ArticlePDF Available


The paper (a continuation of a prior paper by this author on the Neolithic settlement of Boncuklu Tarla) discusses the Architecture (floor and site plans) of the temple associated with the archeological site near the town of Dergecit, in the district of Mardin, in South-Eastern Anatolia, Turkey. In spite of the very limited published archeological information available on this settlement, there is enough imagery related documentation to extract some relatively firm conclusions regarding both the temple and its immediate surrounding structures and environ.
Dimitrios S. Dendrinos, Ph.D.
Professor Emeritus, University of Kansas,
School of Architecture and Urban Planning, Lawrence, Kansas, USA
In Residence at Ormond Beach, Florida, USA
May 10, 2021
Part of the Boncuklu Tarla archeological site as reported on November 30, 2019
in reference:
Table of Contents
The Architecture of the temple at Boncuklu Tarla
The site
The temple
Copyright statement
Photo A. Close-up photo of the temple at Boncuklu Tarla. Source:
The paper (a continuation of a prior paper by this author on the Neolithic settlement of Boncuklu
Tarla) discusses the Architecture (floor and site plans) of the temple associated with the
archeological site near the town of Dergecit, in the district of Mardin, in South-Eastern Anatolia,
Turkey. In spite of the very limited published archeological information available on this
settlement, there is enough imagery related documentation to extract some relatively firm
conclusions regarding both the temple and its immediate surrounding structures and environ.
Various phases involved in the buildup of the settlement, to the extent that they fall within the
temple’s immediate spatial-temporal context, are discussed. The entrance to the temple
question is analyzed. A key ratio involving the temple’s interior floor plan dimensions is
established. The question as to why was the temple located where it is and the dimensions in its
site plan, is also addressed, along with its orientation.
This paper is a continuation of the paper by this author in reference [1]. The archeologists
involved in the excavation at Boncuklu Tarla (to be referred to as BT from now on), which
commenced in 2012, see reference [2], hypothesized as late as December 2019, see references
[3] and [4], that the settlement is either contemporaneous or about a millennium older than the
settlement at Gobekli Tepe (GT). Furthermore, they also hypothesized that the public structure,
which they referred to as a “temple”, was constructed approximately five centuries after the site
was settled, see reference [5]. Moreover, they also suggested that the temple at BT bears
similarities to the temples at GT; and, furthermore, that the temple at BT is contemporaneous to
the GT’s Temples. Certain contradictions in the news reports have been pointed out in reference
[1] by the author. The finding about a sewer system (argued by the archeologists of BT to be the
oldest known) is cited in reference [6]. This is issue is revisited at a later section of this paper.
Arguing against the archeologists’ suggested dating, the author recommended that the
settlement at BT must be viewed in the context of Evolution of Early Neolithic Architecture. The
Architecture and Engineering (through its building technology and construction techniques) of
the settlement, as detected by the partially exposed excavated thus far masonry walls and floor
conditions and seen in Photo 1, shows clear evidence of late and advanced Natufian Architecture.
On the Natufian Architecture, and its Evolution during the Early Neolithic, see references [7], [8],
[9] and especially reference [10] where this author presented the architectonic evidence.
The author argued, in reference [1], that BT is a late 8th millennium BC settlement, and its temple
is an early 7th millennium BC structure. That structure predates the GT Temple structure D layer
III construction by more than a millennium and possibly a millennium and one-half. The
arguments were based on the Art, Architecture (construction methods and building techniques),
Engineering, Geography, Topography, Climate conditions, etc., associated with both sites. The
paper also utilized evidence from the settlement (and the late Temple construction) at Nevali
Cori (NC). These arguments, which can also be found in four related papers written by this author
with regards to GT and related early Neolithic sites, see references {7] – [10], will not be repeated
here. Instead, the focus will be exclusively on the Architecture of the temple at BT and its
surrounding structures within its immediate vicinity. Evidence will utilize publicly available
photos and statements by the archeologists involved in the archeological dig.
Photo 1. Aerial view of the partially excavated section of the settlement at Boncuklu Tarla (BT),
in the vicinity of the temple (the rectangle-shaped structure at top center-left). The Photo,
apparently showing the excavated area as of December 4, 2019, is found in reference:
Some lithic evidence, together with amulets and other decorative and pendant items, especially
a multitude of beads, the name assigned to the archeological site means “beads field”, have been
uncovered at BT. Since no stratigraphic evidence is available, it is rather impossible for this author
to evaluate the importance of these artifacts in reference to the structures under consideration
there, and especially the structures’ presumed dating. Thus, this evidence will not be addressed.
However, it is noted that, in general, archeological artifacts, found within the fill or at the vicinity
of a structure, are only at the margins related to the Archeology of the structure. Only artifacts
unearthed directly underneath a structure can, with some degree of certainty, be directly
connected with the dating of the structure. The fact that almost all structures unearthed thus far
at BT seem to have compressed dry clay (rather than solid) floors, complicates matters further.
The Architecture of the temple at Boncuklu Tarla (BT)
The site
In what follows a cursory and preliminary (for lack of more documentation available to the public)
review and assessment will be offered on the section of the site that the author has available
some evidence, as partly shown in Photo 1 (above, page 4). One thing is quite clear from the
above photo, of this partially excavate archeological site: that the ruins the archeologists have
come across do not represent a site “plan”. Structures were added, and existing structures were
transformed, as the centuries run their course and successive generations (of possibly different
cultures and sub-cultures) inhabited and managed the structures of the site.
Moreover, it is highly likely (almost certain) that BT underwent different phases in its lifelong
construction, destruction, reconstruction, and habitation events. Being close to a river, this
settlement must have been constantly vulnerable to flooding, as all riverine communities are
over the centuries and millennia of habitation. Phases in the construction of this site are
recognized by the archeologists, who came across overlapping masonry structures, see ref. [1],
[2], [3], and [4]. Besides flooding, and over the millennia of its habitation, quite likely the
settlement experienced its own share of invasions, ensuing destruction, and then renovation and
reconstruction events and phases. All these environmental, social, and cultural factors must have
contributed to the architectural Evolution of the settlement’s structures, seen through the
(partial at this point in time) excavation of the site. The strong differences in architectonic style
among the site’s structures (extended over a period of more than two millennia) are, to an extent,
shown in the cover page photo, Photo A (in the Table of Contents page) and Photo 1 (page 4).
A close look into the evidence, revealed through the archeologists’ excavation thus far at BT,
provides strong support to the hypothesis that the temple at BT, see Photo A and Photo 1, is a
later construction than its surrounding masonry structures, both irregularly rectangular and arc
(partially elliptical, or quasi elliptical) in floor plan shape. With this conclusion, that the temple is
a later construction than the settlement’s original habitation, the author is in complete
agreement with the BT archeologists, who placed the temple’ construction later than the initial
settlement date of BT. Also in agreement is the author with the archeologists’ belief that the
original settlement predates the temple period by about half a millennium.
However, the author is in disagreement with the archeologists as to what era these structures
(that surround the temple as well as the temple itself) belong. The author is of the opinion that
these structures represent advanced Natufian Architecture, due to the rectangular shapes, and
the thinness of their masonry (tied with mortar) walls. On that basis, he places them in the middle
to late 8th millennium BC time frame. The archeologists believe that these structures were built
in the latter third of the 10th millennium BC. This author finds this belief to be way off mark, by
at least two millennia, and hence respectfully disagrees with the archeologists.
The layered construction, with superimposed generational habitation, but with mudbrick (not
masonry) walls, has been also encountered at Catal-Hoyuk (as already discussed in references
[1], and [7]-[10]), a late 8th to middle 6th millennium BC life span site. Environmental conditions,
Geology, Geography, Topography (ground’s slopes, incline), Architecture and Engineering, are all
instrumental factors in the layering of structures. But the recycling of materials involves also both
Economics and Culture related forces, beyond the Architecture and Engineering building related
technological knowhow to design, transform, and construct such layered edifices. A level, of
sophistication is needed to carry out such construction, destruction, and reconstruction
processes, especially in reference to masonry walls, largely absent from the archeological record
at such far back dates (as is the case with the BT settlement).
There is an outside chance that at the middle of the 8th millennium BC, BT could be a source for
such innovations for masonry buildings, as Catal-Hoyuk might have been the source for mudbrick
multi-layered structures type of building technology during the same millennium (but quite late
in it, circa 7100 BC). The available evidence however overwhelmingly does not support the
proposition that BT was the innovator, although it might have been among the pioneering
settlements to adopt it. This point will be revisited in a bit. The record strongly seems to support
the proposition that such building technology was not in the cards back in the 8th millennium BC
time frame, let alone into the 10th millennium BC period (that the archeologists suggest).
