Falsifiability and Messy Science - Sixty Symbols

A chat with Professor Phil Moriarty on The Scientific Method, inspired by a Sean Carroll paper.
More links and info below ↓ ↓ ↓

Social Physics by Alex Pentland

This project features scientists from The University of Nottingham

Sixty Symbols videos by Brady Haran

File footage courtesy of AP Archive:

[zooblestyx]

You think science is messy now? In 1771, what astronomy had was even Messier.

[cavalrycome]

I think we need to distinguish clearly between the messy creative process of coming up with scientific theories, a process which depends a lot on social factors and chance, versus the more structured and systematic process of testing them, which doesn't. The latter is what we normally mean by 'the scientific method'. The discovery of graphene is no exception. No one would have believed that graphene existed without a systematic followup to that messy cellotape business.

[ozdergekko]

"I go play" is so true. I was a research scientist in biochemistry and molecular biology and in all honesty I did it to play around. It was very satisfying for some years.

[danieljensen]

I think his conclusion at the end could have been more clear. In fact in general this seems like a fun conversation but not necessarily a well structured or thought out video. I'd say falsifiability doesn't necessarily happen on a large scale, but it should on the scale of individual experiments, especially when you're "poking and prodding" as he said. You poke and prod until something interesting happens, you come up with a theory about what is causing it to happen, and then you do everything you can to prove that it actually isn't happening at all and it's just random chance or you messed something up. So maybe we could just add the poking and prodding bit to the beginning of the scientific method. This is worth noting because most science doesn't start with a theory, and even the theories we have all come from previous experiments.

[pcdsgh]

One of the most enjoyable science dialogue I've seen in a while. Thanks! It's refreshing to see something as candid as this.

[tennisdude]

I respect his point of view. But when viewed by people with little to no scientific education, this video could be massively misinterpreted. It would be truly sad if this video was ever used by anti-vaxers or climate change deniers.

[JBLewis]

One of the most mind bending classes I took in my Surveying major was Error Analysis, and coming to grips with the idea that we can't actually measure anything "very well", and the noble pursuit is to work to understand, identify, and quantify the errors in our measurements.

[neurotoxi]

I'm a neuroscientist and the passion that this professor shows for his field inspires me to do better at mine. Thank you.

[Hecatonicosachoron]

I'm so glad that this video has been made!!! It's spot on. Falsifiability is very seldom the motivation behind actively doing science. Now, what gets in the textbooks in experimental sciences has survived experimental scrutiny. But the fact that it has not (yet) been falsified is always noted after the fact.
Laypersons interested in the sciences must realise that there are at least three levels where scientific ideas live. First, there are the textbooks. These tend to represent the consensus at the time when they were written. Then there's the recent literature (recently published papers) - that's the material that is either confirming what's in the textbooks or is on the waiting list for getting into the textbooks. Finally there's the realm of active research. There, anything goes. Active researchers can do anything they find interesting or expedient. It doesn't have to be with a particular hypothesis in mind or with the intention of falsifying a theory.
There are restrictions of course. Usually researchers will do either what they have outlined in a research proposal that has been approved, or follow the instructions of the principal investigator, or work along the general direction that the research group they belong to. But the intention of falsifying a hypothesis is optional - and usually absent.
It's about time to leave popperism behind and admit that research is more flexible and a great deal freer and messier.

[jamesbrowne]

This should be the start of a whole new series on the philosophy of science, with a new cast of characters. I read Popper's books as an undergraduate, and a stack of literature on this debate since then. We even had a seminar course on the topic in grad school. The question here is not the playing and exploration in the lab or out in the field, but rather the structure of the thought process. Are you thinking in terms of a tautology, or is it possible to unroll the question or observation into a form that could even be false or even measurable at all? In one case, Popper declared evolution a pseudo science because the survival of the fittest reduces to the survival of the survivors, an untestable tautology. A research group took up the challenge and defined the measurable and testable approach of inclusive fitness. This becomes the relative number of descends or relatives, not the strength of an individual. Popper recanted after that was published.

Much more could be, and should be, said about this topic. This goes far beyond physics into all other forms of science and logic.

[lorenzocapitani]

Love videos with Prof.Moriarty, more please!!

[kityer]

Such a great video! I think this applies to anything in life, and underlines what might be the source of all misunderstandings in the world - people have different perceptions of the world.

[quahntasy]

We need more informative and long videos like these! Especially the one discussing a paper.

[Rumble-Tusk]

People need to learn how to say, "I don't know." rather than simply believing in things they don't understand or even worse, disparaging people who don't believe in that thing that you yourself don't understand. Just say you don't know. The ultimate scientific statement. It's where science starts. Just repeat after me: "I don't know."

[BillySugger]

I really enjoy Prof Phil’s presentations. I’m just a bit concerned that a lot of people who don’t follow the nuances and detail of what he says will make mischief out of it.

[mitchstilborn]

This might just be my favourite video you've ever done.

[stewiegriffin]

Disable audio, and imagine him saying "Miss Stevenson, my dog ate my homework" at 0:32

[EmilianoHeyns]

The graphene example is wrong. When he talks about "we're not looking for how this doesn't work, we're looking for how it works", he's talking about the context of discovery. Falsification comes in in the context of justification. Still messy, but the graphene example is not a counter example to falsifyability.

[coachj.landham]

This conversation is still just so beautiful to me.

[muratartvin]

Falsifiability is not about whether a) you carry out an experiment to disprove a hypothesis or b) progress in science is messy. It is about whether your hypothesis is immune to refutation. Simplified, psychoanalysis: if you get better, it is confirmed; if not, it is because you show resistance, which psychoanalysis predicts, and it again is confirmed. A bit like the arguments of climate change sceptics. The commentators are correct; the professor is not clear about what he wishes to express. This is the first video from Numberphile or Sixty Symbols below par