Hacker News new | past | comments | ask | show | jobs | submit login

> Science is NOT about consensus.

The entire scientific method is built around consensus. As science is a process subject to continual refinement. Almost all propositions put forth by science are ultimately incompletely, inaccurate in some cases, or limited to specific situations we can test.

So, the scientific process (and the method itself) rigorously challenge all propositions. Some simply fall apart under the weight of their own inconsistency, which is what happens when results can not be replicated. But those which can be replicated achieve consensus until a more accurate proposition can be made.

Thusly, Newtonian physics was supplemented and updated by Relativity. Wave-particle duality gradually replaced competing theories. Et cetera. These are consensus opinions. They are not facts. They're the closest models we can get based off our observations. The core of the scientific method is that things are given a continuous opportunity to be falsified.




So science is like some big dinner party where really smart people decide what best model substitutes for reality?

I always thought it was more like a dinner party where really smart people were usually wrong and engaged in petty groupthink, and the guy who was able to show they were wrong (by reproducible experiment) eventually changed their minds (after quite a bit of trouble, and sometimes by having to wait until they retired or died)

Have you heard of the book "The Structure of Scientific Revolution", probably one of the top ten books on science in the last hundred years? Wasn't the entire point of that book that the way science is sold to kids, ie, a linear process where one good idea comes out and naturally replaces another, was a complete fable? In reality science gets "stuck" in various paradigms and it takes quite a bit of pushing to get them to change.

I've been observing your comments, and I wonder how you make these two things fit together.


> I always thought it was more like a dinner party where really smart people were usually wrong and engaged in petty groupthink, and the guy who was able to show they were wrong (by reproducible experiment) eventually changed their minds (after quite a bit of trouble, and sometimes by having to wait until they retired or died)

I think this is the thing I said. Ideally the change is purely a matter of data being presented and reproduced, but human politics inevitably creep in. But either way, saying "Science is not about consensus!" is at best a misunderstanding and in some cases it's actually a tactic for climate change denialists.

But I'm not sure how my comments require me to reconcile the information you mentioned from "The Structure of Scientific Revolution" with my viewpoint? Could you explain, please?


Perhaps we are violently agreeing? ;)

Here's the thing. There are two concepts here that people mix up quite a bit: the scientific method and the politics of science.

The scientific method is about 1) Abduction. Collecting data and finding patterns. 2) Deduction. Forming the patterns into possible rules, and 3) Induction. Showing through reproducible experimentation that the rules work (or not) and then extrapolating that to the universe at large.

There's not much argument on the scientific method. A lot of philosophers point out it's many problems (induction, for one, is a thorny one. And there's the problem of instrumentation) but in general the scientific method is the light that lets our species see in the darkness. The reason you get into an airplane and trust it is because these three processes have been followed. The reason medicine is fundamentally different than, say, physics, is that in some cases strong correlation between data and induction is all you have -- there is no hypothesis holding it all together (or a very weak one). Different sciences and different subjects have various levels of maturity in all three of these areas. It's important to understand that when talking what the status of those sciences are.

The politics of science is all about consensus, funding, peer reviews, press coverage, political causes, etc. The actual practice of science, because it is full of people and not demi-gods or robots, has a lot of politics built into it.

The interesting questions for any discussion of science are 1) what is the maturity of the science in all 3 of these areas, and 2) are we talking about the scientific method? Or the politics of science? (either one may be important, but you have to know which you're discussing)

Over the years schoolkids are taught some sort of propaganda that mixes all of this into one big pot and stirs in a little hero worship. (I think the hero worship is well-placed. Scientists are some of my greatest heroes). Scientists are these really smart guys who move from one great idea to the next as new information comes out, and science is the process of being the most "enlightened" by being up-to-date on whatever the current consensus is.

But a funny thing happened on the way to nirvana -- Thomas Kuhn started looking at how the work of science gets done. And he found this huge gap between the legend of how science gets done and how it actually gets done. There's really too much there for me to do justice in this format, but as an exaggeration suffice it to say that scientists have turned out to be as human as the rest of us, and consensus is probably very much a lagging indicator of where the actual science is leading. Lagging by perhaps as much as decades.

So when you say "consensus is what separates science from philosophy" or that "science is all about consensus" I find I must interpret that as "the politics of science" for it to make sense to me. But then when you start using that consensus in some sort of functional context just like the real work of science, it doesn't fit any more.

It's probably me. I'm just confused.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: