You Know Too Much

Most of our exploding collection of science information is junk.

By Seth Lloyd|Saturday, April 28, 2007

Is the world about to drown in science? The demographics lookdaunting. In 1800 there were perhaps a thousand people worldwide whocould properly be called scientists. Nowadays, there are millions. Atthis rate, there will be a billion scientists on Earth by 2200. Thetrends in scientific publishing are equally stunning, indicatingexponential growth. Scientists are multiplying like rabbits, with noElmer Fudds to keep them—or scientific knowledge—in check.

That runaway growth has given rise to fears that science is takingover (just watch any Hollywood science-fiction potboiler to see howdeep-seated those feelings are). The fears are based on twoacknowledged features of science. First, it is cumulative: The factsand papers published in thousands of scientific journals just keeppiling up. The leading ones, Science, Nature, and Proceedings of theNational Academy of Sciences, generated more than 4,000 peer-reviewedarticles last year. Second, it is opportunistic: Scientists areconstantly finding ways to make previously unscientific fieldsscientific. The development of the scientific history of the universe,which now threatens religious creation myths, is just one example.

Does this mean that nonscientific forms of knowledge are in danger?Do we face a world in which we will become “Vulcanized” so that, likeMr. Spock in Star Trek, we operate only at the level of rationality andscientific knowledge? Will nonscientific beliefs gradually fade,leading to a brave new world in which all affairs are conductedaccording to cold scientific principles?

Fat chance! And the reason is simple. The vast majority ofscientific ideas are (a) wrong and (b) useless. The briefestacquaintance with the real world shows that there are some forms ofknowledge that will never be made scientific. Students at MIT, forexample, are well known for hacking any process known to man or woman.If there were a scientific method for getting a date on Saturday night,they would have found it. Trust me, they have not. The true challengeposed by the runaway growth in information is to unearth useful bitsfrom this mountain of dross. The faster knowledge piles up, the harderit is to find something new that is actually worth knowing.

I would bet that 99.8 percent of ideas put forth by scientists arewrong and will never be included in the body of scientific fact. Overthe years, I have refereed many papers claiming to invalidate the lawsof quantum mechanics. I’ve even written one or two of them myself. Allof these papers are wrong. That is actually how it should be: Whatmakes scientific ideas scientific is not that they are right but thatthey are capable of being proved wrong. Nobel laureate Murray Gell-Mannis fond of saying, “The job of a scientist is to generate wrong ideasas fast as possible.”

And what of that 0.2 percent of ideas that turn out to be correct?The great majority of them are relatively useless. In those thousandsof scientific journals, the average article is subsequently cited ahandful of times if at all. The lion’s share of citations are gleanedby a few pathbreaking papers. For example, the paper by Eric Lander andhis colleagues that announced the sequencing of the human genome in2001 has been cited 5,356 times so far. By contrast, the hundredth mostcited article on the human genome has been cited a few hundred times.The great majority of the 34,510 papers on the human genome listed inthe Science Citation Index have been cited fewer than 10 times; onesthat were never cited are not even listed.

In other words, the amount of useful scientific knowledge is farsmaller than the number of journals would suggest. Some of my ownresults have accumulated fewer than a dozen citations.

Making the initial discovery in science is like striking gold:Nuggets lie around just for the taking. As more researchers enter thefield, though, heavy shoveling is required to unearth good results.Once the field is mature, large-scale blasting and leaching with themental equivalent of cyanide are required to separate out a few gramsof the worthwhile from the tons of dross. When you enter a scientificfield in which thousands of smart people have been toiling for decades,your chances of striking the mother lode of discovery are tiny.

Comment on this article

Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar

Log in to your account

Email address:
Remember me
Forgot your password?
No problem. Click here to have it emailed to you.

Not registered yet?

Register now for FREE. It takes only a few seconds to complete. Register now »