THE standard keyboard that was invented in 1872 is known worldwide as the QWERTY after the first six letters in the top line of letter keys. The American naturalist Stephen Jay Gould relates in his collected essays Bully for Brontosaurus (W.W. Norton, 1991) that the characters on the QWERTY keyboard were deliberately set to be inconvenient, thus ensuring slower typing speeds. The reason was simple. Typists using the earliest mechanical typewriters could reach such high speeds that the keys were frequently jamming. Unfortunately, the effect could not be seen by the typists until they removed the page whereupon the entire sheet had to be retyped.
Subsequently, as Gould puts it, by some strange 鈥渢echnological continuity law鈥� in common with historical and biological evolution, the QWERTY keyboard survived into the age of electronic keyboards, despite the fact that the jamming problem was no longer relevant. All recent attempts to create a mass market for more efficient keyboards 鈥� for example, the Dvorak keyboard, on which typists can achieve touch typing speeds about 40 per cent faster than on QWERTY keyboards 鈥� were stonewalled.
This QWERTY story is a classic example of a 鈥渓ock-in鈥� effect. It brings to mind the inevitable failure of the 鈥渂etter mousetrap鈥�. The sheer inertia of existing devices and habits often makes it virtually impossible to switch to obviously more efficient (and often readily available) solutions. A Canadian biochemist Donald Forsdyke has shown in an essay entitled 鈥淥n the giraffes and peer review鈥� that, contrary to intuition, 鈥渄esign by evolution鈥� is frequently very far from being optimal (Federal American Societies for Experimental Biology Journal, vol 7, p 619, 1993). Due to widespread and entrenched use, designs that are less than the best often persist because society is unwilling to change them.
Advertisement
An example of such a 鈥渟uboptimal鈥�, but almost universally accepted, practice is the system of anonymous peer review in science. Many scientists on many occasions have attacked the unreliability of anonymous peer review. This unreliablity is especially evident in the dismally unsuccessful record of anonymous peer review in 鈥渇orecasting鈥� the impact of a grant proposal. For example, in the 1960s the Canadian chemist John Polanyi had his paper on molecular reactions rejected by Physical Review Letters as 鈥渉aving no scientific interest鈥�. Any yet, in 1986 he won a Nobel Prize for Chemistry for the work.
Anonymous peer review has proliferated internationally to the point of no return. What is most notorious about anonymous peer review is its central thesis that 鈥減eople give more honest opinions when protected by anonymity鈥�. Such cant has no justification whatsoever. Anonymous letters to the editor are not published. Anonymous criticism generally doesn鈥檛 gain our respect. Even in special circumstances (for example, the protection of informers), we distrust a veil of anonymity, and demand independent evidence. In all other areas of human endeavour (the arts, sports, politics) criticism, even the most harsh, is invariably open. Only science, by some strange twist of its history, has developed anonymous peer review, to its own detriment.
Anonymous peer review has become as much a stumbling block to scientific innovation as the QWERTY keyboard is a stumbling block to fast typing. J.W. Dauben in his biography, Georg Cantor: His Mathematics and Philosophy of the Infinite (Princeton University Press, 1979), claims that Cantor, who lived from 1845 to 1918, was the creator of transfinite set of theory and was one of the most imaginative and controversail figures in the history of mathematics. 鈥淏ecause his views were unorthodox, they stimulated lively debate and at times vigorous denunciation. Leopold Kronecker considered Cantor a scientific charlatan, a renegade, a 鈥榗orrupter of youth鈥� 鈥� Henri Poincar茅 thought set theory and Cantor鈥檚 transfinite numbers represented a grave mathematical malady, a perverse pathological illness that would one day be cured.鈥�
One wonders just what sort of research grant Cantor would receive today if Kronecker and Poincar茅 were sitting as members of a grant committee, or acting as anonymous peer reviewers. The irony of history is that Cantor and Poincar茅 are both seen as forerunners of the modern theories of chaos and fractals. Cantor is best remembered for his famous 鈥淐antor set鈥� (an infinitely complex object with zero density), and Poincar茅 for his first formulation of the paradigm of the 鈥渂utterfly effect鈥� (the high sensitivity of dynamic systems to initial conditions).
We need alternative forms of scientific communication and different schemes for funding research that are not tangled up with anonymous peer review. The simplest way would be for scientists to refuse to remain anonymous when acting as a referee. We call on all our colleagues to send signed versions of their reports/criticisms directly to the authors of the manuscripts and grant proposals that they review. Each individual, positive act of this kind will begin to replace anonymous peer review by a better system.
If we all take responsibility for our opinions, and act with greater thoughtfulness and openness to our colleagues, the pernicious effects of secrecy and the old boy networks will gradually be broken down.