or obsi-
dian for the manufacture of stone tools (e.g., Díaz-del-
Río et al. 2006;Di Lernia et al. 1995;Tarantini and
Galiberti 2011), minerals used as coloring agents
(Goldenberg et al. 2003;Larocca 2012), and various
ores (Ambert et al. 2009;Stöllner and Gambashidze
2011) were exploited for their high technical, social
and symbolic value. The mobilization of groups
involved in mining is expressed in terms of a skilled
labor force and specialization as well as in the organ-
ization of, and even control over, the exchange net-
works of raw materials and of products, which were
frequently sourced over relatively long distances
(Craddock and Lange 2003;Stöllner et al. 2003).
Salt occurs in different natural states such as sea
salt, vegetable salt, brine or rock salt. Frequently
exploited for its dietary, physiological or healing prop-
erties (Denton 1984), salt is also used for a wide
variety of other purposes relating to animal husban-
dry, food storage, milk production or skin treatment,
all of which were economic activities which underwent
significant development beginning in the Neolithic
period. Salt is also a highly valuable product in
ritual practices and in exchange networks, as demon-
strated in several ethnographic and prehistoric con-
texts (Cassen and Weller 2013;Petrequin et al. 2001).
The prehistoric salt mine at Duzdag
˘i (Nakhchivan,
Azerbaijan) is located in a salt-bearing region centred
on the Aras River, between the northwestern part of
Iran and the eastern part of Turkey (Aali et al. 2012)
(FIG.1). The Duzdag
˘i salt source (literally “Salt
Mountain”in Azeri Turkish) dominates the left
bank of the Aras River, seven km from the current
city of Nakhchivan (Azerbaijan). These Miocene salt
deposits are distributed over an area measuring 3 ×
2 km, reaching a height of almost 150 m. Three main
salt layers have been identified from the analysis of
Spot 5 satellite photos (Marro et al. 2010), and they
alternate with clay levels, gypsum layers and strata
containing residual limestone slabs. In the salt layers,
the NaCl content varies from 86 to 98% (Azizbekov
1961).
The Duzdag
˘i site is exceptional in terms of its size
and the density of its archaeological remains. This evi-
dence provides opportunities to study the significance
of salt exploitation for the economy of prehistoric
groups, particularly in regard to the organization of
the activities and the integration of exchange net-
works. From the end of the 19th century, the large
Correspondence to: Caroline Hamon, CNRS, UMR 8215 Trajectoires,
Maison de l’archaéologie, 21 allée de l’Université, 92023, Nanterre
cedex France. Email: caroline.hamon@mae.cnrs.fr
© Trustees of Boston University 2016
DOI 10.1080/00934690.2016.1201615 Journal of Field Archaeology 2016 VOL. 41 NO. 4510
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

number of mining tools discovered in the area
attracted the attention of archaeologists. Based on
comparisons with mining tools found at the Kültepe
I and II dwelling sites, these remains were initially
dated to between the end of the 3rd and the beginning
of the 2nd millennium B.C.(Abibullayev 1982;Aliyev
1991). New investigations were initiated in 2007 by
the French-Azerbaijani Araxe Research Program
directed by C. Marro (Bakhshaliyev and Marro
2009). Thousands of finds recovered during a systema-
tic survey of Duzdag
˘i include pottery sherds and arti-
facts made from obsidian and flint as well as
macrolithic tools such as hammers and grinding
stones.
Several sherds suggest a very early date for the
exploitation of this salt source, possibly the second
half of the 5th millennium B.C., thus making
Duzdag
˘i one of the earliest known salt mines in the
world (Marro et al. 2010;Weller and Figuls 2007).
Based on the most common pottery remains,
however, the main period of exploitation at the site
has been assigned to the Kura-Araxes Culture. This
attribution is also supported by the presence of a
large number of basalt hammers at the Kültepe I
and II sites, dated to the Early and Middle Bronze
Age (Abibullayev 1982;Aliyev 1991). The site was
used during the succeeding Middle Bronze and Iron
Age as well, but with less intensity (Marro et al.
2010). This suggests that the same areas were fre-
quented, although at a smaller scale, over a long
period of time. Medieval salt extraction was carried
out in other sectors of the mine and is characterized
by the absence of hammers and by other extraction
techniques.
The distribution of pottery assigned to the Kura-
Araxes culture at Duzdag
˘i correlates with the distri-
bution of certain mining tools (Marro et al. 2010)
(see below) which reinforces the special role played
by the Kura-Araxes community in the exploitation
of important mineral resources (such as gold at the
Sakdrisi mine [Stöllner et al. 2014]). This contributes
to the definition of the genesis of the Kura-Araxes
culture throughout the southern Caucasus, eastern
Anatolia and north-western Iran (Sagona 2014).
The exact chronological framework for the emer-
gence of the Kura-Araxes culture is still under
Fig. 1. The location of the Duzdag
˘i site at Nakhchivan (Azerbaijan) (maps after M. Sauvage (CNRS, USR 3225, Nanterre) and
Wikipedia (L. Bourrichon)
Hamon Salt mining tools and techniques from Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.
Journal of Field Archaeology 2016 VOL. 41 NO. 4511
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

discussion. Some authors have proposed that it
emerged at the end of the 5th millennium B.C.
(Marro et al. 2014), significantly earlier than the com-
monly accepted date of the middle of the 4th millen-
nium B.C.(Palumbi and Chataigner 2014).
Nevertheless, all authors agree that there is a clear
rupture between the Late Chalcolithic and the Kura-
Araxes period. This break is evident in metal and
ceramic products, with the widespread appearance in
the latter period of handles on typical burnished
dark brown to black ceramics. It is also evident in
changes in domestic architecture (i.e., the abandon-
ment of multicellular rectangular structures) as well
as in burial customs (i.e., the development of collective
burials). Finally, increasingly intensive settlement in
the mountain zones, possibly linked to specialized
sheep and goat pastoralism, and the rapid spread of
the Kura-Araxes “package”(Sagona 2014) far
beyond its initial core area, demonstrate the capacity
of these groups to develop and maintain wide-
ranging exchange networks over large territorial areas.
Our study aimed to reconstruct the techniques and
the organization of salt extraction activities at
Duzdag
˘i, based on technological and functional ana-
lyses of macrolithic (ground stone) tools. Our goal
was to clarify the role played by salt exploitation in
the southern Caucasus, particularly during the 5th to
3rd millennia B.C.
Materials and Methods
Between 2008 and 2011, four survey campaigns were
undertaken at Duzdag
˘i with the aim of identifying
areas of salt exploitation within the mining landscape.
Other goals were to define the extent and the densityof
the prehistoric mining site and to understand better the
chronological framework. Following an extensive
walk-over survey on the southern and western parts
of the salt mountain, four zones were delimited and
subjected to systematic, intensive survey (Marro
et al. 2010). Selection of these zones was determined
by the abundance of archaeological materials and
specific patterns in the landscape. In these zones, geor-
eferenced plotting of all pottery sherds, obsidian arti-
facts and macrolithic tools was conducted directly in
the field using a geographic information system
(ArcGIS) combined with a DGPS system (Marro
et al. 2010). Given that no topographic map was avail-
able, a map was created using Spot 5 satellite photos.
Photographic coverage of four specific zones was
obtained by taking aerial photographs from a kite;
the photographs were then georeferenced (FIG.2).
The main elements of the landscape were documented
Fig. 2. Duzdag
˘i survey zone 1 showing the spatial distribution of the plotted stone tools based on georeferenced aerial
photographs and a Spot 5 satellite photograph © CNES (S. Sanz, CNRS after Marro et al. 2010).
Hamon Salt mining tools and techniques from Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.
Journal of Field Archaeology 2016 VOL. 41 NO. 4512
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

and mapped, in particular the wadi system and the
platforms that yielded most of the archaeological
information. Anthropogenic features visible in the
landscape were also indicated: cones formed by extrac-
tion waste and possible extraction features (e.g.,
depressions, entrances to galleries) were identified
(FIG.3). The combination and exploitation of these
different levels of cartographic data, together with
ceramic- and macrolithic tool distributions, are cur-
rently being undertaken by S. Sanz (CNRS, UMR
5140), and will not be detailed further in this paper.
In addition to plotting the location of each macro-
lithic tool, technical and functional analyses were
carried out with the aim of reconstructing both the
techniques used in their manufacture and the spatial
distribution of salt exploitation activities. The charac-
teristics of the raw materials and the morphologies,
dimensions and types of manufacture of the pieces
were directly recorded in the field and later uploaded
to a project database. Each tool was georeferenced,
then weighed, measured and photographed.
Thirty-five percent of the tools or fragments (287
out of 823 pieces) were sampled for more detailed
technological and use-wear analyses in the laboratory.
For each tool the essential categories and various cri-
teria were recorded (TABLE 1). Modifications for the
purposes of hafting were noted on the hammers
(FIG.4). Use marks were described, both macroscopi-
cally and microscopically, in terms of their mor-
phology, extent, intensity and precise distribution for
Fig. 3. The landscape around Duzdag
˘i: kite view showing salt levels and platforms on which most of the archaeological material
was found (Photograph S. Sanz).
Hamon Salt mining tools and techniques from Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.
Journal of Field Archaeology 2016 VOL. 41 NO. 4513
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

