jump to navigation

My Personal View of Science July 6, 2009

Posted by Ian in Science.

Jonathan on another blog said:

I am also intrigued to find out what you think the limits of science are, and indeed, how you would even define science. If you have some time and inclination, please share. I am genuinely interested in your view on this.

Since this is an issue which I think is very important, I’m happy to oblige.  What follows is my personal view of what science is and what its limits are.

What is science?

There are many different views of just what exactly science is around the world.  Some people hold that science is reductionism, the art of shrinking things to smaller and smaller elements until you get to a complete view of something.  Others hold that science is the process of inducing theory from observation.  Still others hold that science is the systematic evaluation of knowledge. 

It seems to me there are two different realms for science.  The first is a practical, real world description of the act of doing science.  This includes all the elements of science that a cliche’d into the common view such as experimentation, theories, hypotheses, and so forth.  However it also includes all the elements of politics, funding, personal rivalries and so forth are a very real part of science along with the thinking, experimenting and testing.  This form of science is best thought of as an activity, pastime or job.

However there is another place that science fits in as well which is as a principle of discovery.  Despite the illusion of certainty that is created by society, when we are born we know very little about the world around us.  We find ourselves walking around a spectacular universe with a very narrow view of what is happening and with little or no understanding of how anything works. 

Therefore in my opinion science is the act of matching explanations to observations

What is an observation?  Roughly speaking an observation is the claim by an individual that they experienced a certain pattern of behaviour in something.  Now a single observation by a single individual is of limited value but a sequence of similar observations by different individuals becomes useful.  If only one person ever observed something falling and only once then gravity would never have been discovered. 

What is an explanation?  When you get right down to it, explanations are really answers to the question “what patterns are there in observations?”.  They are not answers to “why” questions or even really how questions.  Nowhere is this clearer (to me at least) than in thermodynamics where this mystical quantity of energy is often touted as a reason why things happen.  This is not the case: energy is merely a description of a pattern of observation that has been derived from other patterns of observations.  Energy is what you get when you multiply mass by distance squared and then divide by time squared.  Nothing more and nothing less. 

Based on these definitions it is easy to see that experimentation is the act of testing an explanation to see if it produces the expected observations.

Haven’t you left out the scientific method?

I think the scientific method is actually a wonderful example of science at work.  We have observed a pattern where explanations derived using this particular method seem to be more likely to produce good matches between observation and explanation than explanations derived through other methods.  Over time this method has been refined as patterns of improvement in effectiveness have been observed through following the method.  In other words the method itself is not science but rather an excellent procedure for doing science.  It is not the only one but it works very well.

What are the limits of science?

Since my definition of science is very simple there are only three places that it can have limits. 

Firstly science is limited to what we can observe.  Now it is important to realise this does not automatically limit science to “the 5 senses only” – anything we experience is an observation so if there is a 6th or 7th or 28th way of sensing things then they count too.  However if no-one observes it somehow then science can’t really contribute.  However I do not really see how anything else can do more than science here – the unobserved necessarily remains a mystery. 

Secondly science is limited to where there are patterns.  If there are truly uncaused entirely random events out there then our ability to explain things goes out the window.  However again I do not see how anything else can contribute here either.  If something actually is without any pattern then we can’t do anything about it.

Finally science is limited to our ability to come up with explanations that match the observations.  It is entirely feasible that there are explanations that are simply beyond our ability to comprehend or that we are simply not neurologically wired to figure out.  Yet again if this is the case then I do not see how anything else can help either.

These are the limits to science, but also the limits to cognition in general.  They are, in my opinion, the only limits to science.

What about religion and spiritualism?

A common claim is that there are things out there which science cannot explain but that other modes of discourse can.  I think this is utterly fallacious.

Let us take the rather heavy example of morality where many religious people argue that science cannot explain absolute morals.  If you set aside the fact that I believe there is no such thing, this is wrong for another very important reason:  the very act of recognising that people all seem to follow a universal morality is actually doing science!  You cannot avoid the fact that this is a pattern matched to observations.  It doesn’t matter what the pattern is or what the observation is. 

The same thing applies to ghosts, revelations, and the rest of it.  They are all observations of one form or another.  What are the patterns in them?  How can we link them to other observations?  These are the questions of science.

