Science is NOT about explaining

[Mr Skeptic]

I've had it up to here with people saying that science is about explaining things. It isn't. Science is about predicting things, and any explanation is a bonus. Anyone who's studied the scientific method will realize this is true. If science were about explanations, there would not be any need for experiments to verify the predictions.

 

I realize that when scientist say a theory "explains" something, they are implicitly saying it makes predictions. However, I'd like to ask them to avoid that since it may confuse laymen. For example, "God did it" will "explain" an object falling, but not in the same scientific sense as Newton's theory of gravity explains an object falling.

 

Science is about predicting, and any explanation is a bonus.

[Lan(r)12]

If science were about explanations, there would not be any need for experiments to verify the predictions.

 

Why do you think the earliest scientists even bothered to use *science* in the first place?

To EXPLAIN something previously unexplainable.

What you are arguing is semantics at best.

I certainly see your point as it pertains to modern science, but I do not think Galileo or Copernicus did all the stuff they did to make predictions. I think that their hope was to EXPLAIN something so that they would BE ABLE to make predictions on other matters as they arose.

[Bignose]

Lan, this is more about the people who like to tell a good story and think that that story is all that is needed to be "scientific". In the P&S section, there are two very good examples -- "atoms are cubes" and "EM thread theory". Both the OPs in those threads have claimed that their ideas a completely "logical" so that means that they are right.

 

Both fail to realize that there are many other "logical" explanations out there. "God did it" is the 800 pounds gorilla of them all, bit history is also replete with examples. It was once "logical" to know that the moon is made of cheese. It was once "logical" to know that the Earth was flat. It was once "logical" to know that the earth rode around on the back on an elephant standing on a tortoise. More recently, it was "logical" for some people to believe in N-rays, and it was "logical" to believe heat was an invisible fluid called phlogiston.

 

The point is that logic alone is completely insufficient. Science demands evidence to back up the logic, through experimentation. And in the vast majority of experiments are going to result in precise measurements and numbers, which is why we always ask for the math when someone has a new idea. The math based on the "logic" of the idea should make quantitative predictions which can then be tested via experimentation. If the experiment and prediction agree, then you have something.

 

If you just go with "logic" -- then all you have is a story -- just like the stories of the Earth riding on the back of an elephant and a cheese moon. You need to prove using objective, clear, unbiased evidence every single assumption before you start drawing conclusions from them. Like I wrote in the other thread, an idea based on unproven ideas is like a house built on sand -- one strong wind and it is blown away. An idea based on proven ideas is like a house build on rock or a foundation -- able to stand up to much harsher conditions.

 

Logic alone is just sand. Logic backed up by evidence is rock.

[swansont]

Science does explain things, but if all you have is an explanation, you don't have science. Science is not ad-hoc.

[Mokele]

Science is about explanation, but prediction is what makes us take an explanation seriously, and what makes it useful.

[Bignose]

Science is about explanation, but prediction is what makes us take an explanation seriously, and what makes it useful.

 

I really like this -- one sentence, straight to the point, and is exactly what I agree with.

[Mokele]

Thanks!

[Glider]

Science is about explanation, but prediction is what makes us take an explanation seriously, and what makes it useful.

Exactly. Predictions allow the testing of theory and theory is the strongest construct in science. Theory is explanation.

[D H]

Exactly. Predictions allow the testing of theory and theory is the strongest construct in science. Theory is explanation.

Now we're getting there. Neither prediction nor explanation suffice. A conjecture might be strong in terms of explanatory capability, but if it those explanations don't conform with reality the conjecture is worthless. An ad hoc scientific law might be strong in terms of predictive capability, but the ad hoc nature of the law leaves something wanting: Specifically, why?

 

A scientific theory explains why ad hoc scientific laws arise and when they are valid. Theory is the glue that binds prediction and explanation.

 

I'll use blackbody radiation as an example. Wein's law (1896) does a good job of describing the high frequency spectrum of a blackbody but fails to describe the spectrum at low frequencies. Max Planck, in what was essentially an exercise in curve fitting, found an equation involving two rather arbitrary constants that yielded a much better fit. He presented this in October 1900, but quickly realized the solution was not unique. He rescued his law by inventing what was to him a purely mathematical construct, energy quantization. He presented the modern version of Planck's law in December 1900 (published 1901.)

 

Planck himself saw this quantization as a mathematical fiction whose sole virtue was that it let him rescue his law and tie it to thermodynamics. Planck's law by itself is just that -- an ad hoc scientific law. It remained such for two decades. It took a decade for physicists to realize the true significance of Planck's quantization. It took another decade for physicists to come up with an explanation of why Planck's law is indeed valid. That explanation arises from statistical mechanics using the Bose-Einstein statistics for massless particles.

 

 

Science is all about explaining. Those explanations must of course match reality, and aren't worth much if they can't predict outcomes. Scientific theory marries explanation, post-diction (matching reality), and prediction.

[Mr Skeptic]

Even so, explanations always take a far second place to predictions. A theory that makes good predictions but offers no explanation is still superior to a theory that has a nice explanation but has poor predictive capability. An "explanation" that makes no predictions is no explanation at all, just the illusion of an explanation.

