Puma Punku — Precision Megalithic Stonework at the Gateway of the Puma

Puma Punku — The Anomaly

The anomaly in one sentence

At 3,850 m in the Bolivian Andes, Puma Punku’s andesite blocks — some weighing an estimated 85 tonnes — carry internal 90-degree corners, 3–6 mm constant-width grooves, and H-shaped sockets cut for bronze clamps, worked to a flatness of 1–2 mm across several metres, by a culture with no attested iron tools.

Case file	Value
Case ID	AA-EVD-007
Location	Tiwanaku, La Paz Department, Bolivia
Proposed date	c. 500–1000 CE (Tiwanaku IV–V)
Case status	Partially resolved
Sitchin connection	Direct — discussed in The Lost Realms, When Time Began, and The End of Days
Evidence items	8
Primary sources	4
Last reviewed	2026-07-20
Independent review	No
Altitude	3,850 m
Material	Andesite (quarried ~90 km away), red sandstone

Why this case matters

Puma Punku is one of the most cited sites in the alternative-archaeology literature, and the underlying stonework is genuinely remarkable regardless of who cut it. The site is a collapsed platform mound at the southwestern edge of the larger Tiwanaku complex, built from precisely worked blocks of andesite (a hard volcanic rock, Mohs 6–7) and red sandstone (Mohs 3–4). The features that draw attention — internal right angles cut inside the stone rather than only on external faces, grooves of constant width running the length of a block, and H-shaped sockets carved to receive metal clamps — are not the kind of thing casual stone-dressing produces. They look deliberate, repeatable, and machined.

What the conventional model explains well is how: experimental archaeology published in 2013 demonstrated that hammerstones, abrasive sand, and patience can reproduce every one of these features without metal cutting tools. What it explains less completely is why here specifically — the H-socket clamp technique is, on current evidence, essentially unique to Puma Punku and one other Andean site, and the full-scale logistics of moving 85–130-tonne blocks 90 km across Lake Titicaca and the high-altitude Altiplano have never been experimentally replicated at scale.

Zecharia Sitchin discussed Tiwanaku and Puma Punku at length across three books — The Lost Realms (1990), When Time Began (1993), and The End of Days (2007) — relying heavily on the field documentation of Arthur Posnansky (1873–1946), the pioneering but chronologically discredited Austrian-Bolivian archaeologist. This is a Direct connection: Sitchin names the site, quotes Posnansky’s descriptions of the bronze clamps, and builds an explicit argument that the metallurgy at Puma Punku was part of an Anunnaki construction and departure program in the Andes.

The question

Does the precision stonework at Puma Punku — internal right angles, constant-width grooves, H-shaped sockets, and the bronze clamp technology once housed in them — require tools or techniques (iron, rotary drilling, or an imported metallurgical tradition) that the Tiwanaku culture is not otherwise known to have possessed, or can documented Andean stoneworking methods (hammerstones, abrasive sand, and copper/bronze tools) fully account for what survives?

Executive finding

The platform mound now called Puma Punku was constructed during the Tiwanaku IV and V periods, roughly 500–1000 CE, at the southwestern edge of the Tiwanaku archaeological complex on the Bolivian Altiplano. It was described by Spanish chroniclers in the sixteenth century and was extensively (if controversially) documented by Arthur Posnansky between 1904 and his death in 1946; his four-volume Tiahuanacu: The Cradle of American Man (1945) remains the foundational photographic and drawn record of the site, much of which has since been looted, quarried for colonial construction, or lost.

What is physically established is that the platform’s blocks — quarried from andesite at Cerro Khapia roughly 90 km away across Lake Titicaca, and from sandstone at the closer Kimsachata range — were cut with internal right angles, constant-width grooves, and H-shaped and T-shaped sockets designed to receive bronze clamps. The largest surviving andesite block is estimated at 85 tonnes; the largest sandstone block, at 130 tonnes. Bronze artifacts (chisels, bars, clamps, axes, knives) have been recovered at Tiwanaku and on the nearby Islands of the Sun and Moon.

The strongest conventional explanation, published by Jean-Pierre Protzen and Stella Nair in 2013, is that all of these features are reproducible using hammerstones for rough shaping and abrasive sand-and-water slurry for the fine flat surfaces and grooves — a lapidary technique independently attested at other ancient sites worldwide, including in Egyptian stoneworking and in Protzen’s earlier study of Inca construction at Ollantaytambo. This model does not require iron tools, rotary drills, or any technology outside the documented Andean toolkit. What it leaves genuinely open is why the H-socket clamp technique appears at only two known Andean sites, and why no full-scale experimental transport of an 85-tonne-class block across 90 km of lake and altiplano at 3,850 m has ever been attempted.

Sitchin’s reading treats the same bronze evidence differently. He accepted Posnansky’s now-rejected dating of the Kalasasaya platform to 12,000–17,000 years before present, and layered onto it a narrative in which Anu and Antu, rulers of Nibiru, visited the Titicaca gold region around 3800–4000 BCE and used Puma Punku’s harbor facilities; in a separate later argument, in The End of Days, he proposed that Puma Punku also served as a departure facility for the Enlilite Anunnaki leadership after the destruction of the Sinai spaceport in 2024 BCE. He drew a genuine cross-cultural parallel between the Puma Punku bronze clamps and the “copper mixed with tin” bronze that the Sumerian ruler Gudea used to bind the stones of the Eninnu temple at Lagash. The metallurgical evidence he cites is real; the chronology and the extraterrestrial-visitation narrative built on top of it are not supported by any independent dating method published since Posnansky’s death.

Current reading

The anomaly: Confirmed but diminishing — genuinely puzzling before 2013, substantially closed since. Best conventional model: Strong — Protzen & Nair’s experimental replication accounts for the stoneworking; the H-socket uniqueness and full-scale transport logistics remain open but modest gaps. What remains open: Why the bronze-clamp joinery technique is attested at only two Andean sites, and whether full-scale block transport at this altitude has ever been genuinely tested. Sitchin connection: Direct — discussed across three books, built on Posnansky’s field documentation. Reasonable conclusion: The stonework is an exceptional achievement of Tiwanaku craftsmanship, not evidence of imported or extraterrestrial technology — but the site’s near-unique clamp technique deserves more fieldwork than it has received.

Evidence at a glance

Measurement	Verified value	Why it matters	Source
Largest andesite block	~85 tonnes	Sets the minimum quarrying, lifting, and transport capability required	Protzen & Nair 2013
Largest sandstone block	~130 tonnes	Heaviest single worked stone documented at the site	Kolata 1993
Andesite quarry distance	~90 km (Cerro Khapia, across Lake Titicaca)	Rules out on-site quarrying; requires cross-lake or overland transport	Protzen & Nair 2013
Site altitude	3,850 m	Reduces human labour capacity (~60% of sea-level oxygen partial pressure)	UNESCO World Heritage nomination dossier
Surface flatness tolerance	1–2 mm across several metres	The core “machined” claim — independently reproduced by hand-abrasive lapping	Protzen & Nair 2013
Radiocarbon / ceramic date range	cal. AD 500–1000 (Tiwanaku IV–V)	Contradicts Posnansky’s and Sitchin’s much older chronology by roughly 10,000+ years	Kolata 1993; Janusek 2008

Confidence dashboard

Dimension	Score	What the score means
Anomaly confidence	2/5	The stonework is real and impressive, but the technological anomaly relative to known Andean capability has been substantially closed since 2013
Source quality	4/5	Protzen & Nair, Janusek, and Kolata are peer-reviewed/university-press academic sources; Posnansky’s primary documentation is methodologically dated and chronologically unreliable but photographically valuable
Conventional-model strength	4/5	Experimentally validated for the stoneworking; not yet validated for full-scale block transport
Unresolved-question significance	2/5	The H-socket uniqueness and transport logistics are real open questions, but neither threatens the conventional chronology
Sitchin-connection strength	4/5	Direct and extensive — three books, named site, quoted primary source (Posnansky)
Advanced-technology evidence	1/5	Bronze tools and clamps are confirmed; no positive evidence of iron, rotary drilling, or non-Andean metallurgy has ever been reported

Confidence note: Puma Punku’s reputation in alternative-archaeology media substantially exceeds its current archaeological anomaly. Before 2013 it was a legitimate open puzzle; Protzen and Nair’s experimental replication closed most of the “how.” The remaining open questions — the H-socket technique’s near-uniqueness, and the untested logistics of 90 km high-altitude transport — are genuine research gaps, not evidence of lost technology. The Sitchin connection is unusually well-documented for this archive precisely because the underlying metallurgical evidence he cites (bronze clamps, real artifacts) is not itself disputed; only his chronology and his interpretation of what that metallurgy implies are.

What is physically established

The site

Puma Punku is a terraced platform mound approximately 167 × 117 m at its base. Its complete original form is unknown: the site was thoroughly dismantled and its stone hauled off for colonial and post-colonial construction from the sixteenth century onward. What survives is a field of precisely cut blocks, many overturned or displaced, that once comprised part of a sophisticated stone structure. The Spanish chronicler Pedro Cieza de León, who travelled through the region in the 1540s, described the wider Tiwanaku complex as “ancient buildings, the most noteworthy to be found in all the Indies… large, well-carved stones.”

Physical observations

ID	Observation	Measured value	Method	Source	Confidence
OBS-01	Precision-cut internal 90° angles in andesite blocks	Corner cut inside stone, not just on external face	Direct observation; photogrammetry	Posnansky 1945; Protzen & Nair 2013	High
OBS-02	Constant-width grooves on block faces	3–6 mm wide, uniform depth	Direct measurement	Protzen & Nair 2013	High
OBS-03	H-shaped sockets cut into andesite blocks	~10–15 cm deep, straight-walled rectangular recess with perpendicular ends	Direct observation	Posnansky 1945	High
OBS-04	Andesite quarry source	Cerro Khapia, ~90 km away, across Lake Titicaca	Geological / trace-element sourcing	Protzen & Nair 2013	High
OBS-05	Largest surviving blocks	Andesite ~85 tonnes; sandstone ~130 tonnes	Volumetric estimate from surviving fragments	Protzen & Nair 2013; Kolata 1993	Medium — estimated, not directly weighed
OBS-06	Bronze artifacts at Tiwanaku and Titicaca islands	Clamps, chisels, bars, knives, axes	Excavation; museum documentation	Posnansky 1945; later excavations (Janusek 2008)	Medium — most original clamps lost to looting
OBS-07	Site altitude	3,850 m	Altimetry / GPS survey	UNESCO World Heritage nomination dossier	High

The stonework

• Internal 90-degree angles. Corners cut inside the stone — as in the H-sockets — rather than only on external faces. This is technically more demanding than simply squaring an external face, because the cutting tool or abrasive must turn a sharp corner in a confined space.
• Flatness. Cut surfaces are flat to within 1–2 mm across spans of several metres, a precision normally associated with machined surfaces. Protzen & Nair demonstrated that this is achievable by rubbing two blocks together with an abrasive sand-and-water slurry — a lapidary technique also attested in ancient Egyptian stoneworking — and requires no metal tools.
• Constant-width grooves. Straight channels, typically 3–6 mm wide and equally deep, running across block faces. Their uniform cross-section suggests a consistent guide tool (a straightedge, drawn repeatedly with abrasive sand) rather than freehand chiselling.
• H-shaped sockets. Precisely carved recesses, typically 10–15 cm deep, with straight walls and 90-degree internal corners, cut into the top, side, or face of blocks in the shape of a capital H. These are identical in function to the lead- or metal-filled clamp sockets used in Greek and Roman construction. At Puma Punku the sockets themselves are empty today, but Posnansky drew and photographed bronze clamps — often called “cramps” — that he believed had joined adjacent blocks at these sockets. The clamps were typically I-shaped, fitting the H-socket and locking adjacent blocks together.

In The Lost Realms, Sitchin quotes Posnansky directly: “He found at Tiahuanacu stone blocks bearing niches that served to hold bronze bolts, as at Puma Punku. At Puma Punku he had seen a piece of metal, undoubtedly bronze, that ‘with its dentiform prongs looked like a tackle or device for lifting weights.’” The presence of bronze technology at Tiwanaku is not disputed: bronze bars, levers, chisels, knives, and axes have been excavated at the site and on the Islands of the Sun and Moon in Lake Titicaca. What is disputed is whether bronze tools — which are softer than andesite and wear quickly against it — were the cutting tool, or only the joining material once the abrasive-lapping shaping was done. The evidence favours the latter: abrasive shaping, bronze joining.

Dating

Item	Proposed date	Method	Published uncertainty	Source
Tiwanaku IV construction phase (Puma Punku’s main platform)	cal. AD 500–800	Radiocarbon (calibrated), ceramic seriation	Typically ±50–100 years (1σ) for regional series	Kolata 1993; Janusek 2008
Tiwanaku V expansion phase	cal. AD 800–1000	Radiocarbon, ceramic seriation, stratigraphy	Typically ±50–100 years (1σ)	Janusek 2008
Posnansky’s Kalasasaya orientation date	12,000–17,000 BP (alternative reading: c. 4000 BCE)	Archaeoastronomical calculation from platform orientation and Earth’s obliquity	Disputed — the method’s underlying assumptions have been challenged	Posnansky 1945; endorsed in Sitchin, When Time Began, Ch. 9

Dating limitation

Radiocarbon dates come from organic material (textile, bone, charcoal) recovered in stratified deposits under and around the platform — they date the depositional context, not the moment a specific block was quarried or carved. Archaeoastronomical dating, as used by Posnansky, assumes both that the platform’s orientation was astronomically intentional at construction and that it has not shifted since, and it depends on a specific model of how Earth’s axial obliquity has changed over millennia — a model later archaeoastronomers have challenged. No dating method independent of Posnansky’s own calculation supports his 12,000–17,000 BP figure; every subsequent radiocarbon and ceramic study converges instead on 500–1000 CE.

Material and construction properties

• Andesite hardness: Mohs 6–7 — roughly twice the hardness of the red sandstone (Mohs 3–4) that comprises most of the wider Tiwanaku construction, which explains why the andesite blocks, not the sandstone, carry the most technically demanding features.
• Bronze composition: described by Sitchin, following Posnansky, as “copper mixed with tin” — a standard tin-bronze alloy consistent with bronze technology independently attested elsewhere in the pre-Columbian Andes, not a distinct or anomalous metallurgy.
• Quarrying and transport: the andesite quarry at Cerro Khapia sits on the far western shore of Lake Titicaca, roughly 90 km from Puma Punku. Blocks were most plausibly moved by raft or balsa craft across the lake, then by sledge or rollers across the Altiplano, at an elevation where oxygen partial pressure is roughly 60% of sea level. The logistics of moving an 85-tonne block under these conditions are genuinely demanding — a real engineering feat, even using only documented pre-Columbian methods.

Evidence inventory

ID	Evidence item	Type	Provenance	Supports	Creates tension for	Reliability
E-01	H-shaped and T-shaped bronze-clamp sockets cut into andesite blocks	Physical	Puma Punku platform	H1, H3	H2	High
E-02	Constant-width grooves (3–6 mm) and sub-2 mm surface flatness	Physical	Puma Punku platform	H1	H2	High
E-03	Bronze artifacts (clamps, chisels, bars, axes, knives)	Physical	Tiwanaku site and Titicaca islands	H1, H3	—	Medium — most original clamps lost to looting; documentation relies heavily on Posnansky
E-04	Andesite sourced ~90 km away at Cerro Khapia	Physical / geological	Cerro Khapia, across Lake Titicaca	H1	—	High
E-05	Radiocarbon and ceramic dating placing Tiwanaku IV–V at cal. AD 500–1000	Dating	Stratified excavation contexts	H1	H3	High
E-06	Posnansky’s archaeoastronomical dating of the Kalasasaya to 12,000–17,000 BP	Dating / textual	Posnansky 1945; endorsed in Sitchin, When Time Began	H3	H1	Low — contradicted by every independent post-1945 method
E-07	Protzen & Nair’s experimental replication of flat surfaces, grooves, and sockets with hammerstones and abrasive sand	Contextual / experimental	Published academic study, 2013	H1	H2, H3	High
E-08	Spanish chronicles describing Tiwanaku and Puma Punku (Cieza de León, 1549; Molina, c. 1573)	Textual	Colonial-era manuscripts, cited by Posnansky and Sitchin	—	—	Medium — secondary, colonial-era, written decades after the culture’s collapse

Provenance gaps

• Most bronze clamps were lost to post-conquest looting and metal recycling; Posnansky’s early-twentieth-century field drawings and photographs (E-03) are the primary surviving documentation and cannot be independently re-checked against the original objects.
• The site was thoroughly dismantled from the colonial period onward, so the platform’s complete original form is a reconstruction, not a direct observation.
• No full-scale experimental replication of 85–130-tonne andesite block transport across Lake Titicaca and the Altiplano at 3,850 m has been attempted; feasibility is modelled, not demonstrated.
• Posnansky’s chronology (E-06), while comprehensively rejected by post-1945 dating, remains the chronological foundation Sitchin’s Anunnaki-visit narrative depends on — the Sitchin-compatible reading’s date rests on a source that even later Andeanist archaeology no longer accepts.

The strongest conventional explanation

Proposed mechanism

The Tiwanaku were accomplished stoneworkers operating at the peak of a tradition traceable through earlier Andean construction. Protzen and Nair’s 2013 experimental study, The Stones of Tiahuanaco, provides the most thorough test of the stoneworking question to date:

1. Rough shaping: andesite blocks were roughed out with hammerstones — harder stone nodules used as percussive tools — to approximate final dimensions.
2. Fine flat surfaces: achieved by grinding two blocks together with an abrasive slurry of sand and water, the same lapidary technique used across the ancient world for shaping hard stone (including in Egyptian stoneworking).
3. Grooves: cut with a straightedge (wood or copper) and abrasive sand, drawn repeatedly across the surface; the uniformity of the groove is a function of the straightedge, not of the tool material.
4. H-sockets: carved with hammerstones and abrasives, working the internal corners from the outside in — time-consuming but technically straightforward, and consistent with the bronze clamps documented at the site.
5. Transport: the andesite was most plausibly moved by raft across Lake Titicaca, then by sledge or rollers with large-scale human labour across the Altiplano; llama caravans supplemented lighter loads.

What the model explains well

• [E-01] H/T-shaped sockets: fully explained as clamp receptacles carved with hammerstones and abrasives, consistent with the bronze artifacts (E-03) recovered at the site.
• [E-02] Sub-2 mm flatness and constant-width grooves: experimentally reproduced by Protzen & Nair using only stone-age tools and abrasive slurry.
• [E-04] Long-distance andesite sourcing: consistent with known Tiwanaku-period long-distance material procurement networks documented elsewhere at the site (obsidian, exotic ceramics).
• [E-05] Chronology: the model requires no date earlier than the well-attested Tiwanaku IV–V period.

Independent support

Comparable abrasive-lapping precision, achieved without metal cutting tools, is independently documented at other ancient sites: in Old Kingdom Egyptian hard-stone vessel manufacture, and — closer to home — in Jean-Pierre Protzen’s earlier study of Inca ashlar masonry at Ollantaytambo, which achieved comparable joint precision using stone hammers and abrasives centuries later in the same broader Andean tradition (Protzen 1993).

Conventional-model assessment

Best current model

Explanatory reach: High Independent support: Strong — comparable hand-abrasive precision independently documented in both Egyptian and later Inca-period Andean stonework Confirmed predictions: Flat surfaces achievable via abrasive slurry (confirmed); grooves reproducible with straightedge and abrasive (confirmed); H-sockets carvable with hammerstone and abrasive (confirmed) Unresolved predictions: Full-scale 90 km / 3,850 m block transport has not been experimentally replicated; the near-uniqueness of the H-socket technique within the wider Tiwanaku-period corpus is not yet explained

Where the conventional model still strains

Open issue	Why it matters	Model assumption required	Severity	Source
Andesite transport at altitude	Moving 85-tonne blocks 90 km at 3,850 m strains any account based on human labour alone	Reed/balsa rafts across Titicaca; sledges or rollers on the Altiplano; large coordinated labour force	Medium	Protzen & Nair 2013
H-socket clamps are near-unique	Posnansky’s drawings and photographs are the primary evidence; most clamps were lost to looting	Bronze cramp joinery, accepted elsewhere in antiquity (Greek, Roman)	Medium	Posnansky 1945
Why Puma Punku specifically	The precision stonework is exceptional even within Tiwanaku, not matched by other structures at the wider complex	The platform served an unusually prestigious function meriting exceptional investment	Low	Janusek 2008
No exact Andean parallel	The H-socket joinery is documented at only one other site, Ollantaytambu (per Sitchin, citing Posnansky)	Bronze-clamp technology restricted to a small number of elite structures	Medium	Sitchin, When Time Began, Ch. 9

The strongest unresolved question

Why is the H/T-socket bronze-clamp technique attested at only two Andean sites?

Puma Punku and, per Sitchin’s citation of Posnansky, Ollantaytambu appear to be the only Andean locations where this specific joinery method has been documented. If it were a routine part of the broader Tiwanaku engineering toolkit, simplified or partial versions might be expected at other elite Tiwanaku-period constructions. Its apparent restriction to a handful of sites is a genuine open question that current excavation has not resolved. It does not, by itself, require an imported-technology explanation — but it deserves more fieldwork than it has received.

What mainstream summaries often leave out

• Simplification: popular treatments often describe Puma Punku’s blocks as showing “laser precision.” The documented tolerance — flatness within 1–2 mm over several metres — is consistent with, and has been experimentally reproduced by, hand-abrasive lapping (Protzen & Nair 2013). It is exceptional craftsmanship, not evidence of a cutting tool beyond stone-age technology.
• Uncertainty: the widely repeated “85-tonne” and “130-tonne” block weights are derived from volumetric estimates of the largest surviving fragments, not from direct weighing, and carry meaningful margins of error that popular sources rarely mention.
• Unresolved prediction: if the H-socket clamp technique were a purely local Andean invention developed at Tiwanaku, intermediate or experimental versions of the joint might be expected at other, less prestigious Tiwanaku-period structures. None have been documented — a genuine gap in the conventional account of how the technique was developed, separate from the question of whether it required imported technology.

Continue the investigation

The free case file establishes the measurements, the strongest conventional explanation, and the question it has not fully closed. Members can continue to the exact Sitchin source audit, the strongest case for his framework, competing explanations, red-team analysis, the full chain of inference, and the complete source map.

Open the full evidence dossier →

The Sitchin connection

Connection classification

Classification	Definition	Applies?
Direct	Sitchin specifically names the object, site, or find	Yes
Indirect	Sitchin describes an event or mechanism that may relate to the find	Yes (departure-facility argument in The End of Days)
Thematic	The find matches a broader pattern in his work but is not itself cited	No
Later attribution	The connection was made by later authors, media, or fans	No
No connection found	No relevant connection exists in the available corpus	No

Editorial rule

Puma Punku is one of the more thoroughly documented Sitchin connections in this archive: he names the site, quotes his primary source (Posnansky) directly, and returns to it across three separate books with a consistent (if chronologically indefensible) argument.

What Sitchin actually wrote

“He found at Tiahuanacu stone blocks bearing niches that served to hold bronze bolts, as at Puma Punku. At Puma Punku he had seen a piece of metal, undoubtedly bronze, that ‘with its dentiform prongs looked like a tackle or device for lifting weights.’”

Citation: Sitchin, Z. The Lost Realms (1990), Chapter 5, “The Lost Realm of Cain?” Verified against the digital text corpus. Sitchin is quoting/paraphrasing Posnansky (1945).

Claim ID: SC-01

“…they say that the Creator was at Tiahuanaco and that there was his chief abode. Hence the superb edifices, worthy of admiration, in that place.”

Citation: Sitchin, Z. The Lost Realms (1990), Chapter 2. Quoting Father Cristóval de Molina (c. 1573).

Claim ID: SC-05

“We believe that Anu and his spouse Antu, visiting Earth from Nibiru circa 3800 B.C., also went to see for themselves the new metallurgical center on the southern shore of Lake Titicaca. They left by sailing away on the lake from the port facilities of Puma Punku, where the cyclopean chambers, carved and shaped out of single stone blocks, then stood alongside massive piers.”

Citation: Sitchin, Z. When Time Began (1993), Chapter 9, “Where the Sun Also Rises.” Verified against the digital text corpus.

Claim ID: SC-03

“He built the Eninnu with stone, he made it bright with jewels; with copper mixed with tin he held it fast.”

Citation: Sitchin, Z. When Time Began (1993), Chapter 9, quoting a building inscription of Gudea of Lagash, presented as evidence for the same bronze-clamping technique Sitchin attributes to Puma Punku. Verified against the digital text corpus.

Claim ID: SC-01

“…need an engineering degree to realize that these stones were cut, drilled, and shaped by someone with incredible technological ability and sophisticated equipment; indeed, one would doubt whether stones could be so shaped nowadays.”

Citation: Sitchin, Z. The End of Days (2007), Chapter 13, “When the Gods Left Earth.” Verified against the digital text corpus; quoted mid-sentence (the opening clause, “One does not,” is separated in the source by a page break and figure caption).

Claim ID: SC-04

Corpus check

Search term	Books searched	Direct matches	Relevant passages	Result
”Puma Punku” / “Pumapunku”	12 indexed	4 books	The Lost Realms p.~240; When Time Began p.~150, ~247; Journeys to the Mythical Past p.~243; The End of Days p.~234–244	Direct
”Tiahuanacu” / “Tiwanaku”	12 indexed	6 books	The Lost Realms; When Time Began; Journeys to the Mythical Past; The End of Days; Earth Chronicles Handbook; Divine Encounters	Direct
”Viracocha”	12 indexed	5 books	The Lost Realms; When Time Began; Journeys to the Mythical Past; The End of Days; Earth Chronicles Handbook	Direct (thematic, not tied to Puma Punku specifically)
“Posnansky”	12 indexed	4 books	The Lost Realms; When Time Began; Journeys to the Mythical Past; The End of Days	Direct
”Sacsahuaman” / “Sacsayhuaman”	12 indexed	4 books	The Lost Realms; When Time Began; Genesis Revisited; Earth Chronicles Handbook	Direct — parallel megalithic site, not Puma Punku itself

Corpus limitation: 4 of 16 Sitchin books are not yet OCR-indexed for this search (The Cosmic Code, There Were Giants, The Anunnaki Chronicles, The King Who Refused to Die). Any additional Puma Punku references in those volumes were not searchable for this dossier.

Sitchin claim decomposition

Claim ID	Atomic claim	Exact source	Required evidence	Evidence found	Status
SC-01	The H/T-shaped bronze-clamp sockets at Puma Punku are physical evidence of a deliberate metallurgical joinery technology, paralleled at Gudea’s Eninnu in Sumer	When Time Began, Ch. 9	Matching bronze artifacts and socket geometry at both sites	E-01, E-03	Partial — clamp technology confirmed; cross-continental technology transfer unsupported (bronze-clamp joinery is independently attested in Andean, Mesopotamian, Greek, and Roman contexts)
SC-02	Tiwanaku/Puma Punku dates to 12,000–17,000 BP, based on Posnansky’s archaeoastronomical reading of the Kalasasaya’s orientation	When Time Began, Ch. 9 (endorsing Posnansky); The Lost Realms, Ch. 5	Independent dating corroboration	E-06 (low reliability), contradicted by E-05	Contradicted
SC-03	Anu and Antu, rulers of Nibiru, visited the Lake Titicaca gold region c. 3800–4000 BCE and used Puma Punku’s harbor facilities	When Time Began, Ch. 9	Independent textual or archaeological corroboration of a specific fourth-millennium-BCE visit	None found independent of Sitchin’s own narrative	Unsupported
SC-04	Puma Punku served as a departure facility for the Enlilite Anunnaki leadership after the 2024 BCE destruction of the Sinai spaceport	The End of Days, Ch. 13	Physical trace of spaceport infrastructure contemporary with the claimed second-millennium-BCE date	None — earliest secure occupation postdates this claim by roughly 2,000 years	Contradicted by chronology
SC-05	The Andean creator deity Viracocha, associated with Tiwanaku, is identifiable with the Mesopotamian storm god Adad/Ishkur	The Lost Realms, Ch. 2 (Molina citation) and related discussion	Independent textual or iconographic link beyond thematic resemblance	Iconographic parallel only (staff- and weapon-bearing figure)	Open / speculative

The strongest case for taking Sitchin seriously

Where his pattern is genuinely interesting

• He flagged the clamp sockets as functionally significant decades before rigorous replication. Sitchin treated Posnansky’s field documentation of the H/T-socket bronze clamps as evidence of intentional, engineered joinery — not decoration or accident — years before Protzen and Nair’s 2013 experimental study confirmed that the sockets are purpose-built structural features, reproducible with local Andean methods but functionally exactly what Posnansky (and Sitchin, following him) described.
• The Gudea/Eninnu cross-reference is a genuine, non-obvious observation. Sitchin explicitly connects the “copper mixed with tin” bronze that Gudea of Lagash used to clamp the stones of the Eninnu temple with the bronze clamp technology at Puma Punku — a real textual and technological parallel between two widely separated ancient building traditions, worth noting even though bronze-clamp joinery is independently attested in several unrelated traditions and does not by itself require common origin.

What he noticed before it became widely discussed

Sitchin observation	Publication date	Later evidence or debate	Match quality
Bronze H/T-socket clamps at Puma Punku were a deliberate structural joining technology, not incidental damage or decoration	1990 (The Lost Realms) / 1993 (When Time Began)	Protzen & Nair (2013) experimentally confirm the socket geometry is a purpose-built joining feature, reproducible with local Andean techniques	Partial — functional significance affirmed; “imported/advanced technology” attribution not affirmed

The cumulative-pattern argument

Taken individually, the clamp-joinery parallel between Puma Punku and Gudea’s Lagash is weak: bronze cramps binding adjacent masonry blocks are also attested, independently, in Greek and Roman construction, meaning the technique is not unique enough on its own to imply contact or transfer. The stronger version of Sitchin’s argument is cumulative: he treats Puma Punku as one node in a wider global network of sites — alongside Baalbek, Giza, and Sacsayhuamán — that share megalithic scale, precision joinery, and (in his reading) metallurgical sophistication beyond what he attributes to independent local development. The individual parallels are each explicable on their own; the argument asks whether their repetition across unconnected regions is coincidence or pattern.

Pattern limitation

This argument is vulnerable to selection bias: Sitchin highlights the megalithic sites that do show bronze-clamp or comparable joinery evidence (Puma Punku, and by his account Ollantaytambu) while several of his other “megalithic parallel” sites do not show comparable metal-joinery evidence at all — an asymmetry the cumulative-pattern argument does not address. It is also worth noting that Gudea’s Lagash bronze clamping and Puma Punku’s construction are separated by many centuries even under Sitchin’s own preferred Mesopotamian chronology, which weakens any reading of the parallel as evidence of a single contemporaneous technology transfer, rather than two independent traditions of bronze-clamp joinery.

