Thursday, August 23, 2007

What is Scientific?

In the modern world, as soon as you propose a new idea, the first argument that you will be faced with is that the idea is not scientific. Although, just being scientific does not really make the idea very useful or legitimate it does help a lot to gain respectability if you are able to prove that it really is scientific. So what is scientific? What does it mean to say that something is scientific?

Invariability in Space

Scientific ‘laws’ or equations or statements should be invariable in space. For example, when Newton says that action and reaction are equal and opposite, what he really means is that they are equal and opposite all over the universe – in Europe, in Asia, in America, on the Moon and even on Andromeda. And there is a slight extension to it. In case your statement actually has to do with measuring physical quantity then the scientific idea should be invariable with rotation too. For example, if a proton decays in t seconds when your apparatus points north, it should also decay in t seconds when it points west.

Invariability in Time

Invariability in time is perhaps even more important than in variability in space. It is this that makes scientific statements into ‘laws’. A law is something that always happens. It happens today, it happened yesterday and we have reason to believe that is also happened billions of years ago right back to the Big Bang. Scientific laws should be repeatable.

There is something more to be said about repeatability. Things are repeatable only when the ‘experiment’ is set up in the same way. One cannot expect gravity to act the same way on moon as it does on earth. The masses are different. One has to make sure that all the factors that significantly affect the phenomenon are similar. If you want the same amount of gravity on moon as on earth, you must put in the same amount of mass into the moon.


This condition is perhaps the trickiest of all. And the most difficult to understand, perhaps. But also the most crucial. Often you will be faced with unscientific claims that are invariable in space and time. But they also have to be deniable. Unless they are deniable they are not, by definition, scientific claims.

To state simple, deniability means that it should be possible to set up an experiment that can possibly refute the scientific claim. For example, taking Newton’s example, it should be possible to set up and experiment that can possibly prove that action and reaction are not always equal. One such simple ‘experiment’ would be putting your pen on the tabletop. If action-reaction were not equal-opposite, the pen would fall through the table. But is does not happen.

The idea here is that it should be possible to set up such experiment and not that the experiment would necessarily refute the above claim. In fact, if the claim indeed is a law then it cannot the experiment cannot refute it. However, such an experiment should be possible to set up.

Another thing to note is that merely having such an experiment does not make the claim scientific. You have to ensure invariability in space-time too.

Using these tools let us examine a spurious ‘scientific’ claim. A pundit claims that if you pray to Vishnu with full faith he will cure your cancer. And indeed, he cites many cases (possible concocted by himself) when this has happened. But he says, faith is important, if there is not faith it will not happen. This ‘scientific-ness’ of this claim is easily put to test. Is it possible to set up an experiment where someone prays with full faith and still does not escape from cancer? As soon as you say that, the priest is in trouble because then he will have to define exactly what faith is and what it is that you have to do during the prayer. I’m sure he will not do that.

Because if he does that, then you can immediately perform the experiment to see whether his claim holds or not. As it will turn out, I’m sure, the claim will not be scientific. It is valid as a claim of religion and faith but not of science.


  1. Why dont you state the idea here that bought on this scientific validity per se? :)

  2. I don't understand what you mean. Do you wish me to trace the development of this particular definition of science?

  3. It is not possible to apply a law to every phenomenon of the universe.
    for example, one cannot apply the second law of thermodynamics to understand the high temp. of corona.

  4. Very true, Adit. You did not even have to go to the murky waters of thermodynamics and solar corona's to give an example. The down-to-earth laws of Newton themselves are not applicable to the world of quantum mechanics. However, it is not this kind of universal applicability that I'm talking about. I am talking about universal applicability one all the conditions of the experiment are well defined.

    A lot of things remain tacit while stating laws. When you state Newton's second law what you really mean is something like - when two macroscopic bodies (e.g. bodies having masses few gram to few thousand kg) interact, then Newton's second law holds to the accuracy of one part in a few billