A Three generations ago, 180 young women wrote essays describing why they wanted to join a convent, a religious community of nuns. Years later, a team of psychological researchers came across these autobiographies in the convent’s archives. The researchers were seeking material to confirm earlier studies hinting at a link between having a good vocabulary in youth and a low risk of Alzheimer’s disease in old age. What they found was even more amazing. Although the young women were in their early twenties when they wrote their essays, the emotions expressed in these writings predicted how long they would live: those with upbeat autobiographies lived more than ten years longer than those whose language was more neutral. Deborah Danner, a psychologist at the University of Kentucky who spearheaded the study, noted that the results were particularly striking because all members of the convent lived similar lifestyles, eliminating many variables that normally make longevity studies difficult to interpret. ‘It was a phenomenal finding,’ she says. ‘A researcher gets a finding like that maybe once in a lifetime.’ However, she points out that no one has been able to determine why positive emotions might have such life-extending effects.

B Barbara Fredrickson, Professor of Psychology at the University of Michigan, believes that part of the answer is the ‘undo effect’. According to this theory, positive emotions help you live longer by shutting down the effects of negative ones. Fredrickson’s theory begins with the observation that negative emotions, such as fear and stress, enhance our fight-or-flight response to very real threats. However, even when the emergency is over, negative emotions produce lingering effects. Brooks Gump, a stress researcher at the State University of New York, explains that one of these effects is excessive cardiovascular reactivity. Behaviourally, Gump says, this reactivity is related to excessive vigilance: the state of being constantly on guard for potential dangers. Not only is it physically draining to live in a perpetual state of high vigilance, but high cardiovascular reactivity could be linked to increased chances of a heart attack.

C Fredrickson believes positive emotions work their magic by producing a rapid unwinding of pent-up tension and restoring the system to normal. People who quickly bounce back from stress often speed the process by harnessing emotions such as amusement, interest, excitement and happiness, she says. To test her theory, Fredrickson told a group of student volunteers that they had only a few minutes to prepare a speech that would be critiqued by experts. After allowing the students to become nervous, Fredrickson told them they would not actually have to deliver their speeches. She monitored their heart rates and blood pressure. Not surprisingly, all the students became nervous, but those who viewed the experiment with good-humoured excitement saw their heart rates return to normal much more quickly than those who were angry about being fooled. In a second experiment, Fredrickson reported that even those who were normally slow to bounce back could be coached to recover more quickly by being told to view the experiment as a challenge rather than a threat.

D Fredrickson believes that positive emotions make people more flexible and creative. Negative emotions, she says, give a heightened sense of detail that makes us hypersensitive to minute clues related to the source of a threat. But that also produces tunnel vision in which we ignore anything unrelated to the danger. Fredrickson speculated that, just as positive emotions can undo the cardiovascular effects of negative ones, they may also reverse the attention-narrowing effects of negative feelings by broadening our perspectives.

E To verify her theory, Fredrickson showed a group of students some film clips: some saw frightening clips, while others saw humorous or peaceful ones. They then completed a matching test in which they were shown a simple drawing and asked which of two other drawings it most resembled. The drawings were designed so that people would tend to give one answer if they focused on details and another if they focused on the big picture. The results confirmed Fredrickson’s suspicion that positive emotions affect our perceptions. Students who had seen the humorous or peaceful clips were more likely to match objects according to broad impressions.

F This fits with the role that positive emotions might have played in early human tribes, Fredrickson says. Negative emotions provided focus, which was important for surviving in life-or-death situations, but the ability to feel positive emotions was of long-term value because it opened the mind to new ideas. Humour is a good example. She says: ‘The emotions are transient, but the resources are durable. If you build a friendship through being playful, that friendship is a lasting resource.’ So while the good feelings may pass, the friendship remains. On an individual level, Fredrickson’s theory also says that taking time to do things that make you feel happy is not simply self-indulgent. Not only are these emotions good for the individual, but they are also good for society.

G Other researchers are intrigued by Fredrickson’s findings. Susan Folkman, of the University of California, has spent two decades studying how people cope with long-term stresses such as bereavement or caring for a chronically ill child. Contrary to what one might expect, she says, these people frequently experience positive emotions. These emotions are not there by accident, she adds. ‘Mother Nature doesn’t work that way. I think that they give a person time out from the intense stress to restore their resources and keep going. This is very consistent with Fredrickson’s work.’