each active surface. A Nikon SMZ 800 binocular
microscope (up to 63×) was used for use-wear analysis
(Hamon 2006,2008;Adams et al. 2009).
The categories of mining tools from Duzdag
˘i (see
below) followed classification criteria used for other
mining sites: their morphology (De Pascale 2004),
their spatial distribution and types of modifications
(Garner 2013;Timberlake 2003;Zebrak 1995), their
point shapes, and their use-wear (Figuls et al. 2007;
Timberlake and Craddock 2013;Weller et al. 2007).
At the Duzdag
˘i site, technological and functional cri-
teria were more widely and accurately incorporated
than at other sites enabling us to propose a functional
approach for the macrolithic toolkit within its context
of use.
Technology and function of the hammers
At the Duzdag
˘i site the concentrations of artifacts
were characterized by a strong diversity in the func-
tional categories of macrolithic tools. Although
hammers prevailed (718 pieces representing 86% of
the toolkit), grinding tools (10 grindingslabs and
handstones, 30 pestles) and percussion tools (50 ham-
merstones and six intermediate tools) were also ident-
ified (TABLE 2). This range of tools suggests that rock
salt extraction and processing activities coexisted.
Hammer types
Hammers are extraction tools that were used to strike
the rock face by direct percussion using their distal
ends. Their working surfaces are pointed or triangular
in shape, and arises from shaping or from use. Six
main categories of hammers, each divided into sub-
types (particularly with regard to the modifications
for hafting), were defined by combining the following
criteria: raw material, shape of the tools, shaping tech-
nology, dimensions and modifications carried out to
enable hafting (TABLE 3).
Table 1 Criteria used for the description and classification of macrolithic tools, especially hammers.
Criteria Sub-Criteria Description
raw material rock type sandstones, limestones, vesicular basalt,
alkali basalts, granites, metamorphic rocks
blank cobbles, blocks
morphometry state of preservation complete, half, etc.
morphology square, oval, pointed, etc.
cross-section square, oval
dimensions
weight
number of active surfaces
shaping types removals, pecking, polishing
distribution sides, edges, back, ends
use-wear nature crushing marks, scars, polish
morphology
extent
intensity
distribution
modifications for hafting distribution faces, sides, ends
location central, distal
types notches, grooves
shaping removals, pecking, polishing
Table 2 Table summarizing the numbers of macrolithic tools
recovered from the Duzdag
˘i site (2007–2011 surveys).
Tool Type Tool Sub-Type Number
Hammers 718
A non-manufactured
tools
57
B notched tools 48
C simple grooved tools 97
D complex grooved
tools
52
E basalt tools 332
F massive tools 69
G other 61
H perforated 2
Grinding slabs/
handstones
10
Hammerstones 50
Intermediate tools 6
Pestles 30
Others 6
Total 820
Fig. 4 Locations of the modifications on the hammers (after
Pickin 1990;De Pascale 2004).
Hamon Salt mining tools and techniques from Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.
Journal of Field Archaeology 2016 VOL. 41 NO. 4514
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

The 57 hammers attributed to Type A (7.9% of the
hammers) take the form of simple cobbles, the ends of
which display splintering and bruising marks resulting
from their use. The opportunistic selection of the blanks,
as well as the rather short-term use of these tools, means
that they can be regarded as occasional hammers
probably used for secondary- or low-skilled tasks.
Simple central or distal lateral notches on 48 Type B
hammers (6.7% of the hammers) facilitated hafting
(FIG.5) of tools made from various raw materials.
The removal of one or two flakes from the sides of
the pieces may have been the preliminary stage in the
creation of lateral grooves.
Ninety-seven Type C hammers (13.5% of the
hammers) have one or two central or distal grooves
to facilitate hafting. These grooves are generally sym-
metrical and located on the faces or sides of the
tools of varied shapes and are made of sandstone,
metamorphic rocks, and alkali basalt (FIG.6).
Fifty-two Type D hammers (7.2% of the hammers)
combine at least two grooves on their faces or sides
and those at their ends. They are manufactured from
various materials, mainly basalt, alkali basalt, and
crystalline rocks. These tools evolved over their life-
times from pointed forms to more oval-shaped forms
as the result of use.
Three hundred thirty-two Type E hammers (46.2%
of the hammers) were manufactured from alkali
basalt, a particularly hard and resilient material
(FIG.7). These pieces exhibit particularly intensive
shaping by flaking and pecking and have pointed
distal ends. They bear evidence of intensive use-
cycles and encompass a wide range of weights. They
can be divided into several subtypes: those with cylind-
rical points made by pecking and were abandoned at
the site; medium-sized hammers with peripheral
grooves; quadrangular-shaped hammers with distal
grooves; and hammers having peripheral distal
grooves combined with longitudinal grooves being
the most complex modifications noted.
Sixty-nine Type F hammers (9.6% of the hammers)
have particularly significant weights (ranging from
5 kg to 80 kg) (FIG.8). Some rough massive blocks,
without any apparent wear related to shaping or use,
were intentionally taken to the extraction area, prob-
ably with a view to being used as hammers. Several
subtypes were distinguished: quadrangular-shaped
hammers usually of sandstone and with distal lateral
grooves for facilitating hafting, usually manufactured
from sandstone; quadrangular, intensively shaped
hammers usually of alkali basalt with distal grooves
on one side for hafting purposes, or with peripheral
grooves; spherical hammers usually of sandstone
with distal peripheral grooves to facilitate hafting;
oval-shaped hammers with central peripheral grooves
(various materials, sometimes with two working sur-
faces); and oblong, triangular-shaped hammers with
or without modifications and having significant
removals at one end.
This classification system focuses on level of com-
plexity of the modifications made to the hammers to
allow hafting, and thus, on the underlying operating
mode of the respective tool(s). What is more, those
hammers manufactured from alkali basalt (type E),
and, in a different way, the massive tools (type F),
follow different practices; their characteristics suggest
that these tools held particular status amongst the
mining tools, perhaps related to a specific function
within the operational sequence (see below).
Table 3 Main characteristics of the different types of hammers identified at Duzdag
˘i.
Type Raw material Shape Manufacturing Hafting preparation Use
A varied varied none; natural cobbles none polyvalent tools
B varied varied none; natural cobbles central, distal or lateral notches polyvalent tools
C varied varied none or occasional pecking simple central or distal groove extraction tools
D varied varied none or occasional pecking complex face and end grooves extraction tools
E alkali basalt pointed tools flaking and fine covering pecking varied specialized hammers
F hard rocks massive none or occasional pecking varied specialized hammers
Fig. 5. Type B hammers showing notches made by
percussion, flake removals and macrotraces from use
(Photographs C. Hamon).
Hamon Salt mining tools and techniques from Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.
Journal of Field Archaeology 2016 VOL. 41 NO. 4515
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

Raw materials and hammer manufacture
Two distinct strategies of management of the lithic
blanks characterize the hammers made on cobbles
on one hand, and made of alkali basalt on the other
hand. In most instances the hammers of Duzdag
˘i
were likely manufactured from cobbles. Such stone
blocks are locally available in the outcropping alluvial
layers of the Aras River, still very apparent at the
summit of the Duzdag
˘i site, while the more recent allu-
vial deposits of Nakhchivan çay may have been a sec-
ondary source of procurement (Azizbekov 1961). For
the whole hammer assemblage, rock types identified
included rocks of sedimentary- (9.8% sandstone and
3.7% limestone), volcanic- (48% alkali basalt and 5%
vesicular basalt), plutonic- (1% granite) and meta-
morphic- origin (10%). Due to their highly variable
grain sizes, porosities and structures, these materials
differ greatly in terms of their mechanical qualities
and resistance when used for percussion. This diversity
is also reflected in the use of certain rocks for the
manufacturing of particular types of hammers:
basalts and volcanic rocks were preferred for the
manufacture of hammers with complex hafting modi-
fications, whereas quartzites were used for those requir-
ing only simple modifications (Online Supplement 1).
Petrographic analysis of a Type E hammer for which
the origin and the nature of the rock remained uniden-
tified (Online Supplement 2). was carried out by
G. Fronteau (GEGENAA, Reims University), and
indicated that it was of alkali basalt (Galoyan et al.
2009:FIG.6h). This highly altered basalt contains plagi-
oclase phenocrysts and opaque mesostatic minerals
including chlorite and calcite crystals (the result of
recrystallization). This raw material is characterized
by extreme hardness and resistance when used for per-
cussion. Potential outcrops are located about 20 km
north-west of Duzdag
˘i.
Hammer manufacturing is mainly involved in
making facets on the cobble for hafting but, more
rarely, it involved the shaping of the general form
and the active surfaces. The simplest modifications
for hafting purposes involved the removal of one or
two bifacial flakes so as to create sub-symmetrical
notches towards the center of each side of the
hammers (FIG.5). As pointed out above, it is possible
that the notches visible on certain tools correspond
to the first stage of the shaping of grooves for
hafting. Their shaping is not very sophisticated, the
only real difficulty involved being the search for dis-
tinct symmetry. In some cases the grooves may have
been pre-shaped by means of a series of alternating
flake removals or by directly pecking the natural
Fig. 6. Type C and D hammers showing their morphology and locations of the simple (top) and complex grooves (bottom)
(Photographs C. Hamon).
Hamon Salt mining tools and techniques from Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.
Journal of Field Archaeology 2016 VOL. 41 NO. 4516
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

surface. Regular pecking of the groove was then
carried out in order to control its width and symmetry
(Online Supplement 3). This regularization was made
directly on the natural surface or aimed to rectify the
sharp ridges left by previous shaping removals. In
some cases a finishing stage involving grinding gave
a blunted aspect to the bases of certain grooves. This
is difficult to distinguish from blunting resulting
from friction with the hafting elements. Alkali basalt
hammers have a different type of procurement and
blank management than that used for the other
hammers. Although this rock type occurs in the allu-
vial layers at Duzdag
˘i and Nakhchivan çay, several
technological indications suggest direct exploitation
of primary outcrops of alkali basalt located at least
20 km to the north (after Azizbekov 1961). These
hard and resilient materials were subjected to more
sophisticated shaping and finishing than those used
for the other hammers. At Duzdag
˘i these shaping
sequences are attested by a large amount of manufac-
turing and maintenance waste (FIG.5B), if we exclude
the initial stages of shaping took place outside the
mining area. The following shaping stages are illus-
trated by 52 hammer roughouts made of alkali
basalt at the site (7.2% of the assemblage). The
initial shaping of the alkali basalt tools was carried
out by means of a series of extensive bifacial flake
removals, followed by pecking over all parts of the
tool (point, sides and edges); the aim was to maintain
the symmetry of the piece and also to erase previous
shaping traces (FIG.7A). This stage resulted in the
shaping of distinctive pointed ends by pecking fol-
lowed by polishing. When the end became too flat-
tened through use, it was completely re-shaped (FIG.
7C). This new shaping involved a reduction in tool
size (by up to two-thirds), distinct dissymmetry,
reshaping of all the surfaces except for the butt,
shaping of a new point and, in some cases, the creation
of a second groove next to the first one. The successive
use- and shaping cycles of the alkali basalt hammers
may, therefore, result in the length of the object
being reduced by up to two thirds of its original size
(FIG.7D). The intensive use of these tools depended
largely on the quality and scarcity of the selected raw
material. The maintenance by-products and use-
waste along with fractured hammers (325 complete
tools, i.e., 45% of the assemblage) confirm that inten-
sive stages of use and maintenance succeeded each
Fig. 7. Type E hammers: A) Incompletely shaped tool; B) Shape and profile of a standard hammer; C) Asymmetric points of three
re-sharpened hammers; D) Blunted points of two intensively used hammers; E) double-grooved rejuvenated hammer
(Photographs C. Hamon).
Hamon Salt mining tools and techniques from Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.
Journal of Field Archaeology 2016 VOL. 41 NO. 4517
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

other within the mining area. In addition to the
hammers in the strict sense, numerous flakes originat-
ing from the points and faces (66 specimens, i.e., 9% of
the assemblage) were recovered, sometimes in concen-
trations. These flakes may be linked to the first use of
the objects and they may reveal a particular type of
hafting (during use the object may have been prema-
turely fractured at the point of attachment with the
haft) or they may be shaping waste (the fracture may
have occurred during the pecking of the more fragile
point) (FIG.9).
In sum, the hammers recovered from the Duzdag
˘i
site provide evidence for two forms of blank manage-
ment. Tools made from locally available cobbles of
various rock types were used for opportunistic
shaping; the tools made of alkali basalt were subjected
to more sophisticated shaping involving greater
manufacturing skills. Moreover, the various types of
macrolithic tools attest to shaping and maintenance
operations that took place within the mining area, par-
ticularly in the case of the intensively-used alkali
basalt hammers.
Function of the hammers
Salt exploitation strategies are revealed within the
combination of several criteria: raw materials, tool
sizes and hafting types for the hammers. The variety
of modifications carried out to facilitate hafting may
not be attributed to temporal and cultural factors
given that the spatial distribution of the different
types of hammers does not show concentrations or
strict exclusions relative to the distribution of pottery
remains. Similarly, it is unlikely that the different
hammer types and forms of modification can be
Fig. 8. Type F massive hammers (Photographs C. Hamon).
Hamon Salt mining tools and techniques from Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.
Journal of Field Archaeology 2016 VOL. 41 NO. 4518
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

interpreted as hallmark products of particular work-
shops or specialized craftsmen due to the rather
limited effort invested in their shaping. We, therefore,
examined whether differences linked to the size or to
the modifications of the hammers corresponded to
functional constraints arising from different technical
operations.
The operating mode of the hammers completely
depends on the constraints of the work place: the pos-
ition of the miner and the range and direction of his
movements being entirely determined by the configur-
ation of the workspace. It was necessary, therefore, to
calibrate the lengths of the handles in order to ensure
efficiency. The orientation of the cutting edge of the
hammer (transversal or longitudinal) and the move-
ments required for direct percussion (tangential, perpen-
dicular) during the extraction process also determines
the operating mode. It appears quite unlikely that
hammers with very different weights (massive 5
−80 kg, intermediate 1.5−5 kg, small 300g–1.5 kg)
had identical operating modes and functions.
The degree of use and high frequency of re-sharpen-
ing was apparently not a limiting factor with regard to
the use of one group of small hammers. The second
type of small hammers reflects opportunistic use. In
both cases, the small hammers appear to represent
functional specialization within the sequence of salt
exploitation compared to the medium-sized or
massive hammers.
The hammers weighing several kilos could not have
been hafted, although a groove was created, as was the
case for the small hammers. The most massive
hammers may have been held in rope-slings making
it possible to direct these tools against the rock face
by a backwards and forwards motion. It may also be
possible that they were swung from tripods, as was
the case at other mines in the Caucasus
(Gambaschidze et al. 2001). In fact, the presence of
multiple peripheral or lateral grooves suggests a
double suspension system. Other massive hammers
with complex modifications exhibit slightly “non-sym-
metrical”grooves forming semi-circles at their pointed
ends (FIG.8B). Juxtaposed coarse impacts on the tools
reflect perpendicular direct percussion. The most
massive hammers were probably smashed against the
rock face in order to expose the salt layers and to dig
horizontal galleries or vertical shafts and pits. It
should be noted that some of the more massive stone
blocks have only slight traces of use; in other contexts
they have sometimes been interpreted as having cere-
monial functions. It is possible that the medium-
sized hammers, particularly those made of alkali
basalt, were also operated by being smashed against
the rock face with the help of ropes.
The grooves on the massive hammers clearly differ
from those on the small and medium-sized hammers
(Online Supplement 4). Eighty-two percent of the
massive hammers display simple lateral or peripheral
grooves and 4% exhibit complex grooves that include
a peripheral groove, which allowed the suspension of
the hammers. In contrast, the lateral grooves,
especially when they are part of a complex system,
are thought to indicate the presence of hafting. The
function of these complex modifications is to fix the
small hammers to their haft. This hypothesis is sup-
ported by various types of hafts that have been pre-
served or reconstructed for other mining contexts
and our own experiments (Online Supplements 5, 6
and 7). They were generally made from various types
of wood, using binding systems of hemp, leather or
bark ropes (Craddock et al. 2003;Garner 2013;
O’Brien 1994;Tarantini and Galiberti 2011).
Fig. 9 Flakes from points of alkali basalt hammer stones and fracture types of the point observed during the first use cycles
(Photographs C. Hamon).
Hamon Salt mining tools and techniques from Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.
Journal of Field Archaeology 2016 VOL. 41 NO. 4519
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

Amongst the medium-sized hammers, those manu-
factured from alkali basalt clearly have a particular
status compared to the other hammers, which
display more opportunistic manufacture and use.
The intensive use of tools manufactured from exogen-
ous and very resistant materials that, in addition, were
subject to particularly sophisticated shaping, almost
certainly involves a specific conception of the mining
work. The regular reshaping of the point made it poss-
ible to cut the rock in a more efficient and precise way,
the material ensuring significant shock-resistance and,
therefore, a longer lifetime for the tool.
Hammer Use-wear Analysis
The working of the rock (digging of pits, galleries, out-
crops, etc.), the carving out of salt blocks, their extrac-
tion and shaping do not necessarily require the same
tools, operations and skills. We have, therefore,
attempted to identify whether the degree of specializ-
ation could be indicative of a possible segmentation
of the exploitation operations and of complementary
operating modes related to salt.
Use-wear analysis was carried out on a selection of
hammers with the aim of accurately describing the
nature, the orientation and the extent of the use
traces visible on the active ends of the tool. The aim
was to determine whether the traces resulted from
the use of the hammers for a specific stage of salt
exploitation.
The use-wear traces, observed macro- and micro-
scopically on about thirty hammers, all indicate the
processing of a semi-hard salt-type mineral material:
this is particularly the case for blunting characterized
by surface deformation of the grains making up the
rock (Online Supplement 8).
However, the active ends of the 30 hammers studied
mainly bear traces resulting from direct percussion:
crushed areas (54%), splintered areas (5%) and scars
(40%). Crushed areas are characterized by impact
scars of various shapes and diameters (circular, oval-
shaped) and they mainly form on convex zones
having sharp angles; their depth directly depends on
the intensity intensities of the blows. Splintered areas
and scars form in zones with steep angles and com-
prise secondary splinters, associated with percussion
impacts, and large step scars (FIG.10).
The formation and the aspect of use-wear is largely
dependent on the raw materials of the tools. Hammers
manufactured from the metamorphic rocks and alkali
basalts, which are more prone to conchoidal fracture,
exhibit step scars and splintering. In contrast, granite
and quartzite stones generally display crushed areas,
the fineness of which directly depends on the accuracy
of the direct percussion.
The extent of use-wear depends on the intensity and
the duration of use of the hammers. The sizes of the
scars and splinters more directly depend on the inten-
sity of the blows during the extraction of the salt
blocks: large scars result from heavy direct percussion;
small splinters are generated by more localized direct
percussion.
Lastly, the morphology of the use-wear traces also
mirrors the development of the morphology of the
active surfaces. Splintered areas are characteristic of
pointed hammers, whereas coarsely crushed areas are
more likely to occur on rounded ends. Partially splin-
tered or split ends tend to be continuously reshaped
by new removals in contrast with rounded ends,
which tend to preserve the outline of the original
surface with use. The morphology of the traces, there-
fore, strongly depends on the morphology, the ergo-
nomics and the penetration of the active surfaces
into the material (Online Supplement 9). The very
first stages of use of alkali basalt hammers only
Fig. 10. Use-wear traces: A) bruising and percussion marks on hammers stone; B) secondary bruising and splintering on
hammer stone; C) large flake removals on hammer stone; D) step scars on hammer stone (Photographs C. Hamon).
Hamon Salt mining tools and techniques from Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.
Journal of Field Archaeology 2016 VOL. 41 NO. 4520
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

