Words That Survive

Sheldon Greaves, Ph.D.

They had been digging in this mound since 1964, and after four sun-parched years of excavations, archaeologists from the University of Rome found a cuneiform tablet written in Akkadian that identified the name of the ancient city as Ebla. Scholars learned little else about this mysterious ancient city, less than an hour's drive south of Aleppo, Syria . Then on a magical day in 1974, the excavators struck archaeological gold: the palace archives. Some 18,000 clay tablets were found buried where they had lain for some 4,300 years. Not only did these tablets contain the records of one of antiquity's larger cities (by their own records, Ebla supported a population of nearly 200,000), many were written in a Semitic language no one had ever seen before. Fortunately, there were vocabulary list tablets in Sumerian and the new “Eblaite” language, so now scholars can follow the history and inner workings of a major ancient city.

What is particularly remarkable about this find was the way in which it was preserved. Around 2250 BCE, the city was attacked and destroyed by Naram-Sin of Akkad . The palace archives were shelved with wood and the tablets themselves stored in baskets. Most clay tablets in antiquity were sun-dried, but the palace was burned when the city fell, and in the ensuing blaze the tablets were fired, rendering them even more durable. For purposes of archaeology, fired clay is virtually indestructible; by destroying the palace, Naram-Sin preserved its contents practically for all time. Thus when the team of diggers finally exposed the archives to a sun over four millennia older than the one by which they were written and dried, their translator, Giovanni Pettinato, was able to read them with relative ease—once he had mastered Eblaite.

Figure 1. A clay tablet written in Old Babylonian (2nd millennium BCE). Photo U.S. Department of State, Bureau of Educational and Cultural Affairs.

There are other instances of writing preserved for long periods of time; papyrus scrolls and wall paintings in Egypt have survived for astonishingly long periods, and the Dead Sea Scrolls remained hidden in their caves at Wadi-Qumran for two thousand years. In both cases the preservation of these texts was due to the extreme dryness of the climate. Once the Dead Sea Scrolls were removed from the site and taken to more humid climates for study, they immediately began to deteriorate. When found, these parchments looked as though they had been written yesterday. Now, many of these same texts have been reduced to blackened scraps that will only yield their secrets to the most sophisticated imaging techniques. Others have decayed beyond recovery. Fortunately, an American archaeologist, John Allegro, made excellent photographs of most of the major scrolls within weeks of their discovery.

Inscriptions in stone are another robust form of writing that can outlast centuries. Even graffiti scratched on a wall can last for thousands of years. Ostraca are pieces of broken pottery—the scratch paper of the ancient world—often used to draft out documents. Those document have long since vanished, but many ostraca “drafts” survived. But if you want to see “writing” in a really durable form, consider the tracks of ancient creatures made hundreds of millions of years ago in soft primordial mud and still discernable today.

The Problem: What of Our Written Legacy?

Now, consider the plight of a future archaeologist who stumbles across one of our cities. He or she might find some inscriptions on public monuments or gravestones, but what might they find in our libraries or our archives? When so much our daily doings are recorded on magnetic tape, floppy disks, CDs and DVDs, what are the chances that historians of the future will be able to read any of it? Much of the information that defines our society is passed down from one technology to another; film to video to one digital format after another; all of it stored on relatively perishable media. If the power failed permanently tomorrow, and our civilization basically closed up shop and was consigned to the dust, much of what we are would soon vanish forever.

The biggest enemy of information in the modern age is obsolescence. I vividly recall how I finally located a long-sought article in a university library. It was printed on a storage medium called an opaque microform that apparently never really caught on. This was a text reduced to near-microscopic size, but on a sheet of special card stock, not a translucent medium like a microfilm or microfiche. And as “luck” would have it, the only machine in the entire university library capable of reading that opaque microform did not have the capacity to make a photocopy as other microfiche readers did. So, because I needed a copy of the whole article, I had to transcribe it by hand from the display screen like a medieval monk. As I spent the afternoon scrawling in frustration, I couldn't help thinking of an anonymous note made by such a scribe at the end of a medieval manuscript: "He who does not know how to write supposes it to be no labor; but though only three fingers write, the whole body labors."

Migration is the word used by archivists to describe the process of moving data from one storage technology to another. It is not unlike the process by which ancient manuscripts were copied from generation to generation, as old copies wore out or faded or were damaged by mildew, insects or other enemies of parchment and paper. Archives by their very nature often have trouble keeping ahead of the obsolescence of hardware and software. In many ways the preservation of information is driven by a society interested in preserving it, whether it be an ancient “singer of tales” who transmits information by oral tradition, or a modern archivist migrating computer data from magnetic tape to the latest long-term storage solutions.

Just how robust are our modern media? Actually the answer depends partly on some of the same factors that influence the longevity of ancient records. The quality of materials used, the process by which the information is recorded, and, above all, the storage conditions can make all the difference.

Magnetic media, such as digital magnetic tape, can last for 10 to 30 years, or longer in a controlled environment. Most optical media is expected to last from 20 to 30 years, but with top-quality materials and equipment estimates range up to 50 to100 years, vastly outstripping the life span of the hardware and software systems needed to read them. Photographic film's performance varies greatly, partly because of the rapid and drastic changes in technology. The old nitrate celluloid used in early movies was highly unstable, and as a result some 70% of the old silent movies are gone. Even acetate film can be problematic from an archivist's perspective. On the other hand, good old-fashioned paper that is acid-free and has a high rag content will last for centuries, especially if kept at relatively low humidity and constant room temperature. Another potentially good long-term storage medium is microfilm, if it is polyester microfilm and is stored under proper conditions. Theoretically it should last as long or longer than paper. Another plus for microfilm is that it is relatively technology independent. You just need to illuminate it under magnification to read the contents.

This raises the problem of estimating longevity. This usually involves testing in which the medium is subjected to extremes of heat, cold, and humidity. Sometimes the results are useful predictors of longevity, but sometimes not. I've read some comments by archivists who are very skeptical of the usefulness of these estimates.

But even under ideal conditions the experts don't hold out much hope that most of these technologies will last for very long; certainly not compared to the humble clay tablet or stone inscription. In fact, paper-based records seem to offer the greatest longevity if, like their papyrus ancestors in Egypt , they are stored under the right conditions.

Where modern media really shines is in information density. One page of a medieval manuscript might contain as many as 40 or 50 lines of text. A compact disk takes up less space than a single parchment page and could easily contain the texts of an entire medieval monastic library. Ironically, it seems that as the technology for packing more and more information into a given volume increases, the longevity of that information diminishes.

A Challenge: You Against Time

So here is a question: How might one store information in a modern medium that is durable, but can still offer some of the compactness of modern media? In other words, what improvements can modern technology make on the stone inscription or the clay tablet?

Let's get down to specifics:

Information density. This technique should not have an information density per unit area of less than a standard, printed page in any trade book in octavo format—roughly 20 x 14 centimeters.

Technology independent. Use all the high tech you want for creating your text, but not for reading it later. No special readers, display units, or anything else other than simple optical magnification or artificial light, unless you can think of some way to integrate the display technology into the medium itself, and you can show that it should last as long as the storage medium.

Resistance to decay or degradation. This medium needs to last. Humans have been able to write in the conventional sense for about 5,300 years. For the sake of this exercise, let's say that it needs to last no less than 10,000 years under normal storage conditions; at best in a library, at worst buried in the ground. If you can make it resistant to fire or similar hazards, so much the better. I should also acknowledge that so far, we're only talking about text information here. Other information such as images, audio recordings, video, and so on have their own problems and solutions.

Allowances for translation. Languages change. They evolve, grow, assimilate other languages and gradually die or mutate beyond recognition. If you want your text to be understood ten thousand years from now, you'll want to make some educated guesses about what people will be speaking at that time. That's a bit beyond the scope of this article because we're more concerned with technology, so let's just assume that once you've perfected the technology, your text will be written in at least two widely understood languages.

You get extra credit if it can be made in the average home workshop.

What's Already Being Done

Not long after I wrote the above paragraphs I recalled something I'd read about a project along these lines in an old issue of Whole Earth Review . After some fruitless searching I finally located information about “The Rosetta Project,” an effort by the Long Now Foundation to catalog and preserve in some form as many of the planet's languages as possible. This ambitious undertaking is motivated partly by the fact that many of the world's less populous languages are dying off as the number of speakers for a particular language grows smaller and smaller. The Long Now Foundation deserves credit for trying to preserve this knowledge; human language encodes surprising amounts of information about the history of the peoples who used them, and offer many clues into the human phenomenon and psychology of language.

Figure 2. The Rosetta Disk with thousands of pages of linguistic information microscopically engraved onto a nickel disk. Copyright by the Rosetta Project, which accepts donations at http://www.rosettaproject.org/ .

Part of this effort is the Rosetta Disk. This is a part of the Long Now Foundation's “10,000 Year Library.” This disk contains information on world languages and linguistics specifically selected to assist future scholars in deciphering the languages and writings of our times. On the obverse side of this 2.8-inch wide nickel disk (shown in Fig. 2), eight major world languages are written in interlaced spirals starting at eye-readable size and gradually shrinking down to type requiring 1000x magnification to be read. By starting at a readable type size and reducing it as the text progresses, the disk cleverly shows the reader that there is more here if one is just able to magnify it. The reverse side of the disk is inscribed with 15,000 microetched pages of language and linguistic information. This tiny library archives information on over 2,500 languages. The plan is to create many copies of this artifact on the time-tested strategy that many copies will ensure the survival of at least a few.

Another text from our time that will probably outlast all others ever written is the famous metal plaque attached to the Pioneer 10 and 11 space probes. These documents are hurtling through the vacuum of space and will continue to do so for millions of years.

Figure 3. The plaque aboard Pioneer 10 and 11. This text is designed not only to bridge vast measures of time and distance, but facilitate cross-species communication.

But I am still interested to see what kinds of alternatives amateur scientists can devise to this highly sophisticated technology. After all, it's one thing for a foundation with lots of money and access to cutting edge technology to pull together an archive like the Rosetta Disk, but what about the rest of us? Most of the clay tablets from Mesopotamia do not contain grand literary works like the Campaigns of Sennacherib or the Epic of Gilgamesh. They are small pillow-shaped things recording business transactions, offerings made at a temple, or personal correspondence. The ordinary things in history form the context for the larger events. How are you going to tell someone 10,000 years from now about your famous five-alarm chili recipe, your favorite light bulb joke, or anything else about your corner of the world?

Resources

CD and DVD Longevity: How Long Will They Last?
http://www.audioholics.com/techtips/specsformats/CDDVDlongevity.php

Magnetic Tape Life Expectancy 10-30 years.
http://palimpsest.stanford.edu/bytopic/electronic-records/electronic-storage-media/bogart.html

The Long Now Foundation.
http://www.longnow.org/

The Rosetta Project.
http://www.rosettaproject.org