PHILANTHROPY A century of the

COMMENT BRAIN A history of HM, the SPACE Buzz Aldrin’s OBITUARY Remembering
Rockefeller Foundation, most famous patient in blueprint for colonizing Robert Edwards, pioneer
inventor of the grant p.311 neuroscience p.313 Mars reviewed p.314 of in vitro fertilization p.318
PETE ELLIS/DRAWGOOD.COM

Don’t bristle at blunders

Embrace mistakes, urges Mario Livio — they are portals to scientific progress.

I
n a July 1991 Nature paper1, astrono- had not corrected adequately for Earth’s time Lyne withdrew his result, Wolszczan
mers Andrew Lyne, Matthew Bailes motion around the Sun. Lyne’s revelation had performed enough tests to be certain.
and S. L. Shemar made an electrifying of the blunder at a meeting of the Ameri- Blunders are an essential part of the
announcement: the discovery of the first can Astronomical Society that month won scientific process. Research is not a lin-
planet outside our Solar System. To every- him a standing ovation. But the story had a ear march to the truth but a zigzag path,
one’s surprise, it was not orbiting a Sun-like happier ending. involving trial and error. Mistakes are not
star but a pulsar — the dense, spinning Immediately after Lyne’s presenta- the exclusive province of sloppy or inexpe-
neutron-star offspring of a supernova explo- tion, astronomer Aleksander Wolszczan rienced scientists. Even the brightest lumi-
sion. The putative planet gave itself away announced that he and his colleague naries — including Charles Darwin and
by altering the period of radio-frequency Dale Frail had discovered two planets orbit- Albert Einstein — made serious blunders.
flashes given off by the pulsar. ing another pulsar using the same technique. Truly innovative ideas require a willing-
Unfortunately, Lyne and Bailes had to These turned out to indeed be the first dis- ness to embrace risks, and acceptance of the
retract this result a few months later after coveries of extrasolar planets. Wolszczan fact that errors can be portals to progress.
uncovering an error, which they reported2 told me that Lyne’s original paper had acted Although this is well known in some pri-
in Nature in January 1992. The astrono- as a “confidence booster”, convincing him vate companies engaged in research and
mers courageously announced that they that the signals in his data were real. By the development, academics today are slow

1 6 M AY 2 0 1 3 | VO L 4 9 7 | NAT U R E | 3 0 9
© 2013 Macmillan Publishers Limited. All rights reserved
 COMMENT

in recognizing the necessity of blunders.

