Content uploaded by Joseph Davidovits
Author content
All content in this area was uploaded by Joseph Davidovits on Dec 17, 2020
Content may be subject to copyright.
J. Davidovits and F. Davidovits / Geopolymer and Archaeology (2020) 36-43
Ancient geopolymers in South-American Monuments, Part IV(*):
use of natural andesite volcanic sand (not crushed).
Joseph Davidovits and Frédéric Davidovits
Geopolymer Institute, 02100 Saint-Quentin, France.
e-mails: joseph@geopolymer.org, frederic@geopolymer.org.
ABSTRACT
The studies carried out in 2017-2018 on the monumental stones constituting the Pumapunku site in Bolivia (South America) provided
evidence that the stones are ancient artificial geopolymers (Parts I to III). To make geopolymer andesite stone, around AD 600 to AD 700,
the builders could have transported an andesite stony material having the consistence of sand from the Cerro Khapia volcano site, and
added an organo-mineral geopolymer binder manufactured with local biomass ingredients. They did not use the many quadrangular
volcanic blocks, the famous "piedras cansadas”, the tired stones, which are still lying on both sides of the lake Titicaca. The present paper
describes how the builders of Pumapunku / Tiwanaku exploited a natural volcanic andesite sand from the volcano Cerro Khapia,
transported and stored it at the shore village of Iwawe, Stratum (V) in the excavation by Isbell & Burkholder, (2002). For the making of
their andesite geopolymer monuments, they did not need to crush andesite rock. This andesite sand is similar to one of the pozzolana
sands found in the best ancient Roman mortars and coined in Latin “carbunculus”, 2000 years ago.
Keywords: ancient geopolymer; South-American monuments; andesite sand, carbunculus, gas-pipes, piedras cansadas.
© 2020 Institut Géopolymère. All rights reserved.
(*) Part I: Davidovits et al. (2019a);
Part II: Davidovits et al. (2019b);
Part III: Davidovits et al. (2019c).
1. Introduction
Tiahuanaco, (Tiwanaku), in Bolivia, is a village known
throughout the world for its mysterious monolithic Gate of
the Sun (Puerta del Sol) made out of volcanic stone,
andesite. It comprises an earthen pyramid and is located
south-east of Lake Titicaca at 3820 m above sea level. It
belongs to the civilizations of the pre-Columbian Americas.
Archaeologists name this site “Tiwanaku” and consider that
it was built well before the Incas, around 600 to AD 700,
1400 years ago. The site of Pumapunku is right next door
with the ruins of an enigmatic pyramidal temple built at the
same time (see Figure 1).
______________________
DOI registered at Research Gate as Preprint:
doi.org/10.13140/RG.2.2.10021.93929/2
Received 10 November 2020, Accepted 15 December 2020,
Because it is not restored and developed for touristic
activity, Pumapunku (also written Puma Punku) is less
known to the general public. However, there are two
architectural curiosities there: four giant red sandstone
terraces weighing between 130 and 180 tons (Figure 1) and
small blocks of andesite, an extremely hard volcanic stone,
whose complex shapes (the H-blocks) and millimetric
precision are incompatible with the technology of the time
(Figure 2).
And for good reason, since archaeology tells us that the
Tiwanakans had only stone tools and no metal hard enough
to carve the rock. But they would have carved the gigantic
blocks of red sandstone (these ancient blocks are the largest
of all the American continent!) and they were able to carry
these hundreds of tons to the site, then to adjust them
precisely. Also, they would have been able to carve the “H”
blocks made of volcanic andesite, an extremely difficult-to-
carve stone with an incredible finish perfectly flat faces at
exact 90° interior and exterior right angles!
36
Geopolymer reoloy
AJournalonGeopolymerScienceApplieoArcaeoloy
available online a geopolymerorglrrygp
J. Davidovits and F. Davidovits / Geopolymer and Archaeology (2020) 36-43
Figure 1: !"#$%&'()*+,-.&'./)0&'1231-."&4+$%&5+4-.-6#&7Gate of the Sun8&9&:#(-3#.6#;&
Figure 2:&$%)&<-("#0&=>=&?1",60@&A&()$)*&%+B%&-./)0+$)&0$".)C&D"%0&%-*/.)00&EFG&7GHI#-*$J8C&/).0+$K&/HL;MN&6B918;&O+B%$@&3*),+0)&
*)3*"/#,2".&?K&!$P?1)&-./&Q%1)&7ANRL8;&
Figure 3:&D-3&0%"4+.B&$%)&?"*/)*&7+.&*)/8&?)$4)).&:)*#&-./&S"1+T+-C&:#(-3#.6#&9&5+4-.-6#&-./&$%)&T"1,-."&U)**"&V%-3+-C&$%)&0(-11&1-6)&
>#+W-K(-*,-C&3-*$&"<&5+2,-,-&X-6);&S"Y"(&*+B%$C&$%)&0+$)&"<&V-11-(-*6-;&
37
J. Davidovits and F. Davidovits / Geopolymer and Archaeology (2020) 36-43
Archaeologists cannot give any rational explanations on
how this was possible. Therefore, for the general public, the
assumptions generally advanced to explain these wonders
are the achievement by a lost ancient super civilization or
by aliens’ involvement.
To make geopolymer andesite concrete, the builders could
have transported non-consolidated volcanic tuff, which is an
andesite stony material having the consistence of sand from
the Cerro Khapia volcano site, and added an organo-mineral
geopolymer binder manufactured with local ingredients
(Davidovits et al, 2019b). However, we could not determine
the source of this volcanic sand.
The present paper describes how the builders of Pumapunku
/ Tiwanaku found and exploited a natural volcanic andesite
sand.
2. Results: we found the volcanic andesite
sand, it is not crushed stone.
We knew that for the andesite stones, tradition indicated a
geological origin in the volcanic region of Cerro Khapia,
Peru, on the other side of the Laguna Huiñaymarca, part of
the Titicaca Lake (Figure 3). Several archaeologists had
detected in the 19th century at this place probable sources
for the volcanic stones of Tiwanaku / Pumapunku. Stübel
and Uhle (1892) had published their comparative analyses
of samples taken, on the one hand, from the Puerta del Sol
and, on the other hand, from andesite outcrops located at the
foot of the Cerro Khapia volcano. There was a strong
concordance between the stone of the monument and the
geological rock.
Andesite boulders lying on the lake shores: the
“Piedras cansadas”.
But we had to solve a problem that seemed much more
worrisome because it contradicted what we wanted to
demonstrate. Despite our scientific discovery, how could we
believe in the artificial nature of the andesite rock of
Tiwanaku / Pumapunku when there existed everywhere on
both sides of the lake shores, many quadrangular volcanic
blocks, the famous "piedras cansadas”, the tired stones? All
the travelers and archaeologists of the 19th century already
describe them. They try to explain how these blocks of
volcanic rock, weighing from 5 to 10 tons, could have been
transported from the slopes of Cerro Khapia to the shores of
the lake. Then, how they were hoisted on rafts built with
totora reeds, next unloaded on the other bank and finally
transported overland to Tiwanaku / Pumapunku. In Figure
4a, these andesite blocks can be seen on the Peruvian shores
near Kanamarca. Figure 4b shows other blocks located on
the Bolivian shores, near Iwawe.
However, we had difficulty understanding why the builders
of Pumapunku had been forced to transport 10-ton blocks
and then cut them into smaller pieces to produce their
famous "H" sculptures. It would have been much easier to
place these small 300-700 kg blocks on small sleds and
small rafts, rather than undertaking this titanic work. It was
not logical. There must have been another explanation. This
is what we will establish now.
Figure 4: piedras cansadas, volcanic andesite boulders; a) top,
Peruvian shore at Kanamarca; b) bottom, Bolivian shore at Iwawe.
Credit: Ricardo Bardales Vassi (2013) .
Following the discovery of andesite "piedras cansadas" at
Iwawe on the Bolivian side of the lake, Ponce Sangines
(1968) and his Bolivian and American archaeological
colleagues Isbell & Burkholder, (2002), Janusek (2008),
Vranich (2005), Isbell (2013), Protzen (2013) had imagined
a scenario involving the crossing of the lake by boat from
Kanamarca. The village of Iwawe was the port where these
blocks were unloaded from the rafts and prepared for
transport to Tiwanaku / Pumapunku. See the map of Figure
3. For an unknown reason, they were not transported further
to Tiwanaku/Pumapunku.
