• Question: What is the biggest scientific mistake you've ever made?

    Asked by lumiereclair to Mark, Matthew, Mike, Paul, Sabina on 8 Mar 2014.
    • Photo: Sabina Hatch

      Sabina Hatch answered on 8 Mar 2014:

      Luckily so far I have not blown anything up or destroyed any equipment. But usually if you make a mistake in my research then it doesn’t create a big splash… unless you publish results that you find out were a mistake. In that case it could be a big problem!

      I made my biggest mistake probably during the time I was at the synchrotron near Oxford in a place called the Diamond Light Source (http://www.diamond.ac.uk/Home/About.html). Because so many scientists want to use this resource you have to apply for time. I managed to get 3days to do all my research. So I barely slept and tried to get as much data from my material as possible. I was using X-rays to look deep into my material (ZnO nanowires) and see whether I could determine the structure using the light that was emitted. It was the first time that this was ever tried! And it was a great success!! My mistake was that I was also supposed to take photoluminescence measurements for each sample… but I didn’t realise this at the time. I found out later and it meant that I wasted quite a lot of precious time going back and having to remeasure all the samples. It worked out ok in the end…I managed to get enough data to publish my results but it probably could have been much better if I had a bit more time.

    • Photo: Matthew Malek

      Matthew Malek answered on 8 Mar 2014:

      So far, I’ve been pretty fortunate in not having made any serious mistakes. Touch wood! Not that I’ve been perfect by any means, but all my errors have been small, technical details — probably too boring to describe in detail. It’s not like I’ve ever published findings that never needed to be retracted!

      That said, mistakes can be interesting things. When he developed his theory of gravity (General Relativity) in the early 1900s, Einstein added an extra term, which he called a “cosmological constant”. He believed that the universe was static — neither expanding nor contracting — and so this term added a repulsive force to counterbalance the attaction of gravity. In the 1930s, Edwin Hubble discovered the universe was NOT static, it was expanding. Einstein’s assumption had been wrong, so he ditched the cosmological constant and called it his greatest “blunder”.

      Well, fast forward six decades to 1998. Measurements of a certain type of supernova by two different groups showed that the acceleration of the universe is expanding, driven by something called “dark energy”. We don’t precisely know what it is yet… but it looks most likely to be Einstein’s cosmological constant! If that turns out to be the case, then Einstein’s greatest blunder was in saying that he had blundered! 😀

    • Photo: Mark Jackson

      Mark Jackson answered on 8 Mar 2014:

      There is a famous quote by Richard Feynman, “The easiest person to fool is yourself.” And apparently the easiest fool to fool is me.

      My collaborator and I had spent several years developing techniques to calculate quantum interactions during the Big Bang. For our next project, we realized that we could easily calculate something called a trispectrum which would be experimentally measured soon, allowing one of the first direct tests of fundamental physics at such high energies! So we calculated the trispectrum, which showed really interesting behavior, and consisted of three terms: I’ll call them A, B, and C. Thus the total trispectrum was A+B+C, and which was large enough to possibly be detected! We were thrilled.

      But this contradicted some predictions that other groups had made, and there was some controversy over the result. One of these other researchers – an expert in the field – insisted that our result must be wrong and suggested it was for a particular reason called boundary conditions. My collaborator and I quickly convinced ourselves that boundary conditions could not be the cause, and therefore we must be right. I hope you see the flaw in this logic. As the months went by, we continued working out details of our result.

      Finally after more time, and more controversy, I re-checked every line of our calculation again. I discovered that we had made a minus sign mistake, so that the answer was actually A-B-C. Yes, that’s right; there wasn’t any profound physics concept that we had misunderstood, we had just forgotten a minus sign. This wouldn’t necessarily be disastrous, but it also so happened that A = B+C, so the final answer was zero. We retracted our claim and notified the other groups. Even if it wasn’t for the reason they suspected, our answer had been wrong. It was still a useful result for some cases, but not as universal as we had originally hoped.

      What is the moral here? First, no matter how excited you may be about your result, you have to be your own harshest critic because someone else will eventually find the mistake if you don’t. Second, if there is any doubt about a result it’s best to check *before* you elaborate upon them, otherwise you’re just building on quicksand.

    • Photo: Paul Coxon

      Paul Coxon answered on 9 Mar 2014:

      Oh I’ve made plenty of mistakes during my research, the stupidest one which I can remember was when I was at MAXlab, the synchrotron facility in Sweden.

      I was measuring single photons of light in a film of nanocrystals, so all the detector apparatus had to be sealed inside a box, then covered in black cloth to shield if from all the light outside.

      We took a scan, and got a *very* strong light signal – GREAT we thought, the nanocrystals we’re looking at are really interesting, and carried on doing more measurements.

      After the experiment we took the apparatus apart, and I realised I’d left my mobile phone right next to the detector, and we’d spent the afternoon measuring the light from my phone screen!

      Luckily, we realised the mistake, and sorted it out.

Comments