Chemist Linus Pauling knew it. His for- H I STORY R E V I S I T E D
mer postdoc, Jack Dunitz, recalls being told:
“Mistakes do no harm in science because Did Einstein ever say “biggest blunder”?
there are lots of smart people out there who
will immediately spot a mistake and correct Almost any history of Albert Einstein’s George Gamow in an article published in
it. You can only make a fool of yourself and ‘cosmological constant’ mentions his the September 1956 issue of Scientific
that does no harm, except to your pride. If it “biggest blunder”— the introduction of American. Gamow later repeated the story
happens to be a good idea, however, and you this constant to counteract gravity into in his 1970 autobiography, My World Line.
don’t publish it, science may suffer a loss.” equations characterizing the Universe. Einstein was indeed unhappy about
Did Einstein actually say this? After having introduced the cosmological
KNOTTY PROBLEM scrutinizing dozens of documents while constant, saying in a letter to cosmologist
Preposterous ideas can lead to important researching my book Brilliant Blunders Georges Lemaître that he was “unable to
insights. In 1867, the eminent physicist (Simon & Schuster, 2013), I found no believe that such an ugly thing should be
William Thomson (Lord Kelvin) proposed3 evidence that he did. realized in nature”. Calling it the “biggest
that atoms were not point-like but ‘knotted The “biggest blunder” phrase seems blunder” was, in my view, Gamow’s
vortex tubes of the ether’. Ether was the sup- to have come from the colourful physicist hyperbole.
posed fluid that pervaded space, providing a
medium for electricity and magnetism.
Inspired by work on vortices in fluids by and recombine the four ends in various ways, impact-driven atmosphere? I believe it must.
the nineteenth-century German physicist which can be described using knot theory. We should make space for risky scientific
Hermann von Helmholtz, Kelvin identified proposals in grant and evaluation processes.
three characteristics of knotted vortex tubes EXTRAORDINARY CLAIMS Until a decade ago, the committees that
that made them attractive models for atoms. Blunders are sometimes hard to correct. allocated observing time on the Hubble
First, vortices in fluids were astonishingly Modern experiments can be so intricate Space Telescope were encouraged to give
stable — mirroring to Kelvin the “unalter- and require such big investments in time up to 10% of the time to proposals with a
able distinguishing qualities” of atoms — and and funds that replicating them becomes low probability of success but potentially
each knot could be classed according to its prohibitive. When a result is widely assumed high return. A similar philosophy could be
geometrical properties. Second, the variety of to be wrong, few scientists are motivated to adopted more widely.
chemical elements could reflect the “endless repeat the work. One problem is that committees tend not
variety” of knots. Finally, just as smoke rings But there can be rewards for doing so. The to approve risky programmes. Efforts to reach
vibrate, the oscillations of ether vortex tubes sensational claim4 in Science by geomicro- consensus converge to a mean. Such obstacles
might produce atomic spectral lines. biologist Felisa Wolfe-Simon and her col- can be overcome if decisions are left to one
To explain the periodic table, Kelvin leagues to have discovered a bacterium that person. In the case of Hubble, a pool of ‘direc-
needed to classify knots according to their substitutes arsenic for phosphorus to sustain tor’s discretionary time’ on the telescope is
forms, discarding any that could be manipu- its growth brought a wave of criticism. available, for which anyone can apply. From it
lated from one to another. In Kelvin’s theory, A few critics checked the experiment, came the Hubble Deep Field, one of the most
the circular ‘unknot’ represented the hydro- including microbiologist Rosemary Redfield detailed images of the Universe ever made.
gen atom; the triple-looped ‘trefoil’, carbon. at the University of British Columbia in Today, telescopes including Hubble are
Kelvin’s theory of vortex atoms is obvi- Vancouver, Canada, who blogged the pro- turned towards addressing the profound
ously wrong. The ether does not even exist. cess (see go.nature.com/bmb62d). The outcome of another ‘blunder’. Einstein
But these failures did not deter everyone. effort proved fruitful, showing that the bac- regretted his attempt to model a static cos-
Whereas physicists lost interest for a while, terium goes to great lengths to dodge arsenic. mos using a repulsive-gravity force (see ‘Did
knots began to intrigue mathematicians, Redfield and her colleagues detected no arse- Einstein ever say “biggest blunder”?’). But
becoming an active area of study for decades. nic in the bacterium’s DNA to much lower since it was revealed by supernovae obser-
In the 1980s, knot theory reconnected with limits than in the original paper. Molecular vations in 1998 that our Universe is accel-
physics. Mathematician Vaughan Jones dis- biologist Dan Tawfik and his team at the erating, understanding the nature of that
covered an algebraic expression that is unique Weizmann Institute of Science in Rehovot, repulsive force is one of the biggest chal-
for every knot. Physicist Edward Witten Israel, identified the mechanism by which lenges that physics faces today.
linked it to quantum field theory, the branch some of the proteins of this and related bac- Researchers must embrace blunders
of physics that describes fields and the suba- teria bind to phosphate and not to arsenate that come from thinking outside the box.
tomic world. In classical physics, the path of Although one lesson is obvious — extraor- Evaluation processes should allow for orig-
a particle travelling from point A to point B dinary claims require extraordinary evidence inality, even at the risk of false starts and
is determined by Newton’s laws of motion. In — the original paper still had some scientific blind alleys. ■
the quantum regime, one has to consider all value. It stimulated discussion and inspired
the possible paths connecting A to B, includ- curiosity about different possibilities for life. Mario Livio is an astrophysicist at the
ing winding and knotted ways. In the nineteenth century, Scottish author Space Telescope Science Institute, Baltimore,
Subsequent work linked knots, quantum Samuel Smiles wrote: “We often discover Maryland, USA. His book Brilliant Blunders
field theory and string theory, which by what will do, by finding out what will not (Simon & Schuster) is released this month.
describing particles as vibrations of strings, do; and probably he who never made a mis- e-mail: mlivio@stsci.edu
harks back to Kelvin’s idea. Today, knots are take never made a discovery.” His statement 1. Bailes, M., Lyne, A. G. & Shemar S. L. Nature 352,
used in chemistry and biology to analyse the should not be taken as advocacy for slapdash 311–313 (1991).
actions of enzymes on DNA molecules. In a science but as an encouragement to think 2. Lyne, A. G. & Bailes, M. Nature 355, 213 (1992).
process known as site-specific recombina- originally and take calculated risks. 3. Thomson, W. Proc. R. Soc. Edinb. 6, 94–105
(1867).
tion, enzymes align segments of the genetic Can research failure be accommodated 4. Wolfe-Simon, F. et al. Science 332, 1163–1166
sequence, cut the two strands of DNA open in today’s fast-paced, funding-starved, (2011).

3 1 0 | NAT U R E | VO L 4 9 7 | 1 6 M AY 2 0 1 3
© 2013 Macmillan Publishers Limited. All rights reserved