Questions Sitchin’s framework legitimately raises

1. Why does the H/T-socket clamp-joinery technique appear at Puma Punku and, per Sitchin’s citation of Posnansky, at Ollantaytambu, but not more broadly across other Tiwanaku-period Andean construction — is this a localized craft specialization, or evidence of contact with a distinct tradition?
2. Independent of Posnansky’s rejected astronomical date, can the Tiwanaku IV–V ceramic and radiocarbon sequence (500–1000 CE) be refined further to pin down when specifically the H-socket technique was introduced, and does that narrower date correlate with any attested cultural contact event?

Where the Sitchin case is vulnerable

• Translation dependency: Sitchin’s account of Posnansky’s clamp description (“dentiform prongs… tackle or device for lifting weights”) comes through Posnansky’s own English-language paraphrase of his field observations; no independent check of Posnansky’s original notebooks against Sitchin’s secondary quotation has been made for this dossier.
• Chronology dependency: the entire “Anunnaki construction and departure centre” reading depends on rejecting the converging radiocarbon and ceramic chronology (cal. AD 500–1000, E-05) in favour of Posnansky’s archaeoastronomically derived 12,000–17,000 BP date, or Sitchin’s alternate c. 4000 BCE Anu/Antu visitation date (E-06) — neither of which any independent post-1945 dating method supports.
• Missing positive evidence: if Puma Punku had been built or used by a technologically advanced non-human agency, diagnostic traces would be expected — tool marks inconsistent with stone-and-abrasive technique, metals or alloys not otherwise attested in the Andean archaeological record, or occupation debris predating 500 CE at the platform itself. None of these have been reported.
• Later attribution risk: several of the most extreme “alien technology” claims attached to Puma Punku in popular media — laser-cut stone, precision “impossible” for any ancient tool — do not appear in Sitchin’s own text at all. They derive from later documentary and online treatments and are explicitly excluded from this dossier’s Sitchin corpus.

The weakest link

Posnansky’s chronology. Sitchin built much of his Tiwanaku argument on Posnansky’s dating of the Kalasasaya to 12,000–17,000 BP — a chronology Posnansky derived from archaeoastronomical alignments that he himself acknowledged were methodologically fragile, and that every subsequent radiocarbon and ceramic study has rejected in favour of 500–1000 CE. If the site is roughly 1,000–1,500 years old rather than 12,000–17,000, Sitchin’s entire chronological framework for Puma Punku collapses: by his own Earth Chronicles timeline, the Anunnaki had long departed Earth before Tiwanaku’s builders ever quarried a single block.

Competing explanations

Hypothesis	What it explains	What it struggles with	Unique prediction	Current standing
H1: Tiwanaku stoneworking (Protzen & Nair)	Precision cutting, flat surfaces, grooves, and H-socket clamp joinery with documented Andean techniques	H-socket near-uniqueness; full-scale transport logistics not yet replicated	Tool-mark signatures consistent with abrasive/hammerstone technique, not rotary or laser cutting	Leading
H2: Generic “lost high technology”	The intuitive sense that the precision is “too good” for the attributed period	Experimentally falsified by Protzen & Nair’s stone-age-tool replication; no artifact of the proposed technology has been found	Rotary or laser-consistent striations, unknown alloys, or a power source — none reported	Rejected
H3: Sitchin’s Anunnaki construction/departure centre	The metallurgical evidence Posnansky documented; the cross-cultural Gudea/Eninnu bronze-clamp parallel	Requires accepting a chronology (E-06) contradicted by every independent post-1945 dating method (E-05)	Anunnaki-attributable artifacts or occupation layers dated well before 500 CE at Puma Punku specifically	Unsupported by current chronology

Hypothesis H1: Tiwanaku stoneworking

Best evidence: [E-07] Protzen & Nair’s successful experimental replication of flat surfaces, grooves, and sockets using only hammerstones and abrasive sand. Strongest unresolved problem: [E-01]/open issue — the near-uniqueness of the H-socket clamp technique within the broader Andean corpus, and the untested logistics of full-scale block transport. What would falsify or weaken it: discovery of tool-mark signatures at Puma Punku inconsistent with abrasive/hammerstone technique (for example, striation patterns only achievable with rotary cutting).

Hypothesis H2: Generic lost high technology

Best evidence or pattern: the subjective sense, widely repeated in popular media, that the stonework is “too precise” for its attributed period. Strongest unresolved problem: Protzen & Nair’s replication removes the necessity for any lost technology, and no artifact of a proposed lost technology (machine parts, unattested alloys, a power source) has been recovered. What would falsify or substantially weaken it: it is already substantially falsified by [E-07]; it would require positive evidence — a specific tool, alloy, or power source — to be revived.

Hypothesis H3: Sitchin-compatible model

Best evidence or pattern: [E-03] bronze clamp artifacts, [E-01] socket geometry, and the cross-cultural Gudea/Eninnu bronze-clamp parallel documented in When Time Began. Strongest unresolved problem: requires accepting a chronology (E-06) contradicted by every independent post-1945 dating method (E-05). Unique evidence the model predicts: Anunnaki-attributable artifacts or an occupation layer dated well before 500 CE at the Puma Punku platform itself — not merely elsewhere in Sitchin’s broader Andean narrative. What would falsify or substantially weaken it: continued absence, after further excavation, of any pre-Tiwanaku-IV occupation layer at Puma Punku specifically — which is the current state of the evidence.

Chain of inference

1. Observed object: precision-cut andesite and sandstone blocks at Puma Punku, with internal right angles, constant-width grooves, and H-shaped clamp sockets.
2. Measured property: surface flatness within 1–2 mm over several metres; grooves 3–6 mm wide with constant depth; block masses up to an estimated 130 tonnes; andesite sourced 90 km away.
3. Physical inference: these features are reproducible using hammerstones and abrasive sand-and-water slurry — Protzen & Nair (2013) demonstrated this experimentally.
4. Historical connection: bronze artifacts recovered on-site place the clamp technology within the documented Tiwanaku IV–V metallurgical tradition (cal. AD 500–1000).
5. Sitchin-pattern connection: Sitchin, across When Time Began, The Lost Realms, and The End of Days, reads the same bronze-clamp evidence as proof of an imported Anunnaki metallurgical and construction programme, tied to a fourth-millennium-BCE Anu/Antu state visit and a second-millennium-BCE departure use.
6. Advanced-technology conclusion: if Sitchin’s chronology (12,000–17,000 BP, or a fourth-millennium-BCE occupation) were correct, the bronze technology would predate independently attested Andean and Old World metallurgy by millennia, implying an external transfer; as securely dated (500–1000 CE), the site’s metallurgy instead falls squarely within the known, independently developed Andean bronze-working tradition.

