K
A R A H A N T E P E by Andrew Collins
Karahan
Tepe might be described as a sister site
to the more widely known Pre-Pottery Neolithic sanctuary of Göbekli Tepe.
Both are situated in mountainous terrain in southeast Anatolia
(the modern-day republic of Turkey), just a short distance from the ancient cities
of Sanliurfa and Harran. Both feature settings of T-shaped stone pillars, which
are anthropomorphic in nature and bear carvings either in high relief or 3D. Both
were built and then abandoned during the Pre-Pottery Neolithic B period. Yet whereas
Göbekli Tepe has received widespread attention, being excavated since 1995
under the auspice of the German Archaeological Institute in partnership with the
Sanliurfa Archaeological Museum, Karahan Tepe remains relatively obscure.
Karahan Tepe lies just beyond the eastern limits of the Harran Plain within the remote Tektek Mountains (Tektek Dagliari). It is approximately 22 miles (35.5 kilometres) northeast of the ancient city of Harran, 23 miles (37 kilometres) east-southeast of Göbekli Tepe and 28.5 miles (46 kilometres) east-southeast of the city of Sanliurfa (see fig. 1). As an archaeological site it occupies the northern extent of a roughly north-northeast to south-southwest oriented tepe (Turkish for "hill"), covering an area approximately around 13.23 hectares (33 acres).
The
hill, a natural formation of Eocene and Miocene limestone (see fig. 2), is approximately
490 metres (0.3 of a mile) in length and 270 metres (0.17 of a mile) in width.
It rises from a height of 675 metres (2,215 feet) above sea-level in the valley
below to 705 metres (2,313 feet) at its summit.
The
first thing the visitor notices when ascending the hill's northern slope are the
partially exposed heads of stone pillars that emerge into view from beneath a
hard deposit of soil that hides their stems (and sometimes most of their heads).
These sunken pillars climb toward the summit of the tepe for a distance of approximately
50 metres (164 feet), forming what appears to be a stone avenue. Its roughly north-northeast
to south-southwesterly orientation matches not only that of the hill, but also
that of the stones themselves. Two further pillars at the northern base of the
avenue are turned 90º and so perhaps formed an entranceway (like the U-shaped
gateway leading into the dromos feature forming part of Göbekli Tepe's Enclosure
C). At least three other stones nearby have the same alignment, complicating any
interpretation of the layout, and suggesting that the base of the avenue might
have included a built structure. What
might also be stone avenues, combined in some way with more complex features,
are visible on Karahan's eastern and southeastern slopes (no standing pillars
are visible on its western and southern slopes, which are predominantly exposed
bedrock without any substantial covering of soil).
Yet what does seem clear, however, is that all three "avenues" - the northern, eastern and southeastern, whose approximate azimuths are in the range of 15º, 115º and 140º respectively - are aligned toward the same spot; this being an exposed rock ledge or knoll immediately north of the hill's extended summit.
The exposed
surface of this northern knoll marks the beginning of an extensive area of bedrock
covered with groupings of deeply bored cupules, or cup marks, usually 15 to 20
centimetres (6 to 8 inches) in diameter and easily as much in depth (see fig.
5). Similar cupules are found on exposed bedrock at Göbekli Tepe, close to
Enclosure E, the so-called Felsentempel ("Rock Temple"), and
also at other Pre-Pottery Neolithic sites such as Basaran Höyük (Güler,
Çelik & Güler, 2013, 295, fig. 11) and Hamzan Tepe (Çelik,
2010, 262, fig. 6), both located in the Sanliurfa province.
Other,
much larger holes, ranging in diameter from between 40 to 50 centimetres (15 to
20 inches), are also present within the exposed bedrock. At least three sets are
placed together in pairs, giving them the eerie resemblance of dark eyes gazing
up at the beholder, a fact that, regardless of their true function, seems deliberate
(see fig. 6).
Beyond
the carved bedrock immediately behind Karahan Tepe's northern knoll very little
evidence of occupation is visible. The southern and western slopes seem devoid
of any bedrock carving, other than a curved groove, around a metre across, carved
into a horizontal rock face about half way up the side of the southeastern slope
(see fig. 7). It terminates in a deliberate vertical fracture, probably the result
of a pillar being forcibly removed from the bedrock.
Small Finds Various small finds, including carved fragments of mini T-shaped stones and right-angled corner sections of what appear to be porthole stones, like those found at Göbekli Tepe, are to be seen on Karahan's eastern slopes (see fig. 8). Most likely these porthole stones formed vertical or horizontal openings into now lost enclosures. One noticed by the author in 2014 was being used on the summit of the hill, close to the northern knoll, to line a fire pit. Since this fragment of carved stone is likely to be over 10,000 years old, this seems a tragic misuse of such an important relic of the past.
Stone Tools Stone tools, tool fragments and discarded flakes are to be seen everywhere at Karahan Tepe. Those observed include arrowheads (identified as mostly Byblos, Nemrik and Aswad points, see Çelik 2011, 244), scrapers (both side and end scrapers), borers, hammer stones, and sickle blades. These are fashioned mostly from a grey to brown flint (which was also the most favoured type of flint used at Göbekli Tepe), although fragments of stone tools in black flint and what appears to be white quartz have also been noted (although, oddly, these were seen only on the tepe's western face, close to the unfinished monolith - see below). No obsidian tools have been observed, although a fair quantity were found and recorded by Çelik and his team (see Çelik, 2000, 7, and Çelik, 2011, 243-5).
A cursory
examination of stone tools, combined with the complete lack of any pottery shards
at Karahan Tepe, makes it clear that the site was active during the Pre-Pottery
Neolithic, ca. 9500-6000 BC.
In size and appearance the T-shaped pillars at Karahan, which when fully exposed would stand around 2 metres (6.6 feet) in height, are comparable with those found in the youngest structures at Göbekli Tepe, such as Enclosure F, the Löwenpfeilergebaude ("Lion Pillar Building"), and the various cell-like rooms located in younger layers west of the four main enclosures, all of which date to Göbekli Tepe's final phase of construction, Level II, during the early Pre-Pottery Neolithic B period, ca. 8500-8000 BC (similar reduced sized T-shaped stones have been found at other Pre-Pottery Neolithic sites in southeast Anatolia, such as Sefer Tepe, Hamzan Tepe and Gusir Ho¨yuk (see, for example, Güler, Çelik and Güler, 2013 & Çelik, 2010, 258-9, Fig. 2). It thus seems likely that this was the main period of building construction at Karahan Tepe, a conclusion drawn also by Çelik (see Çelik, 2000, 7, & Çelik, 2011, 246).
The
presence of a much larger, unfinished monolith still attached to the bedrock on
Karahan's western slope indicates that even larger, and thus much older, pillars
probably once stood at the site (the general rule at Göbekli is that the
bigger and more sophisticated the pillar, the older it is, with the earliest dating
back to the Pre-Pottery Neolithic A period, ca. 9500-8500 BC).
On
two T-shaped pillars found at Karahan Tepe a carved snake is seen to slither up
its front narrow edge. On one, originally found in 1997 and now removed to Harran
University (see fig. 11), the snake looks like a human sperm, with a round, bulbous
head and wavy body (Çelik, 2000, 6, fig. 1, & Çelik, 2011, 243,
figs. 8,9 & 10), while on the other example, exposed during illegal digging
operations and first observed by Çelik and his team in 2011, only the dome-shaped
head of the creature is visible - the rest of its body remaining unexposed beneath
the ground (Çelik, 2011, 243, 247, figs. 11 & 13). In
addition to this a fragment of a chorite bowl, found at the site and dating to
the same age as the T-shaped pillars, bears the relief of a zigzagging snake (Çelik,
2011, 246, fig. 24:7).
Since
the tepe rises to its maximum height immediately south of the aforementioned northern
knoll the natural inclination when standing there is to look north (see fig. 12).
