another not-April Fool's-post...

Damien Broderick (
Mon, 29 Mar 1999 13:31:59 +0000

...even though it looks like it. I repost here without permission, because the claim is so paradigm-busting and consequential if true that it must have vast implications for many of the topics we usually discuss, including brain function and routes to AI:

>From The Independent Review, March 19th, 1999:

           The memory of molecules

           Can molecules communicate with each other,
           exchanging information without being in physical
           contact? French biologist Jacques Benveniste believes
           so, but his scientific peers are still sceptical. By Lionel

           Jacques Benveniste was once considered to be one of
           France's most respected biologists, until he was cast
           adrift from the scientific mainstream. His downfall
           began in 1988 when he infuriated the scientific
           community with experimental results which he took
           as evidence to suggest that water has a memory. His
           ideas were seized upon by homeopaths keen to find
           support for their theories on highly diluted
           medicines, but were condemned by scientific purists.
           Now, Benveniste believes he has evidence to suggest
           that it may one day be possible to transmit the
           curative power of life-saving drugs around the world
           - via the Internet.

           It sounds like science fiction and Benveniste will
           have a hard time convincing a deeply sceptical world
           that he is right. Nevertheless, he began his campaign
           last week when he announced the latest research to
           come out of his Digital Biology Laboratory near Paris,
           to a packed audience of scientists at the Pippard
           Lecture Theatre at Cambridge University's Cavendish
           Physics Laboratory. Benveniste suggested that the
           specific effects of biologically active molecules such as
           adrenalin, nicotine and caffeine, and the
           immunological signatures of viruses and bacteria, can
           be recorded and digitised using a computer
           sound-card. A keystroke later, and these signals can be
           winging their way across the globe, courtesy of the
           Internet. Biological systems far away from their
           activating molecules can then - he suggested - be
           triggered simply by playing back the recordings.

           Most scientists have dismissed Benveniste as being
           on the fringe, although there were some famous
           names in the audience last week, including Sir
           Andrew Huxley, Nobel laureate and past president of
           the Royal Society, and the physicist Professor Brian
           Josephson, also a Nobel laureate. Benveniste started
           by asking some apparently childish questions. If
           molecules could talk, what would they sound like?
           More specifically, can we eavesdrop on their
           conversations, record them, and play them back? The
           answer to these last three questions is, according to
           Benveniste, a resounding "Oui!" He further suggested
           that these "recordings" can make molecules respond
           in the same way as they do when they react.
           Contradicting the way biologists think biochemical
           reactions occur, he claims molecules do not have to
           be in close proximity to affect each other. "It's like
           listening to Pavarotti or Elton John," Benveniste
           explained. "We hear the sound and experience
           emotions, whether they're live or on CD."

           For example, anger produces adrenalin. When
           adrenalin molecules bind to their receptor sites, they
           set off a string of biological events that, among other
           things, make blood vessels contract. Biologists say that
           adrenalin is acting as a molecular signalling device
           but, Benveniste asks, what is the real nature of the
           signal? And how come the adrenalin molecules
           specifically target their receptors and no others, at
           incredible speed? According to Benveniste, if the
           cause of such biochemical events were simply due to
           random collisions between adrenalin molecules and
           their receptors (the currently accepted theory of
           molecular signalling), then it should take longer than
           it does to get angry.

           Benveniste became the bete noire of the French
           scientific establishment back in 1988, when a paper he
           had published in the science journal Nature was later
           rubbished by the then editor, Sir John Maddox, and a
           team that included a professional magician, James
           Randi. With an international group of scientists from
           Canada, France, Israel and Italy, Benveniste had
           claimed that vigorously shaking water solutions of an
           antibody could evoke a biological response, even
           when that antibody was diluted out of existence.
           Non-agitated solutions produced little or no effect.
           Nature said that the results of the experiment that
           produced the "ghostly antibodies" were, frankly,
           unbelievable. The journal itself came in for criticism
           for publishing the paper in the first place.

           In his Nature paper, Benveniste reasoned that the
           effect of dilution and agitation pointed to
           transmission of biological information via some
           molecular organisation going on in water. This
           "memory of water" effect, as it was later known,
           proved Benveniste's academic undoing. For while
           the referees of his Nature paper could not fault
           Benveniste's experimental procedures, they could not
           understand his results. How, they asked, can a
           biological system respond to an antigen when no
           molecules of it can be detected in solution? It goes
           against the accepted "lock-and-key" principle, which
           states that molecules must be in contact and
           structurally match before information can be
           exchanged. Such thinking has dominated the
           biological sciences for more than four decades, and is
           itself rooted in the views of the 17th-century French
           philosopher Rene Descartes.