This scenario received the support of North American
anthropologist William Isbell, a specialist in Andean
ceram ics who undert ook exc avations at Iwawe i n
2000-2002. Iwawe is a mound not more than 3 meters high.
It consists of 9 archaeological layers numbered (I) to (IX) as
shown in Figure 5. In the first report published by Isbell &
Burkholder, (2002, page 212) one reads: (…) Stratum V is
15 cm deep and it is frequently broken by big pit excavated
through it. (…) Stratum V demand interrogation. It is
38
J. Davidovits and F. Davidovits / Geopolymer and Archaeology (2020) 36-43
virtually sterile and its distinctive material does not seem to
be midden accumulated through domestic activities . At the
suggestion of James B. Richardson we submitted samples of
Stratum V soil to volcanologist Richard Naslund (personal
communication January 1994) of Binghamton University
Department of Geology whose thin section microscopic
examination conclusively revealed volcanic pumice.
Unfortunately, though, the only soil sample available to us
in the USA came from a flotation heavy fraction.
Consequently, the relative sizes of particles in the original
matrix could not be determined. Shanaka de Silva (de Silva
and Francis 1991: 138-155 personal communication), who
will collaborate with the Iwawi research team in the future,
informs us that Stratum V could be volcanic ash rained
down on the Altiplano following an eruption in the
prehistoric past. In that case the most likely source is the
volcano known as Cerro Quemado, 250 km to the southwest
of Iwawi. (…) However, volcanic ash is very durable and it
can be blown by the wind for long distance. So the Iwawi
ash may originate in some other primary tephra.
Figure 5: Stratigaphy of the Iwawe mound revealed in the
archaeological excavations; adapted from Isbell & Burkholder
(2002). Stratum V (andesite sand) is in gray and charcoal in black.
This analysis is confusing and not entirely accurate, because
the American geologist only had the finest fraction at his
disposal. This is why he believed that volcanic ash was
present. However, he stated that this Stratum (V) is not a
domestic waste (a pile of garbage) but a sand of volcanic
origin. How did it get there?
Andesite volcanic sand.
After completing a new study on the site, Isbell wrote in
2013 on the same subject: Isbell (2013, page 175): “(…)
Iwawe first gained fame as Tiwanaku’s port, where great
blocks of andesite stone from across Lake Titicaca were
landed before being hauled inland to the city. It is also an
ancient mound of residential debris about 2.5 m deep on the
shore of Lake Titicaca. Excavations yielded nine strata, the
lowest sterile soil that appears to have been shaped into
raised fields for cultivation before the site was occupied.
Stratum V divides the sequence in two. It is composed of
andesite sand, probably from intensive work shaping
imported stone blocks. Significantly, this grit layer probably
represents a moment when construction in andesite was
intensive, a major construction time at the capital.”
The archaeological excavations had discovered that the
layer Stratum (V) was made up of sand, andesite volcanic
sand. Isbell concluded that this layer was the result of
intensive sculpturing of the blocks that had arrived from the
Cerro Khapia volcano. However, carving of stones produce
rather shards, more or less large or small pieces but no sand.
William Isbell does not mention these andesite splinters in
the layer Stratum (V) which for him is made of sand.
However, he can distinguish between andesite sand and
crushed andesite rock. Indeed, in the stratigraphy of Figure
5 there is a thin layer of crushed andesite between Stratum
(VI) and Stratum (VII). It is labeled: "discontinuous stratum
of crushed rock". He further describes the Iwawe excavation
as follows: “ In the deep Stratum VI through VIII, [older
than Stratum V], the common cooking pot is accompanied
by a wide range of more or less shallow, hemispherical
bowls, some with incised rims. These were probably serving
bowls, for eating and drinking. But they disappear with the
appearance of the sandy debitage that constitutes stratum V
and are completely replaced in higher strata by the two
most common Tiwanaku shapes, the kero and the tazón.”
Isbell confirms with the English term "sandy debitage" that
the blocks were sawed (how?) and produced a waste
material in the form of sand. The Tiwanaku era does not
have any saw capable of cutting andesite volcanic rock
blocks. Therefore, the Stratum (V) does not contain any
splinters or pieces of stone, just sand (in geology, it is called
a volcanic sand).