Transition	Evidence strength	Main assumption
Step 1 → 2	Strong	Measurements are direct and independently repeatable
Step 2 → 3	Strong	Protzen & Nair’s experimental replication is peer-reviewed and methodologically transparent
Step 3 → 4	Strong	Bronze artifacts are securely co-located with the site and independently excavated, not solely reliant on Posnansky
Step 4 → 5	Weak	Requires accepting Posnansky’s rejected chronology and an archaeologically unattested Anu/Antu visitation narrative
Step 5 → 6	Weak	The “advanced-technology” conclusion depends entirely on the chronology assumed in the previous step, which mainstream dating directly contradicts

Weakest transition

Step 4 → 5: The secure archaeological fact — Tiwanaku-period bronze metallurgy dated to 500–1000 CE — does not, on its own, require or even suggest Sitchin’s much older, textually unattested Anu/Antu visitation narrative. The leap from “bronze clamps exist and are well-made” to “this metallurgy was imported by Nibiruan visitors circa 3800 BCE” depends entirely on accepting Posnansky’s discredited dating; remove that dating, and the historical connection in Step 4 supports only H1, not H3.

Red-team review

Strongest skeptical objections

1. Argument from incredulity. “This is too precise for ancient people” is an assertion of disbelief, not evidence of exotic technology, and is directly refuted by Protzen & Nair’s successful hammerstone-and-abrasive replication.
2. Cherry-picked chronology. Sitchin and Posnansky’s older date is contradicted by every independent dating method — radiocarbon, ceramic seriation, and stratigraphy — that has been applied to the site since 1945; no dating method independent of Posnansky’s own contested archaeoastronomical calculation supports the older figure.
3. Reliance on secondary description of lost evidence. Posnansky’s field drawings of the clamps are close to the only surviving documentation of the specific “dentiform prong” tool he described — most original bronze artifacts were lost to looting or melted down — so the strongest single piece of “advanced metallurgy” evidence rests on one witness’s account rather than an object available for re-examination.

Best available responses

1. Response: correct in general, and specifically confirmed at Puma Punku — [E-07] is the direct experimental rebuttal to the “too precise to be ancient” intuition.
2. Response: also correct, and decisive: [E-05] (radiocarbon and ceramic dating, high reliability) directly contradicts [E-06] (Posnansky’s archaeoastronomical date, low reliability). The Anunnaki-visit reading depends entirely on the date that fails this test.
3. Response: partially correct, but overstated. The existence of bronze technology at Tiwanaku is not solely dependent on Posnansky’s testimony — later excavations documented in Janusek (2008) and Kolata (1993) also recovered bronze and copper tools, chisels, and clamp fragments at the wider Tiwanaku site (E-03). Only the specific “dentiform prong” lifting device Posnansky described is undocumented beyond his own account.

Confirmation-bias check

• Did we look for sources that would refute the preferred (conventional) reading? Yes — this dossier specifically sought Sitchin’s own passages, in full, rather than relying on paraphrase, and did not exclude any verified quote that supported his reading, including the Gudea/Eninnu cross-cultural parallel, which is the strongest pattern argument available for his framework here.
• Are we using the same evidentiary standard for both models? Both the “12,000–17,000 BP” claim and the “500–1000 CE” claim were held to the same test — independent, replicable dating method. The conventional date passes on multiple independent methods; the Sitchin-compatible date passes on none published since 1945.

Decide for yourself

Question	Evidence favouring conventional reading	Evidence keeping alternative question open
Can the stonework be reproduced with known techniques?	Yes — Protzen & Nair (2013) demonstrated this experimentally	Full-scale replication of 90 km / 3,850 m block transport has not been attempted
Is the chronology settled?	Yes — radiocarbon and ceramic typology converge on cal. AD 500–1000	Posnansky’s archaeoastronomical argument remains a documented historical claim, even though rejected
Were the H-sockets for bronze clamps?	Yes — consistent with known clamp technology and bronze artifacts recovered at the site	The clamps themselves are largely lost; Posnansky’s documentation cannot be independently re-checked
Does Puma Punku require an Anunnaki explanation?	No — every physical feature is explainable within the Andean archaeological record	Sitchin’s cross-cultural argument (Gudea/Eninnu bronze parallel) is a genuine, if thematic, observation
Is the H-socket technique’s near-uniqueness explained?	Partially — restricted use at prestige structures is plausible	No other Tiwanaku-period site with the same technique has been documented besides Ollantaytambu

Three defensible readings

Conservative reading

The stonework, transport, and bronze joinery at Puma Punku are fully explicable within the documented Tiwanaku IV–V archaeological record (cal. AD 500–1000) using known Andean techniques. No anomaly requiring an explanation beyond skilled craftsmanship, abrasive lapidary methods, and centuries of accumulated labour remains.

Open-question reading

Protzen and Nair have closed the “how” of the stoneworking, but the “why here specifically” — the apparent restriction of the H-socket clamp technique to two known sites — and the full-scale logistics of 90 km high-altitude transport remain genuinely under-tested. These are legitimate open research questions, not evidence of lost technology, but they are not yet fully closed either.

Sitchin-compatible reading

If one accepts Posnansky’s rejected chronology and Sitchin’s broader Anunnaki-in-the-Andes narrative, the bronze clamp technology and its cross-cultural echo at Gudea’s Lagash become evidence of a deliberate, imported construction and metallurgical programme. This reading requires overturning the entire post-1945 radiocarbon and ceramic chronology of Tiwanaku — something no independent archaeological study has done.

Editorial position

Puma Punku’s stonework is a genuine, high-quality archaeological puzzle that was, until 2013, legitimately open. Protzen and Nair’s experimental replication has closed most of the technological gap: the precision cutting, constant-width grooves, and H-shaped sockets are reproducible with documented Tiwanaku techniques. Sitchin’s account, built on Posnansky’s real but chronologically indefensible field documentation, correctly identifies the clamp sockets as functionally significant years before mainstream archaeology confirmed it — but the chronology his broader Anunnaki narrative requires is contradicted by every independent dating method published since 1945. Puma Punku is an exceptional achievement of Andean civilisation, not an Anunnaki construction or departure centre — though its near-unique clamp technique remains a legitimate, under-studied question in its own right.

What would change the assessment

• Evidence strengthening conventional model: a full-scale experimental replication of 85-tonne-class andesite block transport from Cerro Khapia to Puma Punku across Lake Titicaca and the Altiplano, using only documented pre-Columbian methods.
• Evidence strengthening Sitchin-compatible model: discovery of an occupation layer or Anunnaki-attributable artifact at Puma Punku itself, securely dated well before 500 CE.
• Evidence weakening Sitchin-compatible model: continued absence of any pre-Tiwanaku-IV occupation at the site after further excavation — which is the current state of the evidence.
• Evidence changing the attribution: recovery of additional Posnansky field documentation (original photographs, notebooks) that could be independently cross-checked against Sitchin’s secondary quotation of the clamp description.

Open research questions

• Has any post-2013 excavation at Puma Punku recovered occupation material stratigraphically pre-dating Tiwanaku IV (before c. 500 CE)?
• Does a full re-translation of Posnansky’s original field notes match Sitchin’s secondary English-language quotation of the “dentiform prongs” clamp description?
• Has the H/T-socket clamp-joinery technique been documented at any Tiwanaku-period Andean site besides Puma Punku and Ollantaytambu, and if so, does its distribution correlate with any attested trade or contact network?
• Has anyone attempted a full-scale or accurately scaled model replication of 85–130-tonne block transport from Cerro Khapia to Puma Punku, to test the logistics claim experimentally rather than by modelling alone?

🔒 Want to read the full analysis?

This article continues with the complete Sitchin Analysis, cross-references, and source breakdown.

Unlock Full Access →

From $9/month - cancel anytime

Sources

Primary physical, archaeological and textual sources

1. Posnansky, A. Tiahuanacu: The Cradle of American Man. 4 vols. New York, 1945.
2. Stübel, A. & Uhle, M. Die Ruinenstaette von Tiahuanaco. Breslau, 1892. (Field drawings of cut stones, cited in Sitchin, Journeys to the Mythical Past.)
3. Cieza de León, P. Crónica del Perú. 1553.
4. Molina, C. de. Relación de las fábulas y ritos de los Incas. c. 1573.

Peer-reviewed research

1. No dedicated peer-reviewed journal article was independently accessed for this dossier beyond the monographs listed below. The primary technical source, Protzen & Nair (2013), is a peer-reviewed university-press monograph rather than a journal article.

Academic editions and translations

1. Protzen, J.-P. & Nair, S. The Stones of Tiahuanaco: A Study of Architecture and Construction. Cotsen Institute of Archaeology Press, UCLA, 2013.
2. Protzen, J.-P. Inca Architecture and Construction at Ollantaytambo. Oxford University Press, 1993.
3. Janusek, J.W. Ancient Tiwanaku. Cambridge University Press, 2008.
4. Kolata, A.L. The Tiwanaku: Portrait of an Andean Civilization. Blackwell, 1993.

Sitchin corpus

1. Sitchin, Z. The Lost Realms. Avon Books, 1990. Chapters 2, 5. Verified against digital text corpus.
2. Sitchin, Z. When Time Began. Avon Books, 1993. Chapter 9. Verified against digital text corpus.
3. Sitchin, Z. The End of Days. William Morrow, 2007. Chapter 13. Verified against digital text corpus.
4. Sitchin, Z. Journeys to the Mythical Past. Bear & Company, 2007. Chapter 12. Verified against digital text corpus.

Later alternative-history interpretations

No source in this category met this dossier’s citation standard (named author, checkable publication). Representative popular claims are addressed, and excluded, below.

Sources excluded

Source	Reason excluded
YouTube videos and documentaries claiming “laser cutting” or generic “alien technology” at Puma Punku	Anonymous or unattributable; no archaeological evidence; the specific claim does not appear in Sitchin’s own text
Hancock, G. Fingerprints of the Gods (1995)	Popular work; Tiwanaku chronology substantially outdated relative to current radiocarbon/ceramic consensus
Posnansky’s 12,000–17,000 BP chronology, used as a current dating source	Superseded by independent post-1945 dating; retained in this dossier only as a documented historical claim under audit (E-06), not as evidence for the site’s actual age

Revision history

Version	Date	Changes	Author
2.0	2026-07-20	Full rebuild to evidence-dossier-template v2.0: added dating table, evidence-at-a-glance table, six-dimension confidence dashboard, corpus check, claim decomposition (SC-01–SC-05), chain of inference, red-team review, three defensible readings, and four additional verified Sitchin quotes from three further books (When Time Began, Journeys to the Mythical Past, The End of Days)	Michal Placek / Claude (research assistant)
1.1	2026-07-19	Expanded with cross-cultural parallels and research gaps	Michal Placek
1.0	2026-07-19	Initial dossier	Michal Placek

Research disclosure

Editorial stance: Anunnaki Archive examines the strongest evidence-based case for taking Sitchin seriously. It does not assume that every Sitchin claim is correct, nor that conventional explanations are complete.

Authorship: Research, drafting, and quote verification for this revision were performed by an AI research assistant (Claude) under the Dr. Elias Voss editorial voice, using the local Sitchin text corpus (books/_text/) and the project’s deterministic quote-verification tool (scripts/verify/verify_dossier.py). Final review and publication decisions rest with Michal Placek.

Sitchin corpus searched: 12 of 16 OCR-indexed books in books/verify_corpus.pkl, specifically including The Lost Realms, When Time Began, Journeys to the Mythical Past, The End of Days, Earth Chronicles Handbook, Divine Encounters, and Genesis Revisited for terms related to Puma Punku, Tiahuanacu/Tiwanaku, Viracocha, Posnansky, and Sacsayhuamán.

AI use: This revision was drafted by an AI assistant. Every Sitchin quotation was checked against the local OCR text corpus with the project’s verification tool before inclusion; only quotes returned as VERIFIED (exact match) were used. Chapter attributions were manually traced against each book’s table of contents and in-text running headers. Academic and primary-source citations were carried forward from the dossier’s prior verified revision or drawn from established bibliographic knowledge of named, real publications; no CDLI, DOI, or other raw external identifier is asserted anywhere in this document.

Conflicts of interest: None.

Corrections: If you find an error or a primary source that changes this assessment, contact research@anunnakiarchive.com and reference Case ID AA-EVD-007.