A Quantum Diaries Survivor

Bee and Stefan at backreaction have just published a very nice post listing 10 (well, 11) fairly well-known physics effects, from the photoelectric effect to the MSW effect. The descriptions are concise and matter-of-fact, with good references for further reading. Definitely a pleasant grab-and-go read.

An “effect”, a reader has argued in the comments thread of the post, should be something experimentally observed, and then (or contextually) explained by a theory. Instead, there are things such as Hawking radiation -the thermal emission from black holes- which is dubbed “Hawking effect” despite having never been observed (a fact that is likely to continue in the future).

Rather than discussing the largely nominalistic issue of what is in earnest an “effect” (which we can do at Bee’s and Stefan’s blog), I would like to elaborate here on the observation that science has become increasingly bold in the XXth century. Perhaps the cause of this effect is our getting accustomed to the routine exploration of realms to which our senses have utterly no access: very few XXIst century men and women would be willing to negate the existence of atoms, although we have never seen one (waiving a few electronic microscope pictures which arguably do not show anything directly either). Can we do the same with cosmic strings ? Surely not, we need proof. But what is an acceptable proof, these days ? If we abandon some of the foundations of scientific investigation - reproducibility, falsifiability, direct testing - through confidence in our means, strength of our prejudices, and mastery in the practice of our science, aren’t we becoming sorcerers ?

Once the foundations are gone, Science is in danger. We are approaching the dangerous terrain where to progress one is required faith. Faith in a theory, faith in conjectures, faith in methodologies. I see this trend quite clearly in particle physics, my research field. On one side, we have started during the last twenty years to peruse multivariate methods such as Neural Networks, which indeed work wonders but contain a good measure of magic within; as an example, CDF and D0 both have evidence in their Run II datasets of single top production  -a process which the standard model predicts with great accuracy, and thus must exist- with neural networks playing a major role. On the other side, theorists get enamoured of concepts such as fine tuning, renormalizability, unitarity - things that strictly speaking are not physical but mathematical arguments - to justify assumptions and build or kill theories.  

I know I am being a bit provocative here. But I have a point: I do see a trend. Humanity faces big challenges in the future, challenges for its own existence. Is global warming a prejudice or a scientific fact ? If we cannot even all agree on something like that, we are bound to extinction. And maybe there is some logic in it.

UPDATE: I hate when this happens. After writing something that I think has some originality, or at least some personality, I stumble in another recent post discussing quite similar matters in more length, more depth, and in front of a larger audience. It happened before, and it is awkward: I have to swear I did not read the other piece first… In any case, Sean’s piece is worth a look (I learned about it in a post at NEW - so I have to thank Peter for keeping us all updated on most of what is worth being aware of from around the web).