According to Isbell and Burkholder (2002), the stratigraphy
and the position of Stratum (V), the probable date of the
construction period in Pumapunku / Tiwanaku, would be
around AD 600. This discovery of volcanic sand was
misinterpreted by all archaeologists and anthropologists at
the time. Indeed, it is necessary to possess serious
knowledge in geology associated with geopolymers in order
to understand the implication of this discovery. Apparently,
the geologists/volcanologists who were consulted by W.
Isbell did not mention that andesite sand was frequently
found in association with hard volcanic rock. Such sand can
be detected through the phenomenon referred to as "gas
pipes". We shall thus try to explain it.
Carbunculus: the volcanic sand in ancient Roman
mortar.
The reader should know that, for us, this type of material is
not a mystery or an unfamiliar part of our archaeological
research. We have known it for more than 30 years. As a
39
0
100
200
300
cm
top soil
Stratum I
Stratum II
Stratum III
Stratum IV
Stratum V
Stratum VI
Stratum VIII
Stratum IX
Stratum VII
andsit sand
2000
AD
600
AD
200
B
J. Davidovits and F. Davidovits / Geopolymer and Archaeology (2020) 36-43
matter of fact, it was one of the main constituents (the so-
called pozzolana, or “harena fossicia”) of the best mortars
developed by the ancient Roman civilization. It is found in
the mortars of the large monuments erected in Rome (Italy),
like the Coliseum, that date back to the time of the Roman
emperors Julius Caesar, Augustus, Nero or Constantine, 2nd
century BC to 5th century AD. According to the Roman
architect Vitruvius, the ancient Latin name for this volcanic
sand was “carbunculus" (Davidovits F., 2020).
In 1994, we started a research program funded by the
European Union, called GEOCISTEM (1997), whose
partners were the Geopolymer Institute, the B.R.G.M. -
Bureau de Recherches Géologiques et Minières in France,
the Geology Faculties of the Universities of Barcelona
(Spain) and Cagliari (Italy), University of Caen, France (see
the caption of Figure 6). The objective of our research was
the creation of new ecological geopolymer cements, which
would use geological materials like volcanic tuff,
Davidovits et al. (1999) and Gimeno et al. (2003). To
achieve this goal, we had chosen to use the same deposits as
the ones commonly exploited in Roman antiquity 2000
ye a rs ag o, in I taly a nd Sp ain. This r equ i red a
multidisciplinary approach, bringing together specialists in
geopolymers, geologists studying volcanic rocks, and a
researcher familiar with the history of ancient Roman
building material techniques. Here is what one of the
authors, FD, wrote in 2007 in his doctoral thesis entitled:
Geology and construction in the De architectura of
Vitruvius. The excerpt is available at Davidovits F. (2020):
(...) To determine the nature of the "materia" carbunculus
(volcanic sand), we have to use a discovery that was
unexpected - not from a geological point of view, but rather
from the humanities - during the GEOCISTEM program, in
which we participated for the sampling of Roman mortars.
During a meeting in Cagliari (Sardinia, Italy) in September
1996, which involved the five geologists of the partner
universities in this project, we were able to visit the volcanic
stone quarry of Paringianu, exploited for the extraction of
hewn stone (Sheet and Map 3). The "Paringianu tuff", as it
is locally called, is a volcanic rock, resulting from a
pyroclastic flow.
The local volcanic context consists of ignimbrite and
rhyolite. The rock is very indurated, i.e., it is solid. It is
composed of plagioclase, potassium feldspar, pyroxene, a
vitreous matrix and montmorillonite. (...) During the visit,
which was instructive on the nature of the Carbunculus
"materia", we saw a very striking curiosity for these
specialists of volcanic materials: while some tens of meters
away, very hard cut stone was being extracted, the
geologists showed us an unexploited area of the quarry. And
with good reason: the tuff had the same mineralogical and
chemical composition as the very hard rock and it contained
crystals of identical dimensions, but it disintegrated into
sand when we ran our fingernail or finger over it. They
explained to us that during the cooling of the volcanic
stratum, which has to be done slowly for the rock to harden,
a sudden degassing in this tuff layer left columns through
which the gases escaped: the stone did not have time to
properly solidify as it cooled down. This demonstrated the
degree of cohesion between the two types of stone: one
cooled slowly and acquired some consistency, while the
degassing turned the other into a soft, not very hard rock.