slightly affect the points; they appear to be only
slightly crushed, even micro-splintered. In the course
of use, the bruising of the end and the formation of
a large number of step scars contribute to a widening
and flattening of the active surface (FIG.11).
Use-wear traces are also a particularly effective
means of studying the use cycles and repair of the
hammers at the mining site. They also highlight their
degree of specialization and their position within the
operational sequence of salt extraction (FIG.12). The
ends of unmodified or notched hammers (Types A
and B) most frequently have scars (70%). These tools
thus had a highly material-consuming and short inten-
sity/duration of use and reflect violent and poorly-
controlled movements linked with salt exploitation.
It is difficult to associate these tools with a particular
function or a particular stage within the operational
sequence of salt exploitation.
Hammers with simple and complex grooves (Types
C and D) exhibit equal proportions of scarred and
crushed areas. Solely on the basis of use-wear traces,
they probably had a comparatively “polyvalent”
(multi-purpose) function during salt extraction, being
used for the carving out and the extraction of the
blocks.
Many hammers manufactured from alkali basalt
(Type E) have crushed areas (60%). The need to main-
tain a pointed active surface can only be explained by
the requirement that the tool penetrate into the
material being worked. Although the Type E traces
clearly indicate different degrees of use, the prevalence
of splintered areas and scars confirms their use for
operations of greater precision, for example cutting
salt blocks. The massive hammers (Type F) more
rarely exhibit scars because the angulation of their
large active surfaces do not favor fracturing. Their
role likely corresponds to the first stages of the
digging of shafts or galleries and not necessarily to
the extraction of the salt.
Thus, at the Duzdag
˘i mine, salt exploitation is
thought to have involved a combination of opportunis-
tically used hammers (Types A and B), polyvalent
hammers (Types C and D) and hammers with a
more specialized use within a particular stage of the
Fig. 11. Schematic evolution of the morphologyof the active surfaces of the alkali basalt hammer stones in relationship with the
visible use marks (Photographs C. Hamon).
Hamon Salt mining tools and techniques from Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.
Journal of Field Archaeology 2016 VOL. 41 NO. 4521
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

operational sequence, particularly for the working of
the rock face (Type F) and the cutting of the salt
blocks (Type E). The lack of correlation between the
main use-wear categories and the types of tool modifi-
cations supports the idea that there was a compara-
tively low level of functional specialization among
the hammers coupled with low segmentation of the
operations involved in salt extraction.
Current field research is now exploring this oper-
ational sequence in detail, but two types of hammers
are thought to have had a more specialized function.
Massive hammers, would have been used for the
working of the outcrop faces or to begin the digging
of galleries/shafts. Intermediate hammers (particu-
larly Type E) may have been used for the contour
cutting and the extraction of the salt blocks following
two operating modes (hafted or suspended).
Hammerstones and intermediate pieces
Two additional tool types used for direct percussion
were found at the Duzdag
˘i salt mine (TABLE 2). Fifty
discoid or oval-shaped percussion tools, frequently
made of cobbles of siliceous rocks (quartzite, sand-
stone) or of metamorphic rocks, were documented.
They range from 8 cm to 20 cm in length. Their per-
ipheral edges are finely splintered and crushed;
however, they were apparently not intensively used
(FIG.12). In some cases, fragments of the stone
blocks also displayed peck marks at the ends.
Experimental work has shown that metamorphic
rocks, alkali basalts and quartzite stones were most
suitable for the manufacturing of the hammers. In
addition, the shaping of alkali basalt hammers
occasionally generated notches on the edge of the per-
cussion tool, matching the observation made on one of
the pieces collected during the archaeological survey.
Four oblong-shaped intermediate pieces were man-
ufactured from metamorphic rocks; they have quite
different sizes and weights (ranging from 10 to 20 cm
in length and 400 g to 2.1 kg in weight). They exhibit
two relatively strongly pecked and/or splintered
ends. In combination with a hard hammer, this kind
of tool allows great precision in the blows delivered;
it may possibly have been used as a wedge or as a
tool for carving out the blocks of rock salt during
the extraction process.
Given the presence of percussion marks on the hard
stone materials from which they were manufactured,
these two tool types were probably used for the
shaping (knapping, pecking) and/or the maintenance
of the hammers used for salt extraction. However,
one cannot rule out the possibility that they played a
role during the processing of the rock salt (for
example the shaping of the salt blocks).
Grinding tools: grinding slabs and pestles
Thirty pestles were also recovered. These were made of
sandstone or siliceous rocks, or from alkali basalt in
the case of the most sophisticated examples. They are
conical to cylindrical in shape and are manufactured
from cobbles or fragments of collected blocks. Their
sizes are rather standardized: on average they
measure 10 cm in length and 5 cm in diameter. Use-
wear analysis of four pestles, revealed fine percussion
marks at their ends, occasionally masked by periph-
eral, surface smoothing resulting from the grinding
of soft mineral materials (FIG.13). This observation
would exclude their use for the shaping of salt blocks
and instead indicates that they were used for grinding
operations in order to obtain salt powder. Such an
operation may have been carried out to produce
crystal salt for domestic consumption or for wider dis-
tribution. Other tools, manufactured from sandstone
and vesicular basalt, were used for grinding by using
a back-and-forth movement or were re-used as crush-
ing implements. The saddle querns (3 pieces), gener-
ally having an oval shape, are to be distinguished
from the riders (4 pieces). A small handstone, the func-
tion of which remains undetermined, was also discov-
ered. Despite the apparent absence of mortars or
grindstones associated with pestles, the presence of a
quern re-used as a grinding slab for the crushing of
hard mineral materials, should be mentioned (FIG.
14). Its active surface exhibits cereal gloss linked
with its primary use and three percussion pits (one of
these on its rear side) that served for the crushing of
hard mineral matter. The diameter of the querns
matches that of the pestles with which they were
Fig. 12. Examples of pecking stones found at Duzdag
˘i
(Photographs C. Hamon).
Hamon Salt mining tools and techniques from Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.
Journal of Field Archaeology 2016 VOL. 41 NO. 4522
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

probably associated. One grinding stone presents two
successive surface states. Each end exhibits an area
presenting an abraded plateau and a strongly granular
aspect resulting from the grinding of semi-hard
mineral matter, which was relatively powdery and
very abrasive. At the center of the object, unpolished
but slightly and regularly crushed rough areas coexist
with large percussion pits the maximum length of
which ranges from 1.5 to 1 cm. In most of the fissures,
salt incrustations are visible, sometimes reaching a sig-
nificant depth. It is likely that this grinding stone was
used as a grinding/crushing slab for semi-hard salt-
like mineral matter. The use of this tool with a pestle
remains a possibility. A second grinding stone, primar-
ily employed for cereals, was used as a grinding/crush-
ing slab for semi-hard mineral matter as evidenced by
the irregular, crushed and picked fissures at its center.
Despite the number of grinding tools that were
identified, they represent a small proportion of the
overall number of tools recovered at the site and indi-
cate that salt grinding was not intensively practiced.
Nonetheless, the presence of these grinding tools is
evidence that extraction was not the sole activity
undertaken at the mining site and that some salt
processing also occurred. This fact raises a number
of questions. It implies that salt grinding stages were
carried out in distinct, dedicated areas associated
with these extraction areas. It remains to be deter-
mined if these were relatively temporary dwelling
areas or, more probably, salt processing workshops
directly associated with the extraction activity itself.
This type of association of extraction and processing
tools conforms to discoveries made in other salt or
mineral exploitation contexts (Garner 2013;Weller
et al. 2007).
Discussion
The macrolithic tools from Duzdag
˘i thus constitute
genuine activity markers in that they enable us to
identify the nature of the technical operations carried
out as well as the degree of specialization and segmen-
tation of the salt extraction and processing activities.
In addition to the preponderance of extraction tools,
corresponding to the different stages of the operational
sequence (working of the rock faces, digging, carving
out, extraction, shaping), other tools were used for
the maintenance of the mining tools and for the
direct processing of salt (grinding).
Opportunistic versus specialized toolkit
Two types of hammers exhibit low investment of effort
in terms of the shaping and procurement of a specific
module or raw material (Types A and B). They exhibit
use marks that are randomly distributed. These
characteristics suggest that these hammers were
almost certainly employed as auxiliary tools at a
given exploitation stage. It is also possible that they
were used in operational sequences in which stone
tools played a minor role, either because the pro-
cedures did not require extensive use of these
hammers or because other tools were manufactured
from hard animal-derived materials, and wood or
metal were preferred. These types of hammers, gener-
ally manufactured from cobbles were consumed in sig-
nificant quantities regardless of whether the chosen
raw material was hard (Ambert et al. 1998) or rela-
tively soft (Garner 2013;Stöllner et al. 2010). They
suggest that the mining activity, did not require a
specialized toolkit and that the skills associated with
extraction focused on other stages of the operational
sequence. Some authors consider that specialized
tools are not necessary in order to extract rock salt
and that a pointed end is the only really indispensable
criterion (Figuls et al. 2007). Others suggest that a
single tool may have been used for several different
operations (Timberlake and Craddock 2013).
Type C and D hammers have simple or complex
modifications. They generally exhibit little effort in
shaping and degree of use. Type C hammers with
simple modifications resemble mining tools generally
Fig. 13. Examples of pestles found at Duzdag
˘i (Photographs
C. Hamon).
Hamon Salt mining tools and techniques from Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.
Journal of Field Archaeology 2016 VOL. 41 NO. 4523
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