Three arc-based structures that surround the temple are of great interest: the part of an ellipse-
shaped structure at bottom right, see Photo 1; an arc-shaped structure at bottom middle; and a
quasi-elliptical structure at middle left. These three, along the remnants of the structure at right
and that at left of the long sides of the temple, indicate that, beyond the temple being older than
all five of these distinct structures, it is also a reformatted structure from the two, earlier phase
construction, open rectangular structures at its two flanks.
What the above photo 1 seems to suggest is that there were at least three phases of construction
within the frame shown. The arc-shapes structures (the oldest) are phase I possibly of the
settlement at BT, and represent typical (advanced) Natufian type Architecture; the two open
rectangular structures at the temple’s flanks likely represent phase II construction at the BT
settlement, and definitely correspond to late Natufian Architecture; and finally, phase III of
construction is the temple phase. The overall time span, encompassing the initial construction of
these three types of structures, must have covered in toto about half to one millennium. The
entire lifespan of this section of the BT site must not had exceeded two millennia, lasting well
into the 7th millennium BC (and possibly the very early part of the 6th millennium) time period.
However, these are speculative statements, as not much of concrete evidence has been
produced, so far, by the archeologists of BT regarding either the time span of its habitation, its
successive transformations and numerous destruction events through the millennia of it being
inhabited, and the conditions of its eventual abandonment. This author estimates that the
abandonment of the BT site must have preceded or (at the latest) coincided with the initial
construction period at GT. This estimate is grounded on the observation that the BT Architecture
(in its entirety) predates that of the initial phase of construction (pegged by this author to have
occurred at the middle of the 6th millennium BC) at GT.
Diagram 1. Boncuklu Tarla, the site plan of the temple’s immediate vicinity. Source: the author.
Like all such physical structures, these structures too must have undergone numerous
transformations, possibly layering, as succeeding generations used them, each generation
adjusting them intermittently to fit its needs, and respond to the environmental and cultural
challenges facing it. Structures of all three phases evolved, as they were built, inhabited and
maintained under different climatic conditions (indicated in the differing thickness of their walls
and their masonry construction techniques and building engineering methods). They served
different cultural and social purposes, which must have also evolved, marginally and at times
significantly, possibly in the temporal scale of the two millennia (the 8th and 7th millennia). The
length of habitation at BT is estimated by the archeologists to have lasted about three millennia,
reference [4]. This author disagrees with this estimate, as he estimates this settlement to have
lasted at most two millennia, placing this time frame approximately in the middle to late 8th
millennium to the early to middle 6th millennium BC period, see also reference [1].
Length in the habitation of a settlement, as well as high turnover in its cultural composition are
elements of cultural turbulence, as well as of not a prolonged habitation. Hence, a three millennia
habitation of the BT settlement, as suggested by the BT archeologists, is highly unlikely. Three
millennia is about half of the Neolithic in duration. Such durability is not likely for any Neolithic
settlement, and especially so for this part of Eurasia. Socio-cultural turmoil has been abundant in
the Anatolian and Mesopotamian Regions over the millennia. Dynamical stability over multiple
generations is not an attribute of the Levant, it has never been. On this basis, the author feels
comfortable in assessing the durability of the BT settlement to not have exceeded two millennia.
Then, there is another feature of the BT settlement that somehow assists in its chronology. The
sewer system that the archeologists discovered, see ref. [6], must have been set up (depending
on environmental and cultural conditions) in the middle period of the settlement. The practice
(also encountered in the settlement of NC) must have been widespread at the time and at that
Region of the Levant. There is absolutely no reason to suggest that BT was the innovator for such
a system (as is the case with the layering of residential and other public structures). BT was not
at the center of a large system and network of settlements, neither did it have the economies of
scale to be the originator of such innovation(s), not according to what the excavation has
uncovered thus far (and has been communicated to the public), besides the archeologists’ simple
(expressed in the press) opinions.
At this point, a close look into the three arc-shaped structures (speculated to have been built at
a prior phase than the temple) will be taken, see Diagram 1 (page 7), and Diagram 2 (page 10). A
note of caution: in all drawings, lines and arcs associated with remnants of masonry walls are
approximately placed. Due to lack of any specific measurements, ratios have to be estimated
and computed from photographs that to some degree they distort shapes. On top of this
photographic distortion related difficulty, the ground level (either floors of structures or the
outside ground level) have not been totally uncovered and appropriately graded. Hence, the
lines’ domains are fuzzy, not strictly and exactly determined (or at least to any satisfactory
enough degree of accuracy). Especially this is the case with imaginary (suggested) lines associated
with the orientations, alignments, and shapes that arcs, quasi-elliptical and open rectangular
shapes seem to form.
Having provided these caveats, it must also be noted that a few aspects of the structures’
Architecture are reasonably clear and close enough to their actual state that lend themselves to
reasonable analysis that derives some relatively safe and satisfactory conclusions.
In Diagram 1, the site drawing of the area surrounding the apparently public space, designated
as a “temple” by the BT archeologists and as such being referred to in this paper (and in ref. [1]),
is shown (structure T). What impresses the observer of this site arrangement of the four
structures in question (setting aside for the moment the two open rectangular structures at the
temple’s flanks), is that the three (partly) arc-shaped (quasi elliptical) structures (designated as
A, B, and C in Diagram 1) seem to point towards a specific area inside the floor plan of temple T.
Moreover, the (approximately drawn) axes emanating from the centers of two arc-forming,
peripheral to the temple, structures (A and C) and cutting across their (approximate) middle (lines
L1 and S1 correspondingly) are parallel to the (approximately drawn) sides of the rectangular in
shape floor plan of temple T. It leaves little doubt that temple T’s structure (it will be analyzed in
the next section) was located there, and it was given by its architects the specific orientation and
dimensions it had in its life span, for two at least reasons. First, because that seems to have been
the location of some (at least two) points (designated as P1 and P2 in Diagram 1) of social and
cultural interest (the point of conversion of the three lines L1, S1 and S2, as well the point of
conversion of lines L1, L2, and L3). And second, because the center of the circular stone structure
inside building A (possibly a Neolithic hearth?) is approximately at a distance from the side of T
facing A equal to the length of T; and the center of building C is at a distance from the side of T
facing C approximately twice the width of T. Hence, the location of T relative to the pre-existing
structures A, B, and C, was not random. Arcs and quasi-elliptical floor plans, to this author, signify
intent by the architect of a structure to indicate the presence at a site of a Bull Cult. The three
axes must have had broader alignments of interest relative to the horizon, which the author can’t
identify (as the source Photo 1 of this site plan doesn’t offer a compass). L1 and L3 are at 80.
The temple
In Diagram 2, page 10, the temple T (of Diagram 1) is shown, in larger scale (obtained by enlarging
it from Photo 1). The scale can’t be drawn as it is not known; the archeologists have not made
public, to this author’s knowledge, the dimensions of this temple. It is again noted that the
possible photographic distortions from photo 1 are magnified a bit as a result. The archeologists
in charge of the excavation at BT have left a lot to be desired in terms of information available to
the public related to the Architecture of T (and of all other structures at BT as well). Hence,
dimensions of the structure are relative, preliminary and in need of confirmation. The temple
was discussed in ref. [1]; hence, the reader is also directed to that reference for more analysis,
as some points made there about it will not be repeated here. Because of the uncertainty
regarding the width of the temple’s walls, the focus will be directed towards the temple’s interior.
The first question anyone, not just an architect, would ask for any structure is this: how did people
get inside this building? Where is the entrance (ingress)? The author speculates that the original
intended door to the structure was the (now seemingly blocked) larger recess at the upper-left
side of the structure. Apparently, at the time of T’s abandonment, the doorway was blocked. T
could have had two entry points, the one where the width symmetry axis (line L2 in Diagram 2)
goes through; as well as the recess below it at left. Both have later been blocked. This element
of T, the blocked entry (or entries), may signify that the building was abandoned before the
settlement was itself abandoned. Possibly, a different culture invaded the settlement and either
buried the temple, or partially destroyed it.