           Nature's attempted debunking exercise failed to find
           evidence of fraud, but concluded that Benveniste's
           research was essentially unreproducible, a claim he
           has always denied. From being a respected figure in
           the French biological establishment, Benveniste was
           pilloried, losing his government funding and his
           laboratory. Undeterred, he and his now-depleted
           research team somehow continued to investigate the
           biological effects of agitated, highly dilute solutions.
           The latest results are, for biologists, even more
           incredible than those in the 1988 Nature paper.
           Physicists, however, should have less of a problem as
           their discipline is based on fields (eg gravitational,
           electromagnetic) which have well-established
           long-range effects. If Benveniste's claims prove to be
           true - which is far from certain - they could have
           profound consequences, not least for medical

           Benveniste's explanation starts innocuously enough
           with a musical analogy. Two vibrating strings close
           together in frequency will produce a "beat". The
           length of this beat increases as the two frequencies
           approach each other. Eventually, when they are the
           same, the beat disappears. This is the way musicians
           tune their instruments, and Benveniste uses the
           analogy to explain his water-memory theory. Thus,
           all molecules are made from atoms which are
           constantly vibrating and emitting infrared radiation
           in a highly complex manner. These infrared
           vibrations have been detected for years by scientists,
           and are a vital part of their armoury of methods for
           identifying molecules.

           However, precisely because of the complexity of their
           infrared vibrations, molecules also produce much
           lower "beat" frequencies. It turns out that these beats
           are within the human audible range (20 to 20,000
           Hertz) and are specific for every different molecule.
           Thus, as well as radiating in the infrared region,
           molecules also broadcast frequencies in the same
           range as the human voice. This is the molecular
           signal that Benveniste detects and records.

           If molecules can broadcast, then they should also be
           able to receive. The specific broadcast of one
           molecular species will be picked up by another,
           "tuned" by its molecular structure to receive it.
           Benveniste calls this matching of broadcast with
           reception "co-resonance", and says it works like a
           radio set. Thus, when you tune your radio to, say,
           Classic FM, both your set and the transmitting station
           are vibrating at the same frequency. Twitch the dial a
           little, and you're listening to Radio 1: different
           tuning, different sounds.

           This, Benveniste claims, is how millions of biological
           molecules manage to communicate at the speed of
           light with their own corresponding molecule and no
           other. It also explains why minute changes in the
           structure of a molecule can profoundly alter its
           biological effect. It is not that these tiny structural
           changes make it a bad fit with its biological receptor
           (the classical lock-and-key approach). The structural
           modifications "detune" the molecule to its receptor.
           What is more, and just like radio sets and receivers,
           the molecules do not have to be close together for
           communication to take place.

           So what is the function of water in all this?
           Benveniste explains this by pointing out that all
           biological reactions occur in water. The water
           molecules completely surround every other molecule
           placed among them. A single protein molecule, for
           example, will have a fan club of at least 10,000
           admiring water molecules. And they are not just
           hangers-on. Benveniste believes they are the agents
           that in fact relay and amplify the biological signal
           coming from the original molecule.

           It is like a CD which, by itself, cannot produce a sound
           but has the means to create it etched into its surface.
           In order for the sound to be heard, it needs to be
           played back through an electronic amplifier. And just
           as Pavarotti or Elton John is on the CD only as a
           "memory", so water can memorise and amplify the
           signals of molecules that have been dissolved and
           diluted out of existence. The molecules do not have
           to be there, only their "imprint" on the solution in
           which they are dissolved. Agitation makes the

           So what do molecules sound like? "At the moment
           we don't quite know," says Didier Guillonnet,
           Benveniste's colleague at the Digital Research
           Laboratory. "When we record a molecule such as
           caffeine, for example, we should get a spectrum, but it
           seems more like noise. However, when we play the
           caffeine recording back to a biological system sensitive
           to it, the system reacts. We are only recording and
           replaying; at the moment we cannot recognise a
           pattern." "But," Benveniste adds, "the biological
           systems do. We've sent the caffeine signal across the
           Atlantic by standard telecommunications and it's still
           produced an effect."

           The effect is measured on a "biological system" such
           as a piece of living tissue. Benveniste claims, for
           instance, that the signal from molecules of heparin - a
           component of the blood-clotting system - slows down
           coagulation of blood when transmitted over the
           Internet from a laboratory in Europe to another in the
           US. If true, it will undoubtedly earn Benveniste a
           Nobel prize. If not, he will receive only more scorn.

           Benveniste's ideas are revolutionary - many might
           say heretical or misguided - and he is unlikely to
           persuade his most ardent critics. Although his ideas
           may seem plausible enough, he will win over his
           enemies only if his results can be replicated by other
           laboratories. So far this has not been done to the
           satisfaction of his many detractors. 


posted by

Damien Broderick