encountered in other types of mining sites in a wide
range of temporal and geographical contexts
(Domergue 1990: 116–117; De Pascale 2004).
Comparisons can be made between the Type D
hammers with complex modifications and specimens
used during the emergence of copper metallurgy in
the British Isles (Pickin 1990). Their ubiquity suggests
that the types of exploitation (outcrops, galleries,
shafts, pits) had more of an effect on the choice of a
tool type than the nature of the exploited rock. Type
D tool types may, therefore, be explained by the
absence of a need for specialized tools for the extrac-
tion of rock salt and, in certain cases, by the adap-
tation of ore extraction techniques and tools for use
in salt extraction. Some authors have even interpreted
these tools as being polyvalent and used as picks,
pestles, hammers or as sharpening stones (Castel and
Soukiassian 1989: 119–120). On the other hand, the
Type E pointed hammers manufactured from alkali
basalt are almost unknown from other mining
contexts. The use of a similar rock, also of exogenous
origin, for the manufacture of grooved mining tools is
known from the Early Bronze Age in Iran (Holzer and
Momenzadeh 1971). In addition, hammers entirely
shaped by pecking are attested at other protohistoric
sites (Tarantini et al. 2011: 258). At the prehistoric
salt mine of La Cardona (Spain), the use of well-
shaped axes and chisels for the extraction of salt was
proposed on the basis of a technological and macro-
wear study of 300 hundred tools from private collec-
tions (Weller et al. 2007). Type B and C macro-traces
on the edges of these tools clearly fit with the splin-
tered areas and the scars observed on the Duzdag
˘i
hammers of type B, which would indicate a similar
gesture to that used in throwing percussion.
However, in contrast to the Type E hammers from
Duzdag
˘i, all of the lithic tools found at the prehistoric
salt mine of La Cardona were interpreted as extraction
tools, and none of them would have had a specialized
function (Weller et al. 2007).
Fig. 14. Example of a grinding slab found at Duzdag
˘i (Photographs C. Hamon).
Hamon Salt mining tools and techniques from Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.
Journal of Field Archaeology 2016 VOL. 41 NO. 4524
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

The Type E hammers at the Duzdag
˘i site show the
highest investment of effort with regard to shaping
and they also present the most complete range of use
marks. They range from points that are scarcely modi-
fied by violent knapping of the distal ridges to
hammers with a completely blunted and rounded
point that occurs adjacent to the modified surface
due to the significant reduction of their length. The
substantially worn aspect of these tools suggests sig-
nificant duration of use. All of these characteristics
suggest that these hammers had a very particular
status. Their use is probably related to the practice of
specialized skills, with part of the mining group man-
ufacturing and using its own salt-exploitation tools. In
fact the manufacturing of such a toolkit from alkali
basalt suggests that mining work was an activity
with special status within Bronze Age societies.
Moreover, the design of such a “point”may evoke,
copy or prefigure the use of metal picks for the extrac-
tion of salt during the Iron Age, as, for example, at
Chehrabad Douzlakh (Iran) (Aali et al. 2012).
Tool types as indicators of operations
The identification of different functional categories of
tools reflects at least three major operations: cutting
the rock face and digging out the rock; carving out
the blocks; and grinding the salt. The working and
digging of the rock faces was probably carried out
using the largest hammers in a back and forth
motion. Within the mining contexts, such an operating
mode was proposed for the use of grooved hammers in
the Chalcolithic mines of Rudna Glava (Jovanovic
1989: 17). The digging of pit-type features, vertical
shafts or other types of galleries is thought to have
been carried out using less specialized tools, i.e., poly-
valent hammers of varying sizes.
For carving out the salt blocks there was a need for
tools that were easy to handle and having pointed ends
enabled optimal penetration into the rock. The
hammers manufactured from alkali basalt would
have been perfectly adapted for this type of operation.
Some hammers with blunted or rounded tips may have
been employed for the regularization of the salt blocks
into standard sizes.
Clearly, a portion of rock salt production was then
transformed into crystal salt by crushing and grinding
for local consumption by the miners or livestock, and
less probably for exchange. The use of salt grinding
stones is attested in other contexts (Erdogu and
Ozbasaran 2007). More generally, tools such as
pestles, hammers, anvils and pitted stone tools were
used for ore processing at mining sites dated to the
4th and 3rd millennia B.C.(Conrad and Rothenberg
1980: 170; Craddock 2008;Domergue 1990;Garner
2013;Hamon et al. 2009;Ottaway 2003;Stöllner
et al. 2011). At the Duzdag
˘i site the absence of a
strict temporal attribution for each tool type, forces
us to assume that the tools were possibly involved in
several operational schemes which may have varied
not only from one period to another but also from
one exploitation area at the site to another.
Organization of mining activities
Based on the concentrations of macrolithic tools, three
complementary areas of activity at the site may be dis-
tinguished: the actual areas of extraction, the areas
used for the manufacturing and the maintenance of
the extraction tools, and, finally, the areas reserved
for the processing of rock salt products.
The presence of particularly massive hammers may
be indicative of extraction areas organized around fea-
tures such as outcrops or galleries. The concentrations
of hammers of all types probably indicate exploitation
zones. The operations related to the carving out of
blocks from the rock face or the shaping of salt
blocks may have been carried out in several oper-
ational extraction sequences.
The co-existence of broken hammers (points,
flakes), of hammer roughouts, and of percussion
tools leads to the suggestion that the manufacturing
and repair of the hammers took place at the mining
site itself. It also suggests the presence of organized
workshops in addition to the operations related to
salt extraction. The emerging hypothesis is that these
workshops were in part autonomous and ensured the
preparation of the tools prior to their use and their
repair after use. A large number of tools (particularly
tools of Type E) were abandoned in situ; it is likely that
the production of the workshops not only satisfied the
immediate needs of the miners with regard to repair or
manufacturing of new tools, but also anticipated, to a
limited extent, future needs with regard to the toolkit.
Lastly, the salt processing zones as evidenced by the
presence of pestles and grinding slabs were apparently
more or less directly connected with the tool mainten-
ance and manufacturing workshops. Two types of
workshops, therefore, may have coexisted in several
suitable areas. These various spaces, which were
located close to each other, frequently occur in other
salt exploitation sectors at the Duzdag
˘i site despite
the different topographical settings of the sectors.
This spatial segmentation of the activities and rela-
tive specialization of the tools with regard to several
extraction and processing operations indicates that
the main period of exploitation, which has been
assigned to the Kura-Araxes Culture on the basis of
pottery remains, was not solely determined by simple
ad hoc needs. The manipulation of the most massive
hammers probably required a dedicated work force
in a situation which differed greatly from the opportu-
nistic exploitation of the most directly accessible out-
crops. Depending on the duration and the areas
Hamon Salt mining tools and techniques from Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.
Journal of Field Archaeology 2016 VOL. 41 NO. 4525
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

available for each site, one must assume that there was
relatively significant mobilization of groups composed
of several workers.
The segmentation of the activities suggests that
there were separate groups of people with specific
skills for the manufacturing and maintenance of the
toolkit, for extraction, and finally for salt processing.
These skills may have been relevant during the entire
operational sequence, from the choice of the materials
and the manufacturing techniques up to the processing
and the transport of the extracted salt. This kind of
organization is similar to that found earlier at the
Neolithic flint mines of Gargano (Tarantini et al.
2011: 262).
Duration and timing of mining activities
The presence of loci still visible at the surface of the
exploitation area may testify to rather isolated and
short-term activities, which were mutually exclusive.
Moreover, their presence confirms that less steep
areas were exposed to less extensive taphonomic pro-
cesses and that the movements created by the accumu-
lation of the vestigial remains were restricted to local
areas. This configuration enables us to take advantage
of the archaeological information that confirms the
existence of different areas of exploitation. Several
types of spatial organization seem to have co-existed
at the exploitation site: successive frequentations at
the same place, occupation of a new platform or a
new area of the platform each time the exploitation
was resumed, or a combination of these two cases.
As is the case for most mining sites from this period,
we are probably dealing with seasonal activities that
were conducted during the least severe periods in
terms of water flow and terrain stability. The existence
of specialized activities particularly during the Bronze
Age does not mean that they were full-time, in fact, the
existence of seasonal constraints related to this type of
activity suggests that these may have been “part-time”
exploitations (Tarantini et al. 2011).
This succession of exploitation episodes of rather
short duration would also explain the abandonment
of large quantities of hammers on the site, as new
tools were probably shaped for each new episode of
extraction. It also suggests that the “seasonal”aspect
of the exploitation was clearly not inconsistent with
the evident capacity of the mining community to
acquire the raw material, produce new tools and trans-
port them on the site.
The earliest hammers with grooves emerged during
the 4th millennium B.C. within an area encompassing
southeastern Europe and the Middle East
(Domergue 1990: 117) and would have spread
towards western Europe during the 3rd millennium
B.C. According to some authors, the significant use
of stone tools during the Bronze Age would have
been replaced by the use of metal tools at the end of
the Bronze Age and the beginning of the Iron Age in
different regions (Garner 2013;Rothenberg and
Freijeiro 1980). But we cannot rule out the re-use of
tools from one occupation to another and this may
sometimes have occurred from one period to the other.
Conclusions
The development of an innovative method of georefer-
enced mapping of the technical and functional data
related to the macrolithic tools at the exceptional site
of Duzdag
˘i has allowed us to propose a dynamic
vision of the modalities of salt exploitation. The
archaeological remains made it possible to date the
beginning of salt extraction at the site to the Late
Chalcolithic period and to attribute the main period
of hammer use to the Kura-Araxes culture (Marro
et al. 2010) between the 5th and 3rd millennia B.C.
Salt mining was part of the emergence of the mining
phenomenon in general in the Caucasus, and high-
lights from a new perspective the intensification of
the exploitation of mineral resources during the tran-
sition between the Late Chalcolithic and Early
Bronze Age.
This integrated study made it possible to highlight
the mining techniques used at Duzdag
˘i as well as the
particular organization of the activities within the
mining area: workshops related to the manufacturing
and repair of the tools, extraction areas and salt pro-
cessing areas. It appears that the mining tools which
are emblematic of the Kura-Araxes culture reflect
the specific status of the salt miners. The use of special-
ized tools for specific operations requires, in itself, a
certain degree of know-how, and also a social back-
ground that permits its transmission over generations.
The use of imported materials—displayed in their
shaping and their long duration of use—implies orga-
nized exploitation which would have mobilized part of
the community. In addition, during episodic seasonal
exploitation, the miners were probably not always in
a position to procure food, and were supported by
their community. In fact, the Kura-Araxes commu-
nities had the technical and economic capacity to
exploit, transform and transport significant quantities
of salt over a relatively long time span. All of this sup-
ports the hypothesis that mining techniques and
mining populations played a role in the development
of the new economic organization within the Kura-
Araxes culture (Stöllner et al. 2014).
It should, therefore, be considered that salt exploita-
tion at Duzdag
˘i, located on a major route linking
the southern Caucasus, Anatolia and Mesopotamia,
played a significant economic role, fully engaged in pro-
found social and economic upheavals (Lyonnet et al.
2012;Marro et al.2011;Sagona 2011). The Duzdag
˘i
site very certainly illustrates a wider phenomenon
Hamon Salt mining tools and techniques from Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.
Journal of Field Archaeology 2016 VOL. 41 NO. 4526
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

related to the exploitation and distribution of mineral
resources during the protohistoric period throughout
the entire region, which probably not only involved
metal products but also unpreserved mineral or
organic products of major economic and symbolic
value.
Acknowledgments
This study was carried out between 2008 and 2011 as
part of the Araxe Research Programme funded by the
French Ministry of Foreign Affairs and directed by
C. Marro, in collaboration with Véli Bashvaliev of
the Azerbaijan National Academy of Sciences at
Nakhchivan. We would like to thank the survey
team: S. Sanz and N. Aliyev who conducted the
survey campaigns, as well as E. Aliyev, R. Berthon,
E. Bouet, J. Cavero, P. Dessaint, K. Dupinay and
N. Gailhard. We would also like to thank the current
engineer of the mine as well as the director of the
Nakhchivan Museum and his team for their warm
welcome. Many thanks to G. Fronteau
(GEGENAA, Reims University) for the petrographic
analysis. We are grateful to the CNES and the ISIS
programme, which enabled us to purchase the Spot 5
satellite photo. The English version of this text was
translated from French by K. Mazurié de Keroualin
and R. Cronin-Allanic.
Caroline Hamon (Ph.D. 2004, Pantheon Sorbonne
University, Paris) is a Researcher at CNRS. She is par-
ticularly interested in food practices, craft activities, and
mineral resource exploitation during the Neolithic and
Metal Ages in different contexts, including north-
western Europe, western Mediterranean, and the
Caucasus.
Supplemental data
Supplemental data for this article can be accessed here.
http://dx.doi.org/10.1080/00934690.2016.1201615
ORCiD
Caroline Hamon http://orcid.org/0000-0003-3314-
9525
References
Aali, A., T. Stöllner, A. Abar, and F. Rühli. 2012.“The salt men of
Iran: the salt mine of Douzlakh, Chehrabad,”Archäologisches
Korrespondenzblatt 42(1): 61–81.
Abibullayev, A. A. 1982.Eneolit i Bronza na Territorii
Nakhichvanskoi ASSR. Baku: Elm.
Adams, J., S. Delgado, L. Dubreuil, C. Hamon, H. Plisson, and R.
Risch. 2009.“Functional analysis of macro-lithic artefacts,”in
F. Sternke, L. Eigeland and L. Costa, eds., Non-flint Raw
Material Use in Prehistory Old Prejudices and New Direction,
15th UISPP congress, Lisbon-September 2006, BAR
International Series 1939. Oxford: British Archaeological
Reports, 43–66.
Aliyev, V. G. 1991.Kul’tura Epokhi Srednei Bronzy Azerbaidzhana.
Baku: Elm.
Ambert, P., V. Figueroa-Larre, J.-L. Guendon, V. Klemm, M.
Laroche, S. Rovira, and C. Strahm. 2009.“The Copper
Mines of Cabrières (Hérault) in Southern France and the
Chalcolithic Metallurgy,”in T. Keinlin and B. Roberts, eds.,
Studies in Honour of Barbara Ottaway, Universitats forschungen
zur Prähistorischen Archäologie, Vol. 169, Metal and Societies.
Bonn: Dr. Rudolf Habelt, 285–295.
Ambert, P., L. Carozza, and B. Léchelon. 1998.“De la mine au
métal au sud du Massif central au Chalcolithique (régions de
Cabarières, Fayet et Villefranche-du-Rouergue),”L’Atelier du
bronzier en Europe du XXe au VIIIe siècle avant notre ère,
Tome 2 (session de Dijon), Du minerai au métal, du métal à
l’objet, Actes du Colloque international Bronze’96′′ Neuchâtel
et Dijon,1996, Paris: C.T.H.S., 59–70.
Azizbekov, C. A. 1961.Geologia Nakhichvanskoi ASSR.
Gosydartstvenoe naytchno tekhnitcheckoe izdatelstvo literatyre
po geologiniox ranenedr. Moskva.
Bakhshaliyev, V., and C. Marro. 2009.The Archaeology of
Nakhchivan: Ten Years of New Discoveries. Istanbul: Ege Press.
Cassen, S., and O. Weller. 2013.“Idées et faits relatifs à la pro-
duction des sels marins et terrestres en Europe, du VIe au IIIe
millénaire,”in J. Soares, ed., Prehistory of Wetlands.
Landscapes of salt, coll. Setúbal Arqueológica, Vol. 14.
Setúbal: Setúbal Museum of Archaeology and Ethnography
of the District of Setúbal (MAEDS), 255–304.
Castel, G., and G. Soukiassian, eds. 1989.Gebel-el-Zeit I Les mines
de Galène (Egypte, IIe millénaire av. J.-C.), Fouilles de l’Institut
français d’archéologie orientale du Caire,Vo l. XXXV. Le Caire:
IFAO, 115–132.
Conrad, H. G., and B. Rothenberg, eds. 1980.Antikes Kupfer im
Timna-Tal. 4000 Jahre Bergbau und Verhüttung in der Arabah
(Israel).Der Anschnitt 1. Bochum: Begbau Museum.
Craddock, B. R., C. R. Cartwright, P. T. Craddock, and W. B. Wray.
2003.“Hafted Stone Mining Hammer from Chuquicamata,
Chile,”in P. Craddock and J. Lang, eds., Mining and Metal
Production Through the Ages. London: British Museum Press,
53–68.
Craddock, P., and J. Lang. 2003. Mining and Metal Production
Through the Ages. London: British Museum Press.
Craddock, P. T. 2008.“Mining and metallurgy,”in J. P. Oleson, ed.,
The Oxford Handbook of Engineering and Technology in the
Classical World. Oxford: Oxford University Press, 93–120.
Denton, D. A. 1984.The Hunger for Salt: An Anthropological
Physiological and Medical Analysis. New York: Springer.
De Pascale, A. 2004.“Hammers From Early Copper Mines: Sintesi
dei ritrovamenti nell’Europa e nel Mediterraneo orientale e
prime considerazioni sui mazzuoli di Monte Loreto (IV millen-
nio B.C. Liguria),”Rivista di Studi Liguri 69: 5–42.
Díaz-del-Río, P., S. Consuegra, N. Castañeda, M. Capote, C.
Criado, M. A. Bustillo, and J. L. Pérez-Jiménez. 2006.“The
Earliest Flint Mine in Iberia,”Antiquity, 80(307). http:// www
.antiquity.ac.uk/projgall/diazdelrio307/
Di Lernia, S., G. Fiorentino, A. Galiberti, and R. Basili. 1995.“The
Early Neolithic Mine of Defensola A (I I8): Flint Exploitation
in the Gargano Area,”International Flint Symposium No. 7,
Poland (1995). Warszawa: Polish Academy of Sciences,
Institute of Archaeology and Ethnology 33: 119–132.
Domergue, C. 1990.Les mines de la péninsule Ibérique dans
l’Antiquité romaine. Rome: Collection de l’Ecole Française de
Rome, 127.
Erdogu, B., and M. Ozbasaran. 2007.“White, Pure, Immaculate and
Incorruptible: Salt in Prehistoric Central Anatolia,”in D.
Morah, G. Dumitroaia, O. Weller and J. Chapman, eds.,
L’exploitation du sel à travers le temps. Piatra Neamt: Edition
Centre International de Recherche de la culture Cucuteni,
Institutt d’archéologie Iasi, 135–145.
Figuls, A., J. Bonache, J. Aranda, J. Sune, M. Vendrell, J. Gonzalez,
A. Funetes, J. Mata-Perello, and J. Sanz. 2007.“Neolithic
Exploitation of Halite at the “Vall Salina”of Cardona
(Catalonia, Spain),”in A. Figuls and O. Weller, eds., Trobada
International d’Arqueologia envers l‘explotació de la sal a la
Prehistória i Protohistória,Acts of the 1st International
Archaeology Meeting about Prehistoric and Protohistoric Salt
Exploitation, Cardona, 6–8Dec.2003. Cardona: IREC, 119–218.
Garner, J. 2013.Das Zinn der Bronzezeit in Mittelasien II: Die
montanarchäologischen Forschungen an den Zinnlagerstätten.
Darmstadt: Zabern Philipp Von Gmbh.
Goldenberg, G., A. Maass, G. Steffens, and H. Steuer. 2003.
“Hematite Mining During the Linear Ceramics Culture in the
Area of the Black Forest, South-western Germany,”in T.
Hamon Salt mining tools and techniques from Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.
Journal of Field Archaeology 2016 VOL. 41 NO. 4527
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