The starkly modern Beinecke Library at Yale University is home to some of the most valuable books in the world: first folios of Shakespeare, Gutenberg Bibles and manuscripts from the early Middle Ages. Yet the library’s most controversial possession is an unprepossessing vellum manuscript about the size of a hardback book, containing approximately 240 pages of drawings and text of unknown age and authorship. Catalogued as MS 408, the manuscript would attract little attention were it not for the fact that the drawings hint at esoteric knowledge, while the text seems to be some sort of code—one that no one has been able to break. It is known to scholars as the Voynich manuscript, after the American book dealer Wilfrid Voynich, who bought the manuscript from a Jesuit college in Italy in 1912.

Over the years, the manuscript has attracted the attention of everyone from amateur dabblers to top code-breakers, all determined to succeed where countless others have failed. Academic research papers, books and websites are devoted to making sense of the manuscript’s contents, which are freely available to all. ‘Most other mysteries involve second-hand reports,’ says Dr Gordon Rugg of Keele University, a leading Voynich expert. ‘But this is one that you can see for yourself.’

It is certainly strange: page after page of drawings of weird plants, astrological symbolism and human figures, accompanied by a script that looks like some form of shorthand. What does it say—and what are the drawings about? Voynich himself believed that the manuscript was the work of the 13th-century English friar Roger Bacon, famed for his knowledge of alchemy, philosophy and science. In 1921, Voynich’s view that Bacon was the writer appeared to win support from the work of William Newbold, Professor of Philosophy at the University of Pennsylvania, who claimed to have found the key to the cipher system used by Bacon. According to Newbold, the manuscript proved that Bacon had access to a microscope centuries before microscopes were supposedly invented. The claim that this medieval friar had observed living cells created a sensation. It soon became clear, however, that Newbold had fallen victim to wishful thinking. Other scholars showed that his ‘decoding’ methods produced a host of possible interpretations.

The Voynich manuscript has continued to defy the efforts of world-class experts. In 1944, a team was assembled to tackle the mystery, led by William Friedman, the renowned American code-breaker. They began with the most basic code-breaking task: analysing the relative frequencies of the characters making up the text and looking for signs of an underlying structure. Yet Friedman’s team soon found themselves in deep water. The precise size of the ‘alphabet’ of the Voynich manuscript was unclear: it is possible to make out more than 70 distinct symbols among the 170,000-character text. Furthermore, Friedman discovered that some words and phrases appeared more often than expected in a standard language, casting doubt on claims that the manuscript concealed a real language, as encryption typically reduces word frequencies.

Friedman concluded that the most plausible resolution of this paradox was that ‘Voynichese’ is some sort of specially created artificial language, whose words are devised from concepts rather than linguistic principles. So could the Voynich manuscript be the earliest known example of an artificial language? ‘Friedman’s hypothesis commands respect because of the lifetime of cryptanalytic expertise he brought to bear,’ says Rob Churchill, co-author of The Voynich Manuscript. That still leaves a host of questions unanswered, however, such as the identity of the author and the meaning of the bizarre drawings. ‘It does little to advance our understanding of the manuscript as a whole,’ says Churchill.

Even though Friedman was working more than 60 years ago, he suspected that major insights would come from using the device that had already transformed code-breaking: the computer. In this he was right—it is now the key tool for uncovering clues about the manuscript’s language.

The insights so far have been perplexing. For example, in 2001 another leading Voynich scholar, Dr Gabriel Landini of Birmingham University in the UK, published the results of his study of the manuscript using a pattern-detecting method called spectral analysis. This revealed evidence that the manuscript contains genuine words rather than random nonsense, consistent with the existence of some underlying natural language. Yet the following year, Voynich expert René Zandbergen of the European Space Agency in Darmstadt, Germany, showed that the entropy of the text, a measure of the rate of information transfer, was consistent with Friedman’s suspicions that an artificial language had been used.

Many are convinced that the Voynich manuscript is not a hoax. For how could a medieval hoaxer create so many tell-tale signs of a message from random nonsense? Yet even this has been challenged in new research by Rugg. Using a system first published by the Italian mathematician Girolamo Cardano in 1550, in which a specially constructed grille is used to pick out symbols from a table, Rugg found that he could rapidly generate text with many of the basic traits of the Voynich manuscript. Publishing his results in 2004, Rugg stresses that he had not set out to prove that the manuscript was a hoax. ‘I simply demonstrated that it is feasible to hoax something this complex in a few months,’ he says.

Inevitably, others beg to differ. Some scholars, such as Zandbergen, still suspect that the text has genuine meaning, though they believe it may never be decipherable. Others, such as Churchill, have suggested that the sheer weirdness of the illustrations and text hints at an author who had lost touch with reality.

What is clear is that the book-sized manuscript, kept under lock and key at Yale University, has lost none of its fascination. ‘Many derive great intellectual pleasure from solving puzzles,’ says Rugg. ‘The Voynich manuscript is as challenging a puzzle as anyone could ask for.’