Undiscovery in physics

By Marcus Wilson 28/11/2012 4


With the recent undiscovery of Sandy Island I’ve begun wondering what other things might be ripe for undiscovery. Wasps, for example. Wouldn’t it be great if we realized that there wasn’t actually any evidence for the existence of wasps after all. Their discovery had been just a mistake made by an entomologist back in the depths of history. We can all tell our children not to worry about them – they don’t exist. Our chickens would love to see the neighbour’s cat undiscovered (as would we – at least from our garden).  I’m sure a variety of places might feature strongly too. Hamilton is bound to be on the list of some people; but, I can assure you, the last time I looked it was still there.

I don’t think in physics there has been a great deal of undiscovery in the last few centuries. I struggle to think of any real undiscoveries.  Sure, there have been changes to our thinking. For example, relativity superseded Newtonian physics, but it would be wrong to say that Einstein undiscovered Newton’s Laws of motion. The latter are still a cornerstone of physics – but their applicability has been reduced to the realm where things aren’t travelling close to the speed of light. That would be more like discovering the coastline of Sandy Island is a bit different to what the maps have it. One might say that the Michelson-Morley experiment undiscovered the aether, but in reality the aether had never been discovered – it was just a well-accepted hypothesis. Likewise Joule’s experiments with heat put pay to the idea that heat was a fluid, but since no-one had claimed (supported by real evidence)  to have observed this fluid, it wasn’t really an undiscovery either.

Underlying modern science (by which I mean Galileo and beyond) is experimental evidence. No change in understanding of science, in any discipline, is going to happen without well collected and well analyzed data. This makes undiscovery of something (by which I mean overturning of some knowledge, theory or principle that has been believed based upon evidence, as opposed to mere hypothesis) unlikely. There have been a few instances of reports of new things that have been made prematurely, with unreliable evidence, such as cold fusion and faster-than-light neutrinos, and these have been embarrassing for the groups concerned and undiscovered very rapidly.  But undiscovery here has happened because they were never properly discovered in the first place.

That said, neither was Sandy Island properly discovered. My spell-checker’s underlining of the word undiscovery may be for good reason.

I’d love to hear readers’ thoughts on this one. Is there any piece of modern science that has been genuinely undiscovered?

 

 

 

 


4 Responses to “Undiscovery in physics”

  • Hmm

    There was a perception that cloning from adult animals was impossible until it succeeded with the sheep ‘Dolly’, but that really wasn’t a discovery or undiscovery. Just the constraints of our knowledge at the time.

  • It seems that a great deal of psychology is under thread of undiscovery. e.g. http://pps.sagepub.com/content/7/6.toc?etoc

    I get the impression that the culture in psychology is that if an experiment has been done, then there is no need to re-do it: attempts to replicate experiments are actively discouraged.

    A lot of results from medical science could be undiscovered too one day, though this is more a matter of research fraud on the part of the pharmaceutical companies. See Ben Goldacre’s new book _Bad Pharma_.

  • Ouch. That Psychology editorial by Pashler and Wagenmakers doesn’t hold back its criticism of the field. It wouldn’t hurt for scientists to take a good, hard look at ‘the way we do science’. If there’s a problem with rewarding researchers who find positive results over those who get (with equally good work) negative results, then it’s surely the scientists’ job to fix it. Sounds like material for another blog entry. I did this entry on fraud a couple of years back http://sci.waikato.ac.nz/physicsstop/2009/05/science-fraud.shtml

Site Meter