Doing so directs the eye north-northwest past the farmhouse to a prominent hill
or tepe, as noted by Bahattin Çelik in his report of the site published
in 2000 (Çelik, 2000, 7). Located exactly 1 mile (1.6 kilometres) away
from Karahan Tepe, the left-hand edge of its summit is at 338º azimuth with
its right-hand edge at 341.25º (see fig. 13). This provides a mean azimuth
for the centre of the hill summit of 339.63º. In
archaeoastronomy a backsight is a position, ideally a prehistoric sanctuary, from
which observations are made toward a foresight, usually a conspicuous geographical
feature over which, or behind which, a celestial body is seen either to rise or
set (see Gil and Belmonte, 2009, for a good example of a proposed prehistoric
foresight/backsight relationship between two sites on Gran Canaria).
Chartered engineer Rodney Hale was asked to investigate the matter. He determined that during the proposed epoch of construction at Karahan Tepe, the bright star Deneb (a Cyg) in Cygnus, the celestial bird, would have been seen to set into the summit of the flat-topped tepe as viewed from Karahan's northern knoll. The optimum period of observation was calculated by Hale to have been between ca. 8685 BC, when Deneb set into the hill summit's eastern edge at an azimuth of 341.25º, and ca. 8375 BC, when the star set into the hill summit's western edge at an azimuth of 338º. This information, based on the established rate of precession of Deneb at an extinction height of 2º, shows also that in ca. 8550 BC Deneb would have set down into the central area of the hill. These three dates-8685 BC, 8550 BC and 8375 BC-are simply estimates, and variations might easily be applied. Yet they do coincide pretty well with the main period of occupation at Karahan Tepe during the early Pre-Pottery Neolithic B period, ca. 8500-8000 BC. However, as Çelik speculates himself, it is possible that construction began here during the late Pre-Pottery Neolithic A period, ca. 9000-8500 BC (Çelik, 2011, 247), the unfinished monolith located on the west side of the hill perhaps being evidence of this fact. Cygnus and the Milky Way The
star Deneb features also in proposed astronomical alignments at Göbekli Tepe
(Collins, 2013 and Collins, 2014a), whereby the mean azimuths of twin central
pillars in two enclosures, C and D, target this same star's setting during their
proposed epoch of construction, ca. 9500-9000 BC (Collins, 2013a, Collins, 2014a,
and see, for instance, Schmidt and Dietrich, 2010, for the radiocarbon dating
evidence for Enclosure D). The significance of this star appears to lie in the
fact that it marks the northern opening of the Milky Way's Great Rift, known also
as the Cygnus Rift. Elsewhere the author proposes that this noticeable fork, or
division, in the Milky Way, which appears to split the Milky Way into two separate
streams, was seen as early as the Palaeolithic age as the entrance to a sky world,
a kind of cosmic womb from which souls emerged from before incarnation, and ultimately
returned to in death. Often this journey saw the soul taking on the form of a
bird, usually a swan, goose, eagle or vulture, all of which are associated not
only with the transmigration of the soul in various Eurasian cultures, but also
with myths and legends surrounding both the Cygnus constellation and the Milky
Way in its capacity as a road or river along which souls were able to reach the
afterlife (see Collins, 2006, Collins, 2014a, and Collins, 2014b).
The Bald Place Several
interesting facts emerged during the author's visit to Karahan Tepe in June 2014.
After leaving the hill our party was invited to share the hospitality of the farmer.
This provided the ideal opportunity to ask a number of pertinent questions about
the site. For instance, the farmer related how the importance of Karahan Tepe
was not realised until Bahattin Çelik's first visit in 1997. In other words,
no one had ever noticed the carved stone pillars, or any other carved object or
worked stone tool, present on or around the hill site. We
determined also that no folklore or legends are known to surround the site, which
is unfortunate as this might have helped us better understand the manner in which
the hill was viewed by past inhabitants of the area.