An alternative point of entry is the edge C in Diagram 2. Photographic evidence, see Photo A, in
page 2. It shows a narrowing in the wall’s thickness at point C (of the temple’s ABCD rectangle),
with a possible accessibility ramp there. There is another point of potential entry into the building
T: the corner point at the lowest part in Diagram 2, at point A, of the ABCD rectangle. However,
it seems that entrance at that corner would had been far too narrow to accommodate massive
ingress to and egress from the structure. In view of the uncertainties surrounding the culture that
built T, and those cultures that used it (or abused it) thereafter, it is rather speculative to further
be more specific regarding the entry to and exit from the structure according to the original
design. However, it can’t be underemphasized that understanding where the entry to this
structure was, is of paramount importance.
Diagram 2. The temple at Boncuklu Tarla (BT); floor plan, an enlargement from Photo 1. The
upper-most stele (top) is the one that is sheered-off, and its top is missing. At point C (see text)
there could be a possible point of entry into the structure. Source of diagram: the author.
It is also noted, that (in spite of any photographic distortion, as well as uneven excavation levels
at different parts of the structure’s floor shown in photo 2) a characteristic feature of this building
stands out: all four sides of the approximately rectangular structure are arcs – not straight lines.
This feature might also provide evidence of some relative degree of primitiveness in construction
techniques, as this author has argued in the past (see ref. [7]-[10]), arc (and quasi or semi-
elliptical) shaped forms preceded both perfectly circular and rectangular shaped floor plans in
the Evolution of Early Neolithic Architecture. In approximation, and in relative terms, one can
estimate the relative lengths of the two (approximately, but not exactly equal) pair of the
structure’s two sides; the two shorter sides that will be referred to as the “width” (W) of the
temple; and the two longer sides that will be referred to as the “length” (L) of the temple. It is
again an indication of primitiveness that the two lengths (L) are not equal, and neither are the
two widths (W).
Given this caveat, it is found that the ratio r of L/W is approximately r=7/5=1.4, a very interesting
(exact) ratio of two prime numbers. Exactness directly implies that this ratio r is just an
approximation of the temple’s actual sides’ ratio. Establishing the approximate building’s floor
area modular structure is then straightforward; and so is the estimate for its modulus (x, y): it is
a rectangle with length x approximately one seventh of the structure’s length; and, equivalently,
with width y approximately one fifth of T’s width. Of course, the ratio x/y=1.4 and one is,
consequently, close to deriving an estimate of the unit of measuring length, l, utilized by the
culture that designed and built this temple, most likely circa 6800 BC, given the actual
measurement of T’s floor area (an approximation to which the archeologists surely know).
Within the temple, four stelae are found. They are all different, possibly indicative of the differing
symbolism they carried by the culture that built the temple, and the differing roles they played
within the temple’s context. Also, possibly, these stelae’s differences can be attributed to the
primitiveness of building techniques used in the extraction and processing of the monoliths they
were carved from at the quarry. Imperfection in building techniques and construction details
(anomalies in uniformity) reveal the degree to which gradual advancements in Architecture, that
had occurred during Neolithic Evolutionary Time, had reached the builders of T.
In the pair of stelae at left in Diagram 2, it can be discerned that the two columns share some
commonalities, as does the pair of columns to the right. The pair at left consists of two stelae
that seem to be more squarish than the clearly rectangular in shape pair at right. Moreover, the
line L3, a line drawn parallel to the temple’s length and tangent to the bottom pillar (stele,
column) from the outside, seems to be tangent to the pillar at right from the inside. On the other
hand, the lines L4 and L5 of Diagram 2 (drawn parallel to the temple’s length) seem to contain
both pillars. Clearly, this difference is due to construction imperfections.
The four pillars (stelae, orthostats, columns, however one wishes to call them) are located off
the modular grid of the T’s floor plan. This might be another indication of the relative
primitiveness in both the design and construction of the temple. However, the four stelae are
located approximately equidistant from the two axes as drawn (the longitudinal axis L1, and
latitudinal axis L2). Again, the degree of approximation is indicative of the degree of primitiveness
found in the temple’s design and construction. However, much more detailed measurements are
needed to corroborate these findings. (Note: The lines L1 and L2 of Diagram 2 are not the same
as lines L1 and L2 of Diagram 1). Very little can be inferred as to the thickness of the temple’s
(and perimetric structures’) walls, from the photographic evidence available. However, the
original thickness, and its mortar cum stone composition, are elements of extreme interest in
determining the structures’ dates (the thicker the walls, the older the structure, quite likely).