Stöllner, G. Körlin, G. Steffens and J. Cierny, eds., Man and
Mining, Studies in Honour of Gerd Weisgerber on Occasion of
his 65th Birthday. Boschum: Deutsches Bergbau-Museum,
179–186.
Hamon, C. 2006.Broyage et abrasion au Néolithique ancien.
Caractérisation technique et fonctionnelle des outillages en grès
du Bassin parisien, BAR International Reports S1551. Oxford:
British Archaeological Reports.
Hamon, C. 2008.“Functional Analysis of Stone Grinding and
Polishing Tools from the Earliest Neolithic of North-western
Europe,”Journal of Archaeological Science 35: 1502–1520.
Hamon, C., P. Ambert, M. Laroche, J. L. Guendon, S. Rovira, and
L. Bouquet. 2009.“Les outils à cupules, marqueurs de la
métallurgie du district de Cabrières-Péret (Hérault) au
Chalcolithique,”Gallia Préhistoire 51: 179–212.
Holzer, H., and M. Momenzadeh. 1971.“Ancient Copper Mines in
the Veshnove Area, Kuhestan-E-Qom, West-central Iran with a
Contribution of G. Gropp,”Archaeologia Austriaca XLIX:
1–22.
Galoyan, G., Y. Rolland, M. Sosson, M. Corsini, S. Billo, C. Verati,
and R. Melkonyan. 2009.“Geology, Geochemistry and 40Ar/
39Ar Dating of Sevan Ophiolites (Lesser Caucasus, Armenia):
Evidence for Jurassic Back-arc Opening and Hot Spot Event
Between the South Armenian Block and Eurasia,”Journal of
Asian Earth Sciences 34: 135–153.
Gambaschidze, I., A. Hauptmann, R. Slotta, and Ü. Yalçın (Hrsg.).
2001.Schätze aus dem Land des goldenen, Vlies. Katalog zur
Ausstellung. Bochum: Deutsches Bergbau Museum.
Jovanovic, B. 1989.“Les premières mines de cuivre des Balkans et le
rôle qu’elles ont joué dans les développements de la métallurgie
en Europe du sud-est,”in C. Domergu, ed., Minería y metalur-
gia en las antiguas civilizaciones mediterraneas y europeas:
Coloquio internacional asociado, Madrid, 24–28 Octobre 1985
Instituto de Conservación y Restauración de Bienes Culturales.
Madrid: Inst. Cons. Rest. De la Bienes Culturales, 13–18.
Larocca, F. 2012.Grotta della Monaca (Calabria, Italia meridio-
nale). Una miniera neolitica per l’estrazione dell’ocra, Actes du
Congres Internacional Xarxes al Neolitic. Circulacio i intercanvi
de matèries productes i idees a la Mediterrània occidental (VII–
III millenni ac), Gavà, Bellaterra 4/2/2011,Rubricatum, Vol.5.
Gava: Revista de Museuo de Gava.
Lyonnet, B., F. Guliev, B. Helwing, T. Aliyev, S. Hansen, and G.
Mirtskhulava, eds. 2012.“Ancient Kura 2010–2011: The first
two seasons of joint field work in the southern Caucasus,”
Archäologische Mitteilungen aus Iran und Turan 44: 1–190.
Marro, C., V. Bakhshaliyev, and S. Ashurov. 2011.“Excavations at
Ovçular Tepesi (Nakhchivan, Azerbaijan). Second Preliminary
Report: the 2009–2010 seasons,”Anatolia Antiqua 19: 53–100.
Marro, C., V. Bakhshaliyev, and R. Berthon, 2014.“On the Genesis
of the Kuro-Araxes Phenomenon: New Evidence from
Nakhchivan (Azerbaijan),”Paleorient 40(2): 131–154.
Marro, C., V. Bakhshaliyev, and S. Sanz, 2010.“Archaeological
Investigations on the Salt Mine of Duzdag
˘ı(Nakhchivan,
Azerbaijan),”TUBA-Ar 13: 229–244.
O’Brien, W. 1994.Mount Gabriel, Bronze Age Mining in Ireland.
Galway: Galway University Press.
Ottaway, B. S. 2003.“Experimental archeometallurgy,”in T. Stöllner,
G. Körlin, G. Steffens and J. Cierny, eds., Man and mining,
Studies in Honour of Gerd Weisgerber on Occasion of his 65th
Birthday, Deutsches Bergbau-Museum Bochum, Der Anschnitt,
Vol. 16. Bochum: Deutsches Bergbau Museum, 341–348.
Palumbi, G., and C. Chataigner. 2014.“The Kuro-Araxes Culture
from the Caucasus to Iran, Anatolia and the Levant: Between
Unity and Diversity. A Synthesis,”Paleorient 40(2): 131–154.
Pétrequin, P., O. Weller, E. Gauthier, and A. Dufraisse. 2001.“Salt
Springs Exploitations Without Pottery During Prehistory:
From New Guinea to the French Jura,”in P. Pétrequin and
S. Beyries, eds., Ethnoarchaeology and its transfert. Actes du
5e Meeting of The European Archaeologist Association,
Bournemouth, 1999.BAR International Series 983. Oxford:
British Archaeological Reports, 37–65.
Pickin, J., 1990.“Stone Tools and Early Metal Mining in England
and Wales,”in P. Crewe and S. Crew, eds., Early Mining in
the British Isles: The Proceedings of the Early Mining
Workshop at Plas Tan y Bwlch, Snowdonia National Park
Study Centre (17–19 Nov. 1989), Plas Tan y Bwlch
Occasional Paper 1. Porthsmadog: Snodonia Press, 39–42.
Rothenberg, B., and B. Freijeiro. 1980.“Ancient copper mining and
smelting at Chinflon (Huelva, SW Spain),”in P. T. Craddock,
ed., Scientific Studies in Early Mining and Extractive
Metallurgy/19th International Symposium on Archaeometry
and Archaeological Prospection, Held in London in March
1979, London : British Museum, 41–62.
Sagona, A. 2011.“Anatolia and the Transcaucasus Themes and
Variations ca. 6400–1500 B.C.E.,”in R. Steadman and G.
McMahon, eds., Oxford Handbook of Ancient Anatolia
10,000–323 B.C.E., Oxford: Oxford University Press, 683–703.
Sagona, A. 2014.“Rethinking the Kuro-Araxes Genesis,”Paleorient
40(2): 131–154.
Stöllner, T., G. Körlin, G. Steffens, and J. Cierny. 2003.Man and
mining, Studies in honour of GerdWeisberger on Occasion of
his 65th Birthday,Deutsches Bergbau-Museum Bochum 2003,
Der AnschnittBeiheft, 16. Bochum: Deutsches Bergbau
Museum.
Stöllner, T., I. Gambaschidze, A. Hauptmann, G. Mindiašvili, G.
Gogoc
ˇuri, and G. Steffens. 2010.“Goldbergbau in
Südostgeorgien. Neue Forschungen zum frühbronzezeitlichen
Bergbau in Georgien,”in S. Hansen, A. Hauptmann, I.
Motzenbäcker and E. Pernicka, eds., Von Maikop bis Trialeti.
Akten des Symposiums Berlin 1–3 Juni 2006. Kolloquien zur
Vor- und Frühgeschichte 13, Bonn, 103–138.
Stöllner, T., and I. Gambashidze. 2011.“Gold in Georgia II: The
Oldest Gold Mine in the World,”in Ü. Yalçın, ed., Anatolian
Metal V. Bochum: Der Anschnitt, 187–199.
Stöllner, T., M. Mireskanderi, and K. Roustaei. 2011.“Mining
Archaeology in Iran. Investigations in Vešnave,”in A.
Vatandoust, H. Parzinger and B. Helwing, eds., Early Mining
and Metallurgy on the Western Central Iranian Plateau: The
First Five Years of Work. Mainz: Philipp von Zabern, 535–608.
Stöllner, T., B. Craddock, I. Gambaschidze, G. Gogotchuri, A.
Hauptmann, A. Hornschuch, F. Klein, I. Löffler, G.
Mindiashwili, B. Murwanidze, S. Senczek, M. Schaich, G.
Steffens, K. Tamasashvili, and S. Timberlak. 2014.“Gold in
the Caucasus: New research on gold extraction in the Kura-
Araxes Culture of the 4th millennium B.C. and early 3rd mil-
lennium B.C.,”in H. Meller, E. Pernicka and R. Risch, eds.,
Metalle der Macht. Halle: Tagungen des landesmuseums für
vorgeschichte, 71–110.
Tarantini, M., and A. Galiberti. 2011.La miniere di selce del
Gargano VI-III millenio a. C. Alle origini della storia mineraria
europea. Firenze: ed. all’Insegna del Giglio.
Tarantini, M., A. Galiberti, and F. Mazzarocchi. 2011.“Prehistoric
Flint Mines of the Gargano: An Overview,”in M. Capote, S.
Consuegra, P. Díaz-del-Río and X. Terradas, eds.,
Proceedings of the 2nd International, Conference of the
UISPP, Commission on Flint Mining, in Pre- and Protohistoric
Times,Madrid, 14–17 October 2009. British Archaeological
Reports International Series 2260. Oxford: Archaeopress,
253–263.
Timberlake, S., 2003.“Early Mining Research in Britain: The
Developments of the Last Ten Years,”in P. Craddock and J.
Lang, eds., Mining and Metal Production Through the Ages.
London: British Museum Press, 21–42.
Timberlake, S., and B. Craddock. 2013.“Prehistoric Metal Mining
in Britain: The Study of Cobble Stone Mining Tools Based on
Artefact Study, Ethnography and Experimentation,”Chungará
(Arica) 45(1). http://dx.doi.org/10.4067/S0717-73562013000
100002
Weller, O., and A. Figuls. 2007.“Première exploitation de sel gemme
en Europe: organisation et enjeux socio-économiques au
Néolithique moyen autour de La Muntanya de Sal de
Cardona (Catalogne),”in A. Figuls and O. Weller, eds.,
Trobada International d’Arqueologiaenversl‘explotació de la sal
a la Prehistória i Protohistória, Acts of the 1ST International
Archaeology meeting about Prehistoric and Protohistoric salt
exploitation, Cardona, 6–8 dec. 2003. Cardona: IREC, 219–238.
Weller, O., A. Figuls, and F. Grandia. 2007.“Première carrière de sel
gemme européenne: le Vall Salina à Cardona (Catalogne,
Espagne) au Néolithique moyen (4500–3500 B.C.).
Technologie, minéralogie et pétrographie de l’outillage lithi-
que),”in D. Morah, G. Dumitroaia, O. Weller and J.
Chapman, eds., L’exploitation du sel à travers le temps. Piatra
Neamt: édition Centre International de Recherche de la
culture Cucuteni, Institut d’archéologie Iasi, 115–133.
Zebrak, P. 1995.“Mining Hammers of Prljusa—Mali Sturac Site,”
in B. Javanovic, eds., Ancient Mining and Metallurgy in
Southeast Europe. Gornji: Archaeological Institute, Museum of
Mining and Metallurgy,37–43.
Hamon Salt mining tools and techniques from Duzdag
̆i (Nakhchivan, Azerbaijan) in the 5th to 3rd millennium B.C.
Journal of Field Archaeology 2016 VOL. 41 NO. 4528
Downloaded by [Caroline Hamon] at 01:04 20 July 2016