Place of Emergence If so, then it implies that Keçili North Tepe, arguably the original Keçili Tepe, was once considered female in nature. Should this prove to be right, it would make sense of the site's apparent synchronization in the ninth millennium BC with the bright star Deneb and the Milky Way's Great Rift. In their dual role as a perceived entry point to a sky world Keçili North Tepe might have functioned as a representational place of emergence of humankind, similar to the concept of the sipapu among the ancient Pueblo peoples of the American southwest. Sipapu was the name given to the portal through which the first peoples were thought to have emerged from the underworld, envisaged in terms of the womb of the Great Mother. The sipapu was identified by the Hopi tribe with a flat-topped hill with a circular depression in its summit located in the Little Colorado Gorge, just upstream from where the Little Colorado meets the Colorado river, in the Marble Canyon area of the Grand Canyon (O'Brien, 2012, and see fig. 16).
Symbolically and ritualistically the sipapu was represented by a small circular hole cut into the floor of the Hopi's kiva huts (see fig. 17). By entering the kiva, the Hopi initiate was able to return to the Great Mother (Leeming, 1996, 28), a process that might well be reflected in the design and layout of Pre-Pottery Neolithic cult buildings at places like Göbekli Tepe.
Circular
holes like those found in Hopi kivas are also seen at Göbekli Tepe. Large,
flat, rectangular stone slabs with porthole-like apertures are located in the
perimeter walls of Enclosures C and D. In each case the holed stones are positioned
toward the north-northwest, exactly in line with the mean azimuths of the twin
central monoliths erected in both structures (see fig. 18).
Karahan's Ancient Population Like
Göbekli Tepe, Karahan would appear to have been abandoned sometime around
8000 BC. Why this happened when it did remains unclear. However, with the emergence
of agriculture and animal husbandry right across the region sometime around 9000
BC, it is possible that religious activities began changing in accordance with
this new way of life.
Unlike the older, much grander, structures found at Göbekli Tepe, such as Enclosures B, C and D, which are all aligned north-northwest to south-southeast, due most probably to its builders' interest in the rising and/or setting of stellar objects in the polar regions, the Lion Pillar Building is aligned almost precisely east-west (see fig. 19). Rearing lions appear in carved relief on the inner faces of twin pillars positioned at the structure's eastern end (see fig. 20), the direction, of course, of the equinoctial sunrise. In
this instance the carved lions perhaps represent the might and influence of the
sun (in much the same way that the lion-headed goddess Sekhmet did in ancient
Egypt). Equally, these leonine creatures could well have symbolised the star group
we know today as the constellation of Leo, the celestial lion of Greek and Babylonian
astronomy (see Belmonte and García, in press). During the ninth millennium
BC the stars of Leo rose heliacally at the time of the spring equinox, exactly
in line with the easterly orientation of the Lion Pillar Building.
The
focus of Karahan Tepe's three stone avenues toward the site's northern knoll,
where the exposed bedrock is covered in cupules, as well as larger twin holes
and a possible water cistern, tells us that this was most likely its principal
area of ritual activity. That this northern knoll might also have acted as a backsight
for observations of the Cygnus star Deneb and the opening of the Milky Way's Great
Rift, which at the time would have been seen to set down into Keçili North
Tepe, suggests that what was occurring here, ca. 8500-8000 BC, was related to
the proposed stellar-based cosmology established at Göbekli Tepe as early
as ca. 9500-9000 BC (that is, before the transition to an agricultural-based economy
among the peoples of the Pre-Pottery Neolithic in southwest Asia).
Barbier,
Edward B. Scarcity and Frontiers: How Economies Have Developed Through Natural
Resource Exploitation. Cambridge, Cambs.: Cambridge University Press, 23 December,
2010. The author would like to extend his gratitude to Rodney Hale, Catherine Hale, Juan Antonio Belmonte, Greg Little, Richard Ward, Hugh Newman, as well as all those who participated in the Origins of Civilization tour in May-June 2014, for the part they played in the creation of this article.
"One Week in Kurdistan" - Andrew's first visit to Karahan Tepe and Göbekli Tepe in June 2004: http://www.andrewcollins.com/page/articles/kurdistan.htm
Hugh
Newman's video report on Karahan Tepe's unfinished monolith Short
article on Karahan's unfinished monolith by Hugh Newman
|