*I want to believe…

There’s a certain type of problem that, as someone who’s spent a lifetime in and around computer systems, I find particularly magnetic. It’s the kind of problem where the output seems to wildly exceed the capabilities of the documented input and processing power. In the world of software, you might find this in a piece of old, elegant code that achieves something you thought required a modern, heavyweight framework. It’s an anomaly that forces you to stop, reassess your assumptions, and dig deeper. For me, the ancient ruins of Puma Punku, high on the Bolivian Altiplano, represent one of the greatest historical and engineering anomalies of this kind.

The site is, to put it mildly, breathtakingly perplexing. Scatterings of colossal, intricately worked stones lie about, remnants of a structure that was part of the great Tiwanaku civilisation, which flourished long before the Inca. And when you look closely at these stones, particularly the andesite ones, the neat, orderly narrative of history begins to fray at the edges. You see surfaces polished to a near-perfect flatness, interior corners cut to a precise ninety degrees, and complex, interlocking shapes known as ‘H-blocks’ that look for all the world like standardised, modular components from some giant, enigmatic construction set. The immediate, almost instinctive question that arises is not just who built this, but how? This simple question is the gateway to one of the most persistent and polarised debates in archaeology, a debate that pits meticulous scholarship against tantalising, and at times fantastical, speculation.

To analyse this properly, we need to function like a systems analyst debugging a complex problem. First, let’s define the system’s parameters: the knowns. Puma Punku is situated at an altitude of nearly 13,000 feet, a harsh, unforgiving environment. The consensus among archaeologists places its construction somewhere around 500 to 600 AD [1]. The primary materials are red sandstone, which was quarried nearby, and the far more challenging grey andesite, a volcanic rock of incredible hardness. Geologists have traced the source of the andesite to a quarry some 90 kilometres away, across Lake Titicaca [2]. So, right away, we have two immense logistical problems before a single stone is even shaped: quarrying massive blocks of one of the planet’s tougher stones, and then transporting them over a considerable distance at high altitude.

The true heart of the debate, however, lies in the processing of that andesite. On the Mohs scale of mineral hardness, diamond is a 10. The tools available to the Tiwanaku culture, as far as the archaeological record shows, were primarily made of stone, wood, and a relatively soft bronze alloy. Andesite sits at around a 6 to 7 on the Mohs scale, whilst bronze is only a 3.5. This is the central conflict in the code, the point where the program seems to crash. How do you use a softer material to carve, with uncanny precision, a much harder one? It’s akin to trying to sculpt steel with a spoon.

This is where the debate splits into two main branches, and it’s here that a logical, evidence-based approach becomes absolutely critical.

The first branch is the mainstream archaeological explanation. It is an argument for human ingenuity, patience, and immense social organisation. The most influential work here comes from figures like archaeologist Jean-Pierre Protzen of the University of California, Berkeley. Protzen didn’t just theorise; he went to the Altiplano and engaged in experimental archaeology. He demonstrated that it was possible to shape andesite using the tools the Tiwanaku had. His conclusion was that the primary shaping tool was not a metal chisel, but another stone. By laboriously pounding the andesite blocks with harder river cobbles, workers could slowly peck and grind the stone into the desired shape [3].

The finishing, he argued, was achieved through a gruelling process of grinding with sand and water. Think of it as a form of human-powered wet-sanding on a monumental scale. This technique, whilst astonishingly labour-intensive, can indeed produce incredibly flat and smooth surfaces over time. It doesn’t require advanced technology, but it requires something just as remarkable: a vast, coordinated workforce and an almost unimaginable amount of time. The logical model here is one of brute force. In computing, we sometimes solve problems not with a clever algorithm, but by throwing immense processing power at it. The orthodox view sees Puma Punku as the ultimate example of a human ‘brute-force’ attack on a geological problem. The precision, in this view, came from meticulous scribing, patient measurement using simple plumb-bobs and levels, and generations of accumulated craft knowledge.

The famous I-shaped metal cramps, which some have pointed to as evidence of advanced metallurgy, have also been analysed. They were cast in place from a nickel-bronze alloy, poured into carved indentations in the blocks to hold them together [4]. A clever technique, certainly, but one that doesn’t necessarily demand a lost high technology, merely skill in smelting and casting, which we know other ancient cultures possessed. The achievement, from this perspective, is not technological but societal. It speaks to a culture with the project management skills and social cohesion to mobilise thousands of people for a single, multi-generational construction project. The wonder, then, is the Tiwanaku’s societal software, not their physical hardware.

Then we come to the second branch of the debate, the one that captures the popular imagination. This viewpoint looks at the evidence and concludes that the orthodox explanation simply doesn’t compute. Proponents of this view, often found on television documentaries and popular internet forums, argue that the precision, particularly the perfectly straight lines, razor-sharp internal angles, and the purported drill holes, are simply impossible to achieve by pounding rocks together. Their argument, in essence, is one from incredulity. As Arthur C. Clarke famously wrote, “Any sufficiently advanced technology is indistinguishable from magic.” They look at Puma Punku and see ‘magic’—or rather, technology so advanced it must have come from somewhere else.

This is where theories of a ‘lost’ technologically advanced ancient human civilisation—an Atlantis-like precursor—or even extraterrestrial intervention enter the picture. The reasoning is that if the known tools couldn’t do the job, then unknown tools must have been used. They point to the H-blocks and see not just modular components, but mass-produced, machine-made objects. To them, the clean lines and complex internal cuts aren’t evidence of patience, but of power tools, maybe even sonic drills or laser cutters.

As someone trained to analyse systems, this is where I find the logic requires careful scrutiny. This alternative explanation is essentially a hypothesis based on a negative. It says, “Because we cannot explain it with X, the answer must be Y.” The problem is that Y—be it ancient aliens or a lost super-civilisation—is a colossal assumption that carries an enormous burden of proof. In system diagnostics, you don’t just invent a new, undocumented component to explain an anomaly. You must first exhaust every possibility within the known system, and if you do propose a new component, you need independent evidence for its existence.

Where are the remains of these advanced tools? Where is the evidence of the power source required to run them? A laser cutter doesn’t work without a generator. An advanced society doesn’t just appear, build one site with baffling precision, and then vanish without leaving any other trace—no lesser factories, no metallurgical workshops, no written language describing their engineering principles. We have the alleged output of the system (the stones), but we have zero evidence for the advanced processing hardware that is being proposed. It’s like finding a flawlessly compiled application but having no trace of the computer, the operating system, or the programming language it was written in. Logically, it’s a far greater leap of faith than believing in the power of human patience.

The debate also highlights a profound, and I think rather sad, cognitive bias we have: a tendency to underestimate the capabilities of ancient peoples. We live in an age of technological dependency, where a difficult task is solved by a better machine. It’s become difficult for us to truly comprehend what can be achieved with time, manpower, and deep, non-mechanised craft skill. We look at the problem through our own lens and, unable to imagine ourselves undertaking such a task, we project that inability onto the Tiwanaku.

Perhaps the most balanced modern view, championed by archaeologists like Dr Alexei Vranich, tries to synthesise these perspectives. Vranich suggests that the Tiwanaku likely used a combination of all the techniques available to them, and that the construction was likely a much more dynamic, and perhaps chaotic, process than we imagine [5]. He’s even used 3D modelling to re-examine how the scattered blocks might have fit together, suggesting Puma Punku might not have been a single, monolithic temple but a series of structures, perhaps built and rebuilt over time. This approach treats the Tiwanaku not as primitive labourers or as conduits for alien technology, but as what they were: clever, adaptable, and ambitious human beings working at the limits of their knowledge.

Ultimately, Puma Punku doesn’t force us to choose between primitives with stone hammers and aliens with laser beams. It forces us to reconsider the very definition of technology. Is technology only about machines and power sources? Or can technology also be a system of social organisation, of passed-down knowledge, of refined manual technique, and of a shared cultural will powerful enough to shape mountains? The Tiwanaku may not have had the hardware we recognise, but their societal ‘operating system’ was clearly capable of executing projects of breathtaking complexity.

The mystery of Puma Punku is not really about who built it. The evidence points overwhelmingly to the Tiwanaku people. The real, more profound question is how they marshalled the ingenuity and collective will to achieve it. Looking at those silent, broken stones on the cold, high plain, I don’t see a definitive argument for ancient astronauts. I see a powerful and humbling testament to a form of human genius we have largely forgotten, and a reminder that the simplest inputs—stone, sand, water, and time—can, with the right human system in place, produce an output that challenges our understanding of the possible. The debate itself is the legacy; it ensures we keep questioning our assumptions, and that is the first step toward any true understanding.

References and Further Reading

1. Goldstein, P. S. (2005). Andean Diaspora: The Tiwanaku Colonies and the Origins of South American Empire. University Press of Florida.
2. Janusek, J. W. (2008). Ancient Tiwanaku. Cambridge University Press.
3. Protzen, J.-P., & Nair, S. E. (2013). The Stones of Tiahuanaco: A Study of Architecture and Construction. Cotsen Institute of Archaeology Press. This is perhaps the most critical text for understanding the mainstream construction argument through experimental archaeology.
4. Lechtman, H. (1998). El bronce y el horizonte medio. Boletín del Museo del Oro, (44-45), 143-186. (An academic look at Andean metallurgy, including the analysis of Tiwanaku cramps).
5. Vranich, A. (2018). Reconstructing ancient architecture at Tiwanaku, Bolivia: the potential and promise of 3D printing. Heritage Science, 6(1), 65.

If this topic has raised your interest, Jean-Pierre Protzen and Stella Nair’s book is a deep and rewarding, if academic, dive into the practicalities of the construction. For a broader context on the civilisation itself, John Janusek’s Ancient Tiwanaku provides an excellent and comprehensive overview.

---