Archaeological Views: Digital Archaeology’s New Frontiers

028

Seventeen years into the “digital millennium”—which supposedly began in 2000—it would be understandable if a certain cynicism arose for everything new, digital, high tech or cyber. Particularly in archaeology—what have digital tools really done for us that earlier, traditional tools did not? Is a digital photograph that much better than a film photograph? Does a laser distance measurer produce better archaeology than established survey tools? What does a computer 3D model do that a hand-drawn or built model cannot? Do we really understand the ancient world more clearly thanks to computers?

As a digital archaeologist, I believe I can affirm the value of digital technology for understanding ancient material culture (the physical evidence left behind by a culture).

First of all, digital archaeology has the power to improve upon the human capacity for discovery. We often praise technology when it mimics human ability (e.g., artificial intelligence, virtual reality), but digital technology has the power to work well beyond the human physical and mental limitations. For example, photography has long been an essential tool for archaeologists. Digital photography could be used as a mere substitute for film photography (perhaps cheaper or quicker). Most digital cameras are designed to mimic the human eye, but some specialized cameras transcend human perception. The human eye is limited in range and resolution of color; on the electromagnetic spectrum, the colors visible to us range from violet to red, excluding ultraviolet and infrared. Archaeologists used infrared photography to capture the reflectance of infrared light—which the human eye can’t see—long before the advent of digital cameras. But the unique power of digital photography becomes clear when we consider that our color resolution is limited to three color receptors. All the colors we are able to see can be expressed as a combination of red, green and blue (RGB). When we look at a spectrum in a rainbow or prism, we see the colors in bands rather than a smooth gradient. These bands in color resolution can be compared to the “pixelation” associated with spatial resolution. A human is considered “color blind” if one of the three receptors malfunctions. Imagine the benefit of adding a fourth, fifth or sixth receptor. And now consider that today’s spectral imaging systems can easily resolve 16 discrete bands of color.

The real power of digital photography for archaeology is that the entire image exists as meaningful numbers. A pixel in one photograph represents the exact same area on the artifact as in the next photograph (as long as the camera and object don’t move). That pixel is represented as a number expressing how much light is reflected from that area (rather than a density of chemical reaction, as on film). For example, today’s specialized cameras (such as MegaVision E7) distinguish 50 million tiny squares and rank the light reflecting from that area on a scale of zero to more than 65,000. A second photograph—and then a third and up to the 16th—could be taken with a slightly different color of light. A human looking at all these numbers would be easily overwhelmed. But crunching numbers is where digital technology excels. Statistical techniques, such as Principal Component Analysis, can easily reduce redundancy. In digital photography, this means finding and ranking the greatest contrasts detectable in the full range of data. Since different materials have different spectral signatures, spectral imaging can distinguish them even if they look identical to the human eye.

To be able to distinguish contrasts has significant029 potential for archaeology. A faint trace of pigment on parchment or pottery can be difficult to read, especially if the parchment is decayed or the ink was deliberately erased (as is often the case with recycled manuscripts, or palimpsests). A forger may have matched the color of pigments in a painting as far as the eye can see, but spectral imaging can easily spot the difference. The spectral signature of a pigment from one region can be distinguished from a pigment with a similar appearance from a different region. Fundamentally, materials do not look the same to spectral imaging unless they are in fact the same.

Besides color contrast, archaeological imaging needs to capture texture. Inscriptions and coins are the clearest examples of small-scale objects in which the primary intended meaning is conveyed through texture. Archaeological publications often use line drawings or photography with a low angle of raking light (bright light shown almost parallel to the photographed object). With raking light from just the right angle, a photograph can show the texture of a cuneiform tablet through the highlights and shadows. Sometimes, however, one angle is not enough. In that case, digital texture imaging can map the entire surface, effectively creating an image in which the viewer can control the angle of light. This procedure, called Reflectance Transformation Imaging (RTI), is more demanding of the photographer, since it amounts to taking 40 or more images of the same object while moving only the angle of illumination. The end result is then a single image in which we know not only the color of each pixel but also its texture. So far, this is not necessarily better than human perception, since a human (assuming we have access to the object) can easily move a nonvirtual flashlight to visualize the texture. But an RTI image is more than a series of raking light images. Since the texture is known on a digital level, it can be amplified or enhanced to be seen more clearly than is possible with natural observation.