According to the geologists who were with us on the site, the
difference in hardness between two rocks of similar
composition is a common phenomenon.
Figure 6: The GEOCISTEM team visited the site of Paringianu,
Sardinia, Italy, on September 27, 1996. From the left: Frédéric
Davidovits (Caen Univ., France), Domingo Gimeno (Barcelona
Univ., Spain), Philippe Rocher (BRGM, France), Carlo Marini
(Cagliari Univ., Italy), Athos Rinaldi (Laviosa, Italy), Joseph
Davidovits (Géopolymère, France), Sandro Tocco (Cagliari Univ.,
Italy), Michel Laval (BRGM, France), Luigi Buzzi (Cementi
Buzzi, Italy), Jean Claude Toussaint (E.U Commission, Brussels,
Belgium), (GEOCISTEM, Final Technical Report, 1997).
40
J. Davidovits and F. Davidovits / Geopolymer and Archaeology (2020) 36-43
In Figure 6, the members of the European project
GEOCISTEM are in front of the quarry of hard volcanic
rock, and then they will see the gas-pipes and the
Carbunculus reproduced in Figure 7. Frédéric continues:
"Looking at the degassing columns, one could see that they
were vertical and that they created a small system of veins
that vertically crossed the entire tufa layer from the bottom
to the surface. This was approximately one man's height,
and these ducts were a few centimeters wide. This
phenomenon is known in geology under the name of "gas
pipe" (...). “
Figure 7: The gas pipes in the Carbunculus layer, volcanic sand
at Paringianu, Sardinia, Italy. Reconstitution after Davidovits F.,
(2007-2020).
The occurrence of the Stratum (V) in the archaeological
stratigraphy leads us to conclude that the ground of the
Iwawe village was covered with heaps or mounds of natural
andesite volcanic sand. This sand would have been
extracted in one or more places in the Cerro Khapia
volcano, transported to the shores of Kanamarka. Then it
would have crossed the lake on rafts and been stockpiled in
the port of Iwawe. We know that the level of the lake could
vary with the seasons. Presumably, this activity took place
during the rainy season, when the lake level was at its
highest. The rafts could easily approach Iwawe's mainland.
The andesite sand was poured on the ground, forming a
layer several tens of centimeters thick. Isbell indicates that
this layer is discontinuous and that it is "frequently broken
by big pits excavated through it". It is thus a storage area in
which workers came to dig and transport the andesite sand
to another place (Tiwanaku / Pumapunku).
The Carbunculus of the ancient Romans corresponds
exactly to the andesite volcanic sand of the Iwawe Stratum
(V). Geologists had told us that, in volcanic rocks, the
phenomenon of juxtaposition of a very hard stone and a
sand was common. But in Iwawe we actually have a
volcanic sand coming from the Cerro Khapia volcano. It is
not crushed andesite stone. Therefore, logically, in one or
more places of this volcano, we should find areas, where
there is the juxtaposition of a hard andesite rock and a sand,
both having the same mineralogical and chemical nature.
But where to do the exploration?
3. Discussion: who transported the piedras
cansadas and when?
We must now find out why, how, when and by whom these
piedras cansadas had been abandoned on the path leading
from Cerro Khapia to Tiwanaku / Pumapunku. For that, we
can consult archaeological data and publications describing
the use of andesite volcanic rocks in Tiwanaku and
Pumapunku. There would be two periods during which
these blocks could have been extracted, transported and
abandoned on the way to Tiwanaku. The first is around
800-1000 AD., 200 to 300 years after the storage of andesite
sand in Stratum (V), either at the end or just after the
"classical" period of Tiwanaku. The second is during the
Inca Empire, around 1400 AD, 700 years after the
construction of Pumapunku.
Although it is difficult to prove it for lack of written texts,
anthropologists think that after AD 800., there was a brutal
change in the governance of Tiwanaku during this epoch.
But the period of the Incas seems to us more appropriate,
because it relies on historical and archaeological facts.