Finally, a few remarks are in order to address the relative size and location of the recesses found
on the upper left part of Diagram 2, running along the length of the temple’s masonry wall, which
have been categorized as possible (blocked) ingress points to T; and the two protruding parts
(possibly seats, or resting places for religious or ceremonial objects) of the facing wall running
along the length of the temple (the bottom corner A of the rectangle ABCD having been thought
of by this author as a possible narrow ingress point). It is obvious that the four segments (two on
each longitudinal temple’s walls) were placed there by design. The intent is not clear, as much
more documentation is needed before any statements can be made regarding their intended
functions. Obviously, estimates about the height of the walls, and the cover of the building must
be part of the factors that influenced these four segments’ relative locations inside the temple.
It is noted however, that line (latitudinal axis of symmetry) L2, although going through the center
of the ingress located approximately at the middle of the upper longitudinal wall, is tangent at
the bottom part of the protruding segment, located close to the middle of the righthand side
longitudinal wall. This asymmetry may also be either an imperfection in construction technique;
or a sign of different functions assigned to the two protrusions (seats). Answers to these queries
likely will never be obtained with a high degree of confidence, as the messages contained in
buildings of the past (especially that far back in History) were intended for an audience long gone,
of cultures that have disappeared from the collective practices, customs, memories of, and have
left few if any traces to, contemporary cultures.
The main conclusion of this paper is that much more is needed to ascertain the key findings of
this paper, surrounding the temple at Boncuklu Tarla, its Architecture and surrounding structures
found at its immediate archeological vicinity. Regarding the temple itself, a key ratio of its interior
length to width was established (1.4 as the ratio of two prime numbers: 7 and 5).
With a high degree of eagerness and anticipation long-awaited and expected documentation will
hopefully soon become available, from the team of archeologists of this interesting site. Then,
reasonable approximations to the structures’ exact location, orientation, and dimensions, such
as the temple’s wall thickness, the length l and the rectangular modulus (x, y) of the temple’s
interior, as well as the form of the surrounding structures walls’ precise location and widths will
be reasonably set and estimated.
The exact location, as well as the form of the surrounding the temple structures, are not only
expected to provide support to the hypotheses expressed in this paper; but also, to document
the reasons as to why was the temple located there; and why it has the apparent interior
dimensions that it has, with a ratio r (approximately equal to 7/5=1.4).
Reference [1]: (PDF) BONCUKLU TARLA: Why evidence also from this Neolithic settlement supports the
theory that Gobekli Tepe is a 6th millennium BC site | Dimitrios S Dendrinos -
Reference [2]:
Reference [3]:
Reference [4]:
Reference [5]:
Reference [6]:
Reference [7]: (PDF) Dating Gobekli Tepe | Dimitrios S Dendrinos -
Reference [8]: (PDF) Gobekli Tepe: a 6 th millennium BC monument | Dimitrios S Dendrinos -
Reference [9]: (PDF) A Primer on Gobekli Tepe | Dimitrios S Dendrinos -
Reference [10]: (PDF) Gobekli Tepe, Tell Qaramel, Tell Es-Sultan: Why is Gobekli Tepe a 6 th millennium
BC site, and Evolution of Early Neolithic Architecture | Dimitrios S Dendrinos -
Copyright statement
Dimitrios S. Dendrinos
The author, Dimitrios S. Dendrinos, retains all copyrights to this paper, with the exception of the
photos, which are in the public domain and have their own references. The rest of the paper
belongs to the author. No section of this paper, in part or as a whole, can be reproduced in any
form (electronic or otherwise) without the explicit and written consent by the author, Dimitrios
S. Dendrinos.
Full-text available
The paper reviews and evaluates the published evidence produced by the archeologists of Boncuklu Tarla regarding the dating of the site. It also addresses a number of issues associated with carbon-14 dating of Neolithic sites’ structures, including Gobekli Tepe and Nevali Cori in addition to Boncuklu Tarla. Furthermore, the paper recommends a rigorous, streamlined, and transparent set of procedures to be established before an archeological team proposes dates (and especially carbon-14 based dates) on the structures of Neolithic monuments and their sites.
ResearchGate has not been able to resolve any references for this publication.