The future of history
The ancient city of Nineveh, on the eastern banks of the Tigris river, was one of the most magnificent cities of the ancient world. Capital of the Neo-Assyrian empire during the 7th century BC, it was the cultural heir of many long-vanished Bronze Age civilisations (recent evidence even suggests that the famous ‘Hanging Gardens of Babylon’ were in fact located here). Despite being a notoriously bloodthirsty power, whose exploits included using the head of the vanquished King of Elam as a decoration in the royal pleasure garden, Nineveh was a highly cultured society for the period. Its final King, Ashurbanipal, was a keen reader and lover of knowledge at a time when his enemies were largely illiterate warlords.
As part of his conquests, he and his armies scoured the known world for cuneiform tablets about topics spanning technology, military tactics, myths and prophecies and stored them in a colossal library; the first of its kind in the world. Astonishingly, some of these tablets were already thousands of years old at the time. This collection was to become a thing of legend. According to the oral histories of the Armenians and Persians, it was the Library in Nineveh that inspired Alexander the Great to build his own centuries later during his conquest of the Middle East. But unlike the Great Library of Alexandria, whose contents were scattered, destroyed and lost to history, the Library of Ashurbanipal was to have a more unusual fate. In the 6th Century BC it was devastated by fire, but rather than destroying the tablets the inferno baked them solid, leaving them relatively undamaged but forgotten for over two and a half thousand years.
Rediscovered in the 19th century, over 34,000 tablets were unearthed and sent to universities and museums across the world, but in the process their contents were muddled, making it difficult to reconstruct. Nevertheless, much of our understanding of the pre-classical world, from Sumerian flood myths to the oldest known ‘Book’ — the Epic of Gilgamesh — have come to us from this collection. Due to the combination of the sheer scale of the find and the botched archeology, the vast bulk of the Library remains untranslated.
Then, in 2014, the Islamic State took control of Nineveh and Mosul, and began the systematic destruction of Assyrian history. As this chaos unfolded, the British Museum partnered with the University of Mosul to digitise and rebuild the entire Library of Ashurbanipal. Their goal was to scan in the many thousands of cuneiform tablets and fragments scattered across the world and reunite them on a vast central website, liberating them from museum basements and opening them to translation by professional and amateur Assyriologists across the world. And so, despite devastation, fire, war and entropy of the ages, the most ancient library in human memory is being resurrected and preserved for future generations.
From clay tablet to 3D scanner, technology has been essential in preserving human culture. As technological innovation accelerates so do the methods that we have at our disposal to do this, and we are already seeing signs that museums, trusts, charities, foundations and other organisations are at the forefront of these innovations. When the tomb of Pharaoh Tutankhamun was discovered in the 1920s, his body was cut in half at the pelvis and then into 18 pieces for transportation (no wonder there are legends of a curse). Contrast this with the 2015 exhibition of mummies at the British Museum, which explored eight mummies non-invasively using CT scanning and visualisation technologies, which identified the treasures inside, and replicated them using 3D printing.
These new techniques are now attempting to make up for Carter and company’s destructive antics. Funded by UNESCO and spearheaded by Oxford University’s Institute for Digital Archeology, the tomb of Tutankhamun is one of a multitude of sites now being thoroughly 3D scanned and printed for future study. Across the world, 3D scanning has exploded in the heritage sector over the last five years. Scottish Heritage and their technology partners are presently undertaking the laser scanning of ‘The Scottish Ten’, in order to better understand how to conserve and manage them.
The scans, which utilise cutting-edge laser scanners that can capture the environment to a resolution of 4mm in geotagged ‘point-cloud’ models, can then be used in “dissemination, learning and engagement”. Indeed, many of the industry leaders in 3D scanning are involved in such heritage projects. Edinburgh-based Luma 3Di on the other hand have scanned the neolithic site of Skara Brae and other objects of historic interest such as a beautiful 10th century Byzantine sardonyx and 17th-century Italian Baroque table — in near-photorealistic detail, while London-based Scanlabs, have embarked on projects such as the digitisation of Admiral Nelson’s jacket and a map of the Roman Underworld.
But uploading the physical world to an electronic hereafter is only one aspect of these trends, and some projects are even more ambitious in scope, aiming to save, resurrect, and reconstruct the past in forensic detail using games, augmented reality and virtual reality experiences. Dutch AR specialist Twnkls, as well as developing genuinely innovative augmented reality tools, have been involved in a number of interesting applications, such as the Museum of the Market Square in Rotterdam, which uses a specially designed viewfinder to overlay a 3D recreation of the medieval town square (it can also be viewed on tablet and smartphone apps).
But uploading the physical world to an electronic hereafter is only one aspect of these trends, and some projects are even more ambitious in scope, aiming to save, resurrect, and reconstruct the past in forensic detail using games, augmented reality and virtual reality experiences. Dutch AR specialist Twnkls, as well as developing genuinely innovative augmented reality tools, have been involved in a number of interesting applications, such as the Museum of the Market Square in Rotterdam, which uses a specially designed viewfinder to overlay a 3D recreation of the medieval town square (it can also be viewed on tablet and smartphone apps).
This encompasses everything from street plans and architectural diagrams, to details of who lived where, to records of all shipments in and out of port and what people were paid for different professions. Such a pioneering project is not without its difficulties, as Kaplan explains:
“The central scientific challenge of a project like this one is qualifying, quantifying and representing uncertainty and inconsistency at each step of this process. There are errors everywhere, errors in the document, it’s the wrong name of the captain, some of the boats never actually took to sea. There are errors in translation, interpretative biases, and on top of that, if you add algorithmic processes, you’re going to have errors in recognition, errors in extraction, so you have very, very uncertain data.”
This problem will always blight any project we have to resurrect the past accurately, which is more acute the further back we travel. Kaplan explains this as the ‘information mushroom’ meaning that the deeper the time, the more will be simulated as opposed to accurately represented.
Take a casual walk around the flea markets of the world and you will find an abundance of century-old photographs; scattered memories of people whose lives and names are all but forgotten. While this is somewhat sad, it is also remarkable that there are any remains of their mayfly lives at all.
Most information produced by humanity in its long existence has succumbed to such information entropy. Like our individual memories, which are ghostly and ever-shifting, our collective understanding of the past is elusive and fragmentary, tied together with mis-rememberings and falsehoods.
To use just one example, we have but a fraction of the works of the ancient Greek and Roman world and as such the picture we have of this period is subject to the ‘survivorship bias’ of their very existence, and the cognitive biases and political motivations of their authors (Caesar’s accounts of the conquest of Britain are both colourful and exciting tales of derring-do and fossilised political propaganda). Historians are painfully aware of these gaps in our knowledge, and mourn over a long list of lost works, such as the sequels to Homer’s Iliad and Odyssey and over a dozen books by the Roman historian Suetonius. We also know of some philosophers by mere fragments of their work, quoted or referenced by others. Then of course there are the unknown unknowns; things we do not know we’ve forgotten, permanent blind spots in our understanding of the past.
Authentic remembering and reconstruction on the past should always recognise that our knowledge will always be imperfect, however, this does not mean that reconstruction of the past is futile. With the many apparatuses of modern science, we can scrutinise the physical fragments of the past in atomic detail to extract new insights and information, while new advances in machine learning and ‘deep convolutional neural networks’ can piece together new patterns to history that have remained elusive to historians. Babak Saleh and Ahmed Elgammal at Rutgers University in New Jersey created an algorithm to mine a database of “80,000 paintings by more than a 1,000 artists spanning 15 centuries” which managed to identify similarities and influences between artists that had been missed by generations of art historians. With access to large enough data sets, who knows what insights future generations of Artificial SuperIntelligence (ASI) may glean.
But all this analysis is for nought if these insights cannot capture people’s imagination and engage them meaningfully, which has been the challenge of generations of historians. However, there is already growing interest in the exploration of authentic history from a comparatively new medium: games. One game, Digital Songlines by Niche Studio, looks to authentically recreate the culture of the Carnarvon Gorge Aborigines, where you can forage for witchetty grubs and even learn their language. While this ‘indie’ game remains niche, in the past decade, there has been a growing interest in video games exploring different historic periods.
Tired of stock sci-fi and Tolkienesque regurgitations, new mega-franchises such as Assassin’s Creed have provided for some of the first introductions to the worlds of Renaissance Florence and Revolutionary France. While the plot was cloaked in the pseudo-history of the franchise, the games do an admirable job (in the context of a premium game title) of trying to rebuild accurate reconstructions of the physical and social structures of the time periods.
Kingdom Come: Deliverance is another upcoming game that seeks to take this hunger for authenticity one step further, by looking to create a fully historically accurate recreation of 15th-century Bohemia during the Hussite Wars. It does of course make some compromises on accuracy for the sake of gameplay; in-game characters do not speak in Medieval Czech — but it does go further than any game of this scale. Importantly, this game was crowdfunded through Kickstarter and has developed in line with audience expectations, such as adding support for VR headset Oculus Rift.
With the levels of immersion offered by even the first generation of VR headsets, combined with accurate 3D models of historic objects and architecture, AI insights combined with human expertise in everything from fashion, to art history, linguistics and etiquette, will in time construct dazzlingly realistic simulacra of the past. The ‘gameplay’ of such worlds need not even be the familiar hack-and-slash role-playing mechanics, but something much more absorbing. In recent times, indie games (particularly those made by Bristol studio The Chinese Room) have pioneered a genre known lovingly as ‘walking simulators’ which have explored new ways to tell stories by environmental exploration. These projects open up tantalising new possibilities to reconstruct the past, allowing us to be tourists in L. P. Hartley’s “foreign countries”.
But if the past is a foreign country, then the future will be another world. The massive and accelerating accumulation of information that characterises the third millennium is only getting started. What we consider ‘big data’ now will seem as charming as computers with 8k of memory in the coming centuries. After the convergence of the ‘Internet of Things,’ the ‘Smart Home,’ and the ‘Smart City’, harvesting every heartbeat and movement in return for fine-grain insight into our own behaviour.
In short, soon everything we do, see and say will be captured and stored in sea-powered server farms, and in increasing resolution. Hand in hand with the imminent boom of VR technology are new cameras which record 3D video, such as Nokia’s Ozo, and the remarkable Lytro Immerge camera, the latter of which does not so much record ’video’, but instead collects information about lightwaves themselves. This allows entire scenes to be reconstructed in VR in a way that allows “six degrees of freedom in a live action environment that would be photorealistic”. In short, we could walk through such recordings as if we were physically there.
The advent of such technologies has implications not just for documentarians and filmmakers, but for historians. Today, Ozo and Lytro Immerge are extremely high-end pieces of kit. One day — perhaps sooner that you think — technologies like these will be ubiquitous, attached to every phone or headset, or whatever devices succeed them, permeating the fabric of our future cities. In the centuries to come, for every instant of time you could conceivably have an electronic snapshot of the entirety of human experience that could be rebuilt and explored virtually. One could be the expert of a single moment, itself a vast time-slice that would take many lifetimes to fully explore.
We could stand on a street corner and travel back centuries to see what it looked like at 9.45am on Tuesday 23 May 2045, or navigate with AI assistants through the causal forces of history that led to a single moment. Together, we could haunt the hallways of the past like ghosts.
But we should not forget that demon of information entropy has not yet been banished. Much of the data we are beginning to accumulate could be lost through incompetence and indifference, and much already has. Google’s Vint Cerf even talks of a “lost century” as we let programs that read old data succumb to extinction. He said:
“We don’t want our digital lives to fade away. If we want to preserve them, we need to make sure that the digital objects we create today can still be rendered far into the future… We are nonchalantly throwing all of our data into what could become an information black hole without realising it. We digitise things because we think we will preserve them, but what we don’t understand is that unless we take other steps, those digital versions may not be any better, and may even be worse, than the artefacts that we digitised… If there are photos you really care about, print them out.”
In other words, we should celebrate but not be complacent about our Digitised Library of Ashurbanipal, which, despite its liberation from the frailty of matter, is an energy crisis or format change away from oblivion. While rematerialising the tablets with 3D printers is one solution, it is again just another link in a long chain or remembering in this information stream, going back to the omens and oral histories of antediluvian Iraq that predate recorded history.
To store information important to their clan, some Aboriginal rock artists still repaint the illustrations made by their long-dead ancestors many thousand of years in the past. Similarly, gallery restoration teams repaint and renew the works of the Renaissance masters, extending their still finite lifespans further into the future. Remembering, preserving and recreating the past require continued and active human involvement, and the mastery of technology.
This was first published as a white paper for The Team