Online Supplement Captions
Online Supplement 1. Selection of rock types for hammer Types A–D in Duzdaği.
Online Supplement 2. Alkali basalt rock used for the manufacture of pointed hammers in
Duzdaği; A) Macroscopic photo of a flake; B) Thin section (Photographs C. Hamon and
G. Fronteau).
Online Supplement 3. Different techniques of grooves shaping on hammers: A) By
pecking of the natural surface; B) By flake removals followed by pecking; C) By fine
pecking and blunting caused by friction at the bottom of the groove (Photographs C.
Hamon).
Online Supplement 4. Representation of the types of modifications for hafting according
to hammer stones size in Duzdaği.
Online Supplement 5. Description of the explanatory experimentations on hammers:
shaping and hafting.
Online Supplement 6. Experimental shaping test carried out on a grooved hammer stone.
A) by flake removals and pecking starting at the edge of a cobble; B) by pecking of the
natural surface of the cobble (photos C. Hamon).
Online Supplement 7. Experimental reconstruction of different operating modes and
haftings of hammers in Duzdaği. A) hammers suspended on ropes to be directed against
the rock face; B–D) different types of hafting and fixing with ropes, wooden handles and
bark; E) cutting of the salt levels at the outcrop; F) creation of a groove for the contour
cutting of a salt block with a hammer of type E.
Online Supplement 8. Use-wear traces on archaeological hammers from Duzdaği:
combination of impacts and blunted area characteristic of the processing of semi-hard
mineral matter (10×).
Online Supplement 9. Distribution of the various use marks by hammer type in Duzdaği.

hammer Type A
hammer Type B & C
hammer Type D
porous basalt 4710
volcanic rock 53316
metamorphic rock 13 38 8
metamorphic limestone 782
sandstone 12 31 5
quartzite 55
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
hammer Type D
hammer Type B & C
hammer Type A

A
B
C

li •
*
M
~
*
,...;
.....
~
*
a-
* "' 1
0
-o
· 0
e::>
C?. C?.
·-
0
....,
1
"'
• •
1 1
"""'
~""
~""
~'v
,<v
x
<,._
·
~<;)
..
<,._
·
"""""
~"'
,<v
~c.,
<-~
c-?
00 00
o
...
.:;.'<;
'<.,
o"
.:,.'<-
o<--
6~
6~
0~ ~0
00
~c.,
""
'<-0
~"
.,
...
c'<-
o<v
~~
0
0~
~'?-
-~0
0
~"?-
~<;)
c-«-
v
~0
'<'?-
,,
~'
"\
x "
.,
<v
'<""
x
~<v
~'
x
""<v
\'v
00 00
o"
0~
0~
~0
~~
(
'">
(,
<v
~0
'<""
~'
~'

Online Supplement 5: Description of the preliminary experimentations on
the Duzdagi hammers: shaping and hafting.
In order to evaluate the mechanical constraints of the selected rocks, exploratory experimental
manufacturing of a dozen hammers representative of the main archaeological types was
carried out. These tests permitted a better understanding of the preferential link between
distinct rock types and specific modifications identified on the archaeological hammers. All
types of modifications were made on all rock types. The selection was oriented towards “pre-
shaped” stone blocks and involved modules and morphologies, which were naturally
favorable to distal modifications.
The shaping of notches was the most rapid and simple modification method implemented.
The most homogenous rocks (metamorphic rocks, sandstone and quartzite) more commonly
exhibit modifications such as notches or simple grooves, which are more easily created by
controlled knapping. In contrast, complex grooves are generally made on tools manufactured
from magmatic rocks with granular textures that are more suited to shaping by pecking.
However, these materials proved less resilient during use as they tend to wear out rapidly. In
addition, a few basalt hammers were recorded whose small size may be linked to the relative
density of these rocks and, therefore, to the need to obtain hammers with a minimum weight.

According to our own experimental tests, a complex groove involving the modification of
the distal end of the hammer enables a more solid fixing than a simple groove. We have
tested the adequacy of hammers of type C, D, and E with several binding and haft types,
connecting wooden handles and bark lashes or ropes (Online Supplement 7). These

reconstructions confirm that the distal and central peripheral grooves were more suitable for
the passage of ropes while the rectilinear grooves on the sides or the edges seem to be more
suitable for fixing a rigid, or semi-rigid, wooden haft.


a
b c d
e f

b