 

For example, Aristotle's theory of gravity explained why things fall (finding their proper place) and made qualitative predictions. Newton found a quantitative relation between mass and gravitational force, but did not offer an explanation as to why it would be so.

 

Explanation is part of the aesthetics of a theory. Given equal predictions, people are going to choose the theory with the nicer explanation. Also, much of the time finding an explanation for the predictions will lead to more predictions.

[darshanchande]

The basic purpose behind science is "to predict" things. But the process of prediction is "explained" in form of a theory, which would be called a scientific theory. I believe the explanation is also as much important as prediction is, because without explanation and experiments there would be no credibility to the prediction and so it would be useless.

 

Science is both, predicting as well as explaining, then only it is complete.

[scrappy]

The basic purpose behind science is "to predict" things. But the process of prediction is "explained" in form of a theory, which would be called a scientific theory. I believe the explanation is also as much important as prediction is, because without explanation and experiments there would be no credibility to the prediction and so it would be useless.

 

Science is both, predicting as well as explaining, then only it is complete.

I must agree with you. But I have to ask: Couldn't the same thing be said about religion?

 

Maybe we should add a third leg to the scientific stool: correcting. The greatest tool science has at its disposal is the peer review. Rigorous skepticism embodies the Darwinian advantage of science. As such, it's the "natural selection" of the scientific process.

[UC]

Science, unlike religion, is a logical system. A system has a framework of basic information that is assumed to be correct (in this case, all sorts of equations that can't be derived from other equations). From these basics, it derives all other content (many equations in physics can be derived if you have enough time on your hands). A system has the power to accurately predict and explain based on it's contents. It is far from a perfect system, which would be complete, have no internal contradictions, and establish the validity of all assumptions (including the validity of deductive reasoning, but that's largely *the* big project for logicians).

 

Religion on the other hand is more like a list of "facts." If I have a question, I consult the large holy book of your choice and look up the answer. If the answer isn't there, I'm screwed.

[proton]

I've had it up to here with people saying that science is about explaining things. It isn't. Science is about predicting things, and any explanation is a bonus. Anyone who's studied the scientific method will realize this is true. If science were about explanations, there would not be any need for experiments to verify the predictions.

 

I realize that when scientist say a theory "explains" something, they are implicitly saying it makes predictions. However, I'd like to ask them to avoid that since it may confuse laymen. For example, "God did it" will "explain" an object falling, but not in the same scientific sense as Newton's theory of gravity explains an object falling.

 

Science is about predicting, and any explanation is a bonus.

Science is about describing nature. It’s circular to say that science “explains” anything (although we often word it as if it does). The typical application of the scientific method starts out with observations, lots of them. From those observations scientists attempt to formulate an hypothesis which describes what was observed. The formulation comes in the form of a "law" (postulate, axiom etc) which attempts to describe the observations. The laws almost always come with the power to make predictions. To be precise one should say Science does not explain; it only describes. Fritz Rohrlich covers this topic his new text Classical Charged Particles in the first chapter which is entitled Philosophy and Logic of Physical Theory. For example, on page 2 the author writes

Having gone through all the trouble that is necessary to find the "right" laws and the "right" theory, no one should be surprised that the predictions of the theory indeed agree with experiment or, for that matter, that a given phenomena can be fully accounted for by the theory. Nevertheless, we say that the theory explains the phenomena. In this sense scientific explanation is circular. The emphasis should really be on the existence and correctness of the deductive system, not on the explanation, for the theory is not derived or derivable from experiments. It is a mental step into the abstract and general, postulating validity in the future and for all experiments including those that have never been carried out before. The existence of a valid theory is therefore the nontrivial and indeed the very remarkable facet of scientific explanation.

We now define the aim of theoretical physics not as an attempt at explaining the phenomena, but as striving for the construction of more and more inclusive physical theories and the exploration of their ramifications.

[mooeypoo]

I must agree with you. But I have to ask: Couldn't the same thing be said about religion?

 

Maybe we should add a third leg to the scientific stool: correcting. The greatest tool science has at its disposal is the peer review. Rigorous skepticism embodies the Darwinian advantage of science. As such, it's the "natural selection" of the scientific process.

I see your point, and I agree.

 

My only contention is that I, personally, for the sake of avoiding a potential endless-fall into religious discussion, would use the comparison to pseudoscience instead of religion.

 

As in, "Couldn't the same thing be said about pseudoscience". Which it could, and it's a good point, only without touching the 'sensitive' religion button.

 

Just saying.

 

I agree with the point, though. The self-correcting method of science is one of the most important aspects of it. That's how we evolve new understanding of our environment, and how theories are replaced with better theories. It's also the main difference between it and everything else.

 

You won't see astrology having a self-correcting methodology to it, even after pluto was redefined as a "plutoid" instead of a planet (and about 100 other bodies of the approximate same size found next to it). You would, however, find that happening in actual scientific subjects.

 

Good point.

 

~moo

 

p.s: If it wasn't clear, I am asking the rest of the readers of this thread to avoid getting into a religious discussion about religion. This thread is about science, and the point made was about science, and used religion as an example, not as a principle. Please keep it that way.

[Royce]

Science to me is about having fun and to make a safer more productive world.

 

We are all we got, and I don't think any of us would be interested in doing scientific research if it wasn't in the best interest of us or others.

 

I will take the Miss America crown now