Ancient Texts as "Fossils": How They Survive

Stefan Lovgren
for National Geographic News

February 28, 2005

Through the ages, ancient texts have survived wars, fires, theft, and neglect.

But so far scholars have only been able to draw upon anecdotal evidence to estimate how many handwritten texts created before the advent of the printing press in the 15th century have survived.

A new study, however, uses population biology to calculate the likelihood that an ancient text has survived from the eighth or ninth century to the present.

"The basis of the model is that manuscripts are like organisms," said John L. Cisne, a paleontologist and evolutionary biologist at Cornell University in Ithaca, New York. "A manuscript is going to behave just like an individual in a population. It can divide and reproduce or it can die."

Cisne's calculations may help historians to better establish the quantity of knowledge and science that we have inherited from antiquity and the Middle Ages. Cisne's study suggests that certain works have a greater chance of survival than has been guessed from anecdotal evidence.

The research is published in the current issue of the journal Science.

Population Growth

Turbulent times such as the Viking invasions, the Reformation, and the French Revolution, claimed untold numbers of medieval manuscript copies of ancient texts. The percentage that escaped has intrigued manuscript scholars for decades.

The late Bernhard Bischoff, considered the foremost authority on medieval paleography (the study of old writings and inscriptions), estimated that one in seven manuscripts produced during the reign of European king Charlemagne and his heirs (780-810 A.D.) has survived to the modern era.

"That's a really well-educated guess," Cisne said. "[Bischoff] based it on an encyclopedic knowledge of all the manuscripts of early medieval Europe."

But instead of examining the Latin manuscripts themselves—a painstaking and training-intensive process performed by paleographers and codicologists (a sort of manuscript archaeologist)—Cisne suggests that equations from population biology can be used to estimate a text's likelihood of survival.

Cisne treated handwritten copies of ancient texts like fossils from an extinct population. He studied the Saint Bede the Venerable's De Temporum Ratione (A.D. 725), a standard early medieval textbook on arithmetic and calendar calculations. Popular from the 8th to the 16th century, the text's finger-counting chapter for years served as an instruction manual for low-tech "pocket calculators."

The models constructed by Cisne show that the number of Bede manuscripts grows exponentially as they are copied, similar to a population of living organisms. "The more individuals there are, the faster the population can grow," Cisne said.

The rate of growth slows as the population reaches its carrying capacity. For living organisms, this would be largely the result of increased competition for resources. For manuscripts, it is the demand for the manuscript and its rate of destruction.

According to Cisne's calculations, the number of surviving Bede manuscripts is close to two in seven. The data suggests that the percentage of manuscripts that survived from antiquity and the Middle Ages could be higher than previously thought.

Families of Texts

As a paleographer and historian of medieval medicine, Eliza Glaze, the co-author of an accompanying commentary in Science, says the new approach will inspire manuscript scholars to look at their work in a new light.

"This is an example of a brilliant paleontologist, who has discovered a whole new range of possibilities for using the models he's already familiar with," said Glaze, who is an assistant professor of medieval history at Coastal Carolina University in Conway, South Carolina.

"None of the textual historians have done this yet," she said. "We tend to look at individual manuscripts and what happened to them in particular circumstances. We don't usually think of the texts as populations, though we do speak of them as belonging to families, based upon their content."

Glaze stresses, however, that textual scholars focus on the quality of the text conveyed by each manuscript, rather than the quantity of texts available.

From Papyrus to Parchment

The leap from papyrus to parchment in late antiquity was one crucial element in the survival of ancient scientific texts in Europe.

"Other factors determining their survival included their availability in a familiar language, like Latin rather than Greek, the level of their practical utility, the clarity of their script, and cultural factors like the popularity or appeal of their content," Glaze said.

Cisne, the study author, also calculates the ultimate probability of survival for a text—the likelihood that sometime in the future there will be at least one copy surviving. "The number [that] one comes up with for that is somewhere around 90 to 95 percent," he said.

That, he says, means we probably have a better selection of the texts that were actually available in the Middle Ages than previously thought. "We probably know about most of the works that were at all popular," Cisne said.

Glaze, however, bemoans the complete loss of some important medical texts, which are known only by title or references that survive in the writings of other authors.

"We'd give just about anything to have those texts back again," she said. "As it is, fragmentary quotations of lost texts, like the isolated bones of extinct and unknown species, are sometimes all that survive."

Don't Miss a Discovery
Sign up our free newsletter. Every two weeks we'll send you our top news by e-mail (see sample).