Final Remarks

Science is not the way it is today via assumptions made at its inception.  Rather it is the way it is because it matches itself to the patterns in what we observe around us.  Materialism and determinism (for example) appear to the casual observer to be inherent assumptions from which science springs but I think this has it backwards.  Materialism and determinism are simply assumptions from which a pattern of more effective explanations has been observed over time and so have been adopted as “good practice”.  If, as the quantum revolution seems to be suggesting, things like determinism are not that solid a ground to work with then we will soon see it dropped as a basis for good practice in favour of a better pattern. 

Finally the remarkable thing about science is that it has created a wonderful positive feedback loop.  By matching explanations to observations we have also managed to find better ways of obtaining observations, which then lets us get better observations, which reveal more accurate patterns, and so forth.  In my opinion it is this feedback loop that explains why our planet took well over 4 billions years to develop life and under 10,000 years to get to where we are today.


1. Wonderist - July 6, 2009

Very good. I would add only one thing; but it is a very important thing. Science is crucially about making predictions. We may match explanations to observations all day long, but if those explanations do not make accurate and reliable predictions, then they are useless as science.

For example, “God did it,” is an explanation that can match any observation. But it cannot make any predictions. That is why it’s not science. F=ma or E=mc^2 are ‘explanations’ that make accurate and reliable predictions. That’s why they are science.

I don’t think that the importance of prediction can be emphasized enough. You imply it, but do not express it as explicitly as I think it should be.

2. Ian - July 6, 2009

Good point Wonderist. I did think about this quite a lot and I think predictions are more part of the scientific method than science itself. Predictions are an excellent way (and perhaps the only way) of testing explanations to see if they match observations but I think the concept of predictions sits in that layer above science called the scientific method.

Of course an entirely valid discussion would be whether there are suitable methods other than the scientific method for determining and testing explanations (I can’t think of any that could come close). If there are not then science and the scientific method could potentially be unified but I think the distinction is still important to convey exactly what science is (to me at least).

3. Ken - July 6, 2009

I like this general description of the scientific method by Neil deGrasse Tyson:”
“Do whatever it takes to not fool yourself when trying to understand the world around you.”

4. Ian - July 6, 2009

deGrasse Tyson is an endless font of great quotes 🙂

5. Wonderist - July 7, 2009

“I think the concept of predictions sits in that layer above science called the scientific method. ”

I disagree strongly. Prediction is the foundation of science, not a layer above it. I base this statement on my interpretation of epistemological pragmatism: Truth is usefulness. And by useful I mean: Able to make accurate and reliable predictions. You see, I am saying that the ‘truth’ of something is literally its ability to help us make predictions. That is the foundation of science, not a result of it.

Here’s a metaphor to help clarify: Truth is like an arrow. When an arrow is ‘true’, we mean that it strikes its targets accurately and reliably. The less ‘true’ an arrow is, the less accurate and reliable it is (it may be warped and fly off in random directions). The more ‘true’ it is, the more accurate and reliable.

And so, when we say a scientific theory is ‘true’, we mean the same thing, metaphorically speaking. It is ‘true’ because it is able to make accurate and reliable predictions. Einstein’s Relativity is ‘more’ true than Newton’s Laws of Motion because it makes better predictions, especially in extreme circumstances.

But, we don’t need science to make predictions. Another ‘useful’ tool to make predictions is the natural, innate ability of our brains to make pretty good guesses. This is intuition. Now, intuition, like any other product of evolution, is flawed and imperfect; but it does work, and works pretty well. For example, when someone is telling you a story, and you realize, without knowing how you know, that the person is lying to you, that is an example of intuition. You could be wrong, of course, but on the whole, humans have the ability to read body language and detect subtle changes in voice that help us to detect when someone’s lying to us.

Clearly, using intuition alone is not as reliable or accurate as using science, but it relies on the same pragmatic epistemology: The ability to make predictions. If our intuitions were not able to make any predictions for us, we would not have become the dominant species on the planet. Building on our basic, innate intuition, we developed the scientific method to weed out the systematic flaws in our natural intuitive abilities (such as all the logical and informal fallacies we continually make as individual humans). But science still relies heavily on human intuition, especially in hypothesis formation. It is just that science is a systematization, aggregation, and correction of human intuition. Thus, I claim that prediction is the foundation of science, not some level above it, or some end-product of it.