We know now that the Incas had undertaken many
restoration works, using andesite blocks, on the site of
Pumapunku. According to the American archaeologist /
Alexei Vranich (2013) these were significant restoration
works. Here is what he wrote: “ (…) To make this history
tangible, the Inca invested a huge amount of energy in
constructions across the entire Titicaca Basin, modifying
and enhancing important ritual places. Tiwanaku became
the terminal point on an important ritual pilgrimage route
that began in the Inca capital of Cuzco. The best-preserved
structure at Tiwanaku, the Pumapunku, was renovated and
refurbished, and a royal and religious settlement, complete
with a palace, bath (essential for Inca rituals), kitchen, and
associated plaza, was built against this temple. The Incas
even intervened in a more aggressive fashion on the
platform, emptying out fill between the retaining walls to
form three chambers overlooking the plaza. Arriving in
Tiwanaku, visitors and pilgrims would be hosted in the
refurbished and Incanized temple, where the mythic history
would be told through ritual performance. (…).
The Incas used a lot of andesite materials to restore
Pumapunku and Tiwanaku. For that, they went to seek their
blocks in the geological sites which naturally possess this
type of quadrangular blocks in the Cerro Khapia crater (see
Figure 8b). But we know that the Incas did not drag their
stones on the natural surface of the land, but prepared
carefully constructed roads. A similar road remains visible
41
J. Davidovits and F. Davidovits / Geopolymer and Archaeology (2020) 36-43
at Kanamarca in Figure 8a connecting the Cerro Khapia
"quarry" to the shores of the lake (see the location of the
village on the map of Figure 2),.
Figure 8: (a) top: marked by the arrow the Inca Trail by which
the andesite blocks were brought down from the crater area of the
volcano Cerro Khapia, (b) bottom: the quarry. Credit: Ricardo
Bardales Vassi (2013).
4. Conclusion.
We now know what geological material was used in the
manufacture of the artificial andesite geopolymer blocks in
Pumapunku / Tiwanaku, Bolivia, around AD 600. It is not
crushed andesite stone but natural volcanic andesite sand,
perfectly adapted to this use. It has the same mineralogical
and chemical constitution as the andesite blocks of the
Cerro Khapia volcano. This sand would have been extracted
in one or more places in the Cerro Khapia volcano,
transported to the shores of Kanamarka in Peru. Then it
would have crossed the lake on rafts and been stockpiled in
the port of Iwawe, Bolivia. To produce the andesite
geopolymer blocks, the workers of Pumapunku / Tiwanaku
added a type of organo-mineral binder manufactured with
local biomass (carboxylic acids extracted from maize and
plants), guano and reactive alumino-silicate minerals
(Davidovits et al, 2019b, 2019c).
In the same way, we had already found the geological
material used in the manufacture of Pumapunku megalithic
terraces. It is a red sandstone that has been disintegrated by
climatic erosion and easily transformed into sandstone sand
(Davidovits et al, 2019a, 2019c). It is associated with
Kallamarka / Callamarca (Bolivia), an historical village that
is part of the UNESCO World Heritage. It is remarkable!
The next step should involve collecting andesite sand
samples from the Stratum (V) of Iwawe and organizing a
geological expedition to the Cerro Khapia volcano. We hope
to discover the exact source of the materials used to build
the monuments of Pumapunku / Tiwanaku.
The construction workers of Pumapunku / Tiwanaku could
easily transport these geological materials over great
distances, in baskets, by rafts, in llama caravans. In short, it
was normal work, on the scale of Homo sapiens and his
ingenuity.
References
Bardales Vassi R., (2013), Wiracocha, El reflejo de un código
sagrado, Capítulo I: Tiahuanacu y las culturas
en el Altiplano; Puno, Perú.
Davidovits J., Huaman L., Davidovits R., (2019), Ancient
geopolymer in South American monuments. SEM and
petrographic evidence, Material Letters 235, 120-124. DOI:
doi.org/10.1016/j.matlet.2018.10.033.
Davidovits J., Huaman L., Davidovits R., (2019), Ancient organo-
mineral geopolymer in South American Monuments: organic
matter in andesite stone. SEM and petrographic evidence,
Ceramics International, 45, 7385-7389. DOI: doi.org/10.1016/
j.ceramint.2019.01.024
Davidovits J., Huaman L., Davidovits R., (2019), Tiahuanaco
monuments (Tiwanaku / Pumapunku) in Bolivia are made of
geopolymer artificial stones created 1400 years ago,
Archaeological Paper #K-eng, Geopolymer Institute Library,
www.geopolymer.org. DOI: doi.org/10.13140/
RG.2.2.31223.16800.
Davidovits J., Buzzi L., Rocher P., Gimeno D., Marini C. and
Tocco S., (1999), Geopolymeric Cement Based on Low Cost
Geological Materials. Results from the European Research
Project GEOCISTEM, Geopolymer '99 Proceedings, pp.
83-96, 2nd International Conference GEOPOLYMERE ’99,
Institut Géopolymère / Geopolymer Institute, Saint-Quentin,
France.︎︎︎︎ Available at Research Gate https://
www.researchgate.net/publication/284757919
Davidovits F., (2020), Carbunculus, extrait de la thèse Géologie et
construction dans le De architectura de Vitruve (2007),
Geopolymer and Archaeology, 10-36. Geopolymer Institute,
Saint-Quentin, France. DOI: doi.org/10.13140/
RG.2.2.26618.72644.
GEOCISTEM, Davidovits J., Rocher P., Gimeno D., Marini C.
Rinaldi A., Tocco S. and Davidovits F., (1997), Cost Effective
Geopolymeric Cements for Innocuous Stabilisation of Toxic
Elements (GEOCISTEM), Final Technical Report, available
online at: http://infoterre.brgm.fr/rapports/RR-39616-FR.pdf.
Gimeno D., Davidovits J., Marini C., Rocher P., and Tocco S.,
Cara S., Diaz N., Segura C. and Sistu G., (2003), Desarrollo de
un cemento de base silicatada a partir de rocas volcánicas
vítreas alcalinas: interpretación de los resultados
preindustriales basada en la composición químico-
mineralógica de los precursores geológicos, Bol. Soc. Esp.
Cerám. Vidrio 42 [2] 69-78. [Development of silicate-based
cement from glassy alkaline volcanic rocks: interpretation of
42
J. Davidovits and F. Davidovits / Geopolymer and Archaeology (2020) 36-43
preliminary data related to chemical-mineralogical
composition of geologic raw materials; project
GEOCISTEM].Available at Research Gate: https://
www.researchgate.net/publication/28301118
Isbell W.H. and Burkholder JoE, (2002), Iwawi and Tiwanaku, in
Andean Archaeology I, Variations in Sociopolitical
Organization, W.H. Isbell and H. Silverman eds., Springer
Science+Business Media, LLC, New York, USA.
Isbell W. H., (2013), Nature of an Andean City: Tiwanaku and the
Production of Spectacles, in Visions of Tiwanaku, A. Vranish
and C. Stanish eds., Monograph 78, Chap. 10, 167-196, Cotsen
Institute of Archaeology Press, UCLA, Los Angeles, USA.
Janusek J.W., (2008), Ancient Tiwanaku, pp. 173-175, Cambridge
University Press, New York.
Ponce Sangines C., (1968), Perspectiva arqueológica, in Las
Andesitas de Tiwanaku, Mille N. and Ponce Sangines C. eds.,
Publicación N° 18, Academia Nacional de Ciencias de Bolivia,
La Paz.
Protzen J.P., and Nair S., (2013), The Stones of Tiahuanaco, A
Study of Architecture and Construction, Monograph 75,
Chapter 6, Construction: Transporting, pp. 180-181, Cotsen
Institute of Archaeology Press, UCLA, Los Angeles, USA.
Stübel A. and Uhle M., Die Ruinenstäette Von Tiahuanaco, Verlag
von Karl W. Hiersemann, Leipzig, 1892. http://digi.ub.uni-
heidelberg.de/digit/stuebel_uhle1892/0004, (accessed 10
August 2018).
Vranich A., Paul Harmon P. and Chris Knutson C., (2005), Reed
Boats and Experimental Archaeology on Lake Titicaca,
Expedition, 47(2), 20–27. University of Pennsylvania Museum
of Archaeology and Anthropology, Philadelphia, USA.
Vranich A., (2013), Visions of Tiwanaku, in Visions of Tiwanaku,
A. Vranich and C. Stanish eds., Monograph 78, Chap. 1, 1-9,
Cotsen Institute of Archaeology Press, UCLA, Los Angeles,
USA.
43