Digital texture imaging has been used by archaeologists for more than a decade. In particular, the West Semitic Research Project at the University of Southern California has built the InscriptiFact Digital Image Library, including RTI texture images of many coins and inscriptions.^a Cultural Heritage Imaging, an independent non-profit, has been promoting the technology more widely.

Two other developments are more recent. First, as the spatial resolution capabilities of digital imaging increase, we can see finer texture than the relatively broad textures intended for human observation (e.g., inscriptions, coins). Today it is possible to observe the texture as fine as the corrosion of parchment due to acidic ink, such that we can read the outline of the letter even if the ink is now completely absent. We can observe the thickness of ink, the hair and flesh sides of parchment065 and the fine nuances of a potter’s reed, a painter’s brush or the technique of papyrus production. The other recent innovation combines the benefits of spectral imaging with RTI texture imaging into a single image set, such that we know both spectral signature and texture of any pixel and can enhance each in combination. This new technology, called Spectral RTI, has recently been developed by the Jubilees Palimpsest Project. Funded by the National Endowment for the Humanities, this project now aims to apply the technology to the Jubilees Palimpsest and to provide tools, documentation and training to museums, libraries and scholars throughout the United States.

Photography is only one area in which digital technology has provided benefits over its predigital counterpart (film). There is nothing wrong with archaeology made easier through digital assistance, but digital archaeology can live up to its potential—and the hype—only when it fully embraces the potential to transcend human ability and even human categories for organizing information. One of the most profound implications of a digital archaeologist being able to see, study and discover more (not just faster) than a traditional archaeologist is that artifacts already in museums and libraries become ripe for digital “excavation.” Tens of thousands of palimpsests become readable; evidence of the techniques and pigments used by ancient artisans can be digitized and compared using powerful tools. The realm of new discovery includes not only what is beyond human perception because it may be buried in dirt, but any artifact in museums and libraries that has been studied only with conventional, human perception or digitized only with conventional photography.

Some may rightly wonder if that066 is a good thing. Do we need another avalanche of data? Anyone who can remember his or her first day wandering through the stacks of a major research library knows that information overload arrived long before the digital era. Uniform standards for categorizing and searching information existed before digital technology, and they are no less powerful with digital technology.

Perhaps the biggest obstacle to discoverability and interoperability is the human desire to protect our “property.” This, too, has a predigital counterpart in the editors who took far too long to publish information or distorted the presentation of the data to fit a skewed interpretation. Openness of intellectual property is progressing as institutions realize that a light under a bushel has less value than one that shines openly.

If previous generations erred in the direction of a few individuals hoarding control of objects of study, we may soon face a problem in which data is published online faster than anyone can study it—even if it is fully open and accessible. This problem can be solved by increasing the number of people working on interpreting the data. Fortunately, one of the fundamental features of digital information is that it can be replicated precisely and transmitted quickly and cheaply. This means that anyone in the world can now do work that was once reserved to only a few elite institutions. Democratizing access to archaeology—not just learning about it but participating in it—will certainly benefit the field and increase the global appreciation for study of ancient material culture.

Seventeen years into the “digital millennium”—which supposedly began in 2000—it would be understandable if a certain cynicism arose for everything new, digital, high tech or cyber. Particularly in archaeology—what have digital tools really done for us that earlier, traditional tools did not? Is a digital photograph that much better than a film photograph? Does a laser distance measurer produce better archaeology than established survey tools? What does a computer 3D model do that a hand-drawn or built model cannot? Do we really understand the ancient world more clearly thanks to computers? As a digital archaeologist, I believe I can […]