In contrast, religion/mysticism takes the flaws in intuition and amplifies them. Religion specifically exploits our innate flaws and weaknesses in intuition. It relies on argument from authority, argument from popularity, etc. as ways to exploit an individuals weaknesses to convince them of *false* beliefs, i.e. beliefs that, when tested do not make good predictions, but serve the religion as control beliefs to influence human behaviour.

So, personally, I think prediction is the most crucial difference between science and pseudo-science/mysticism/religion/etc. It is the one thing that you can rely upon to distinguish truths from falsehoods.

6. Ian - July 7, 2009

Fascinating comment Wonderist. First off I agree entirely that in our current world pseudoscience, religion and the rest of it are not good science and that the lack of predictive ability is one of the main reasons why they are so poor. Secondly I do not think it undermines the power, necessity or importance of prediction to consider it part of the scientific method rather than science itself. In fact I think the power of the scientific method (including prediction) is what makes science so practically useful. However I think science is a slightly more fundamental concept than the scientific method.

To boil it down to semantics: when I look at pseudoscience I think “they are doing science badly” rather than “they aren’t doing science”.

7. Wonderist - July 7, 2009

I think your broad definition of science leaves a gaping hole that creationists, woo woo practitioners, and other charlatans would love to drive their trucks through.

Science today is under threat in the arena of public opinion. Whether it’s intelligent design, homeopathy, global warming denialism, or just postmodern solipsism, there are real people who really do want to undermine science.

I think we have a responsibility to state clearly where the boundaries are between science and non-science. Saying that psychics are practitioners of science, just not ‘good’ science, is like throwing in the towel. We might as well just erase the word science out of the dictionary, since it will no longer have any distinct meaning. The quantum mystics and the Scientologists will rejoice … and will immediately sign up for government funding. After all, it’s all science, right?

If you were in the court room for the Kitzmiller v. Dover case, who’s side would you be on? Would you be saying, “Well, ID isn’t really good science, but it *is* science, and so it’s appropriate to teach it in a science classroom,” or would you be saying, “No, ID is not science.”

I hope you see that I’m not making a big deal out of this just to be contrarian. This issue is very important in a real-world sense. I don’t think it would be helpful to muddy the waters by using such a broad definition of science such that children’s make-believe technically counts as ‘science’, because it is matching observations with explanations, regardless of how useful those explanations are.

8. Ian - July 7, 2009

I agree entirely that we have a fight on our hands to maintain science and that there are dishonest people out there trying to undermine science. I also understand the importance of getting this right and the need for clarity. However I see the solution differently from you I think.

If I say pseudoscience is not science then I immediately open up the opportunity for them to say that since it isn’t science then the scientific method doesn’t apply and they can make whatever claims they like and I can’t touch them. If I say they are doing bad science then I can demonstrate their failings and effectively make them do it properly – there is no escape. If they are included then they have to play by the rules. If they are excluded they can make up their own rules.

The same applies to creationism and ID – I frequently hear people saying that science doesn’t deal with god or morals or whatever – I say science does apply, it must apply. I want to be able to rigorously demonstrate why their explanations are poor matches to the observations rather than letting them falsely exclude science from the discourse.

Of course much pseudoscience is not science regardless of which definition we take. Coming up with a random idea and then touting that as truth regardless of observation is not science whichever way you define it and can easily be shown to be a poor way of doing things. Similarly distorting scientific findings to push forward an idea is not science either.

I want to encourage people to do better science by showing them where their attempt at science fails. I would much rather recognise and improve someone’s scientific skills than simply dismiss their ideas as not science and ignore them. The end result for the idea is the same but the former may well produce a new scientist, while the latter may well produce an enemy of science.

I also worry about dogmatism in science. There is a danger that the scientific method can become a dogmatic way of life rather than simply a well established method with the possibility of improvement (however small). From a tactical point of view there is also the danger that dismissing so-called non-science out of hand can lead to false claims of scientific dogmatism (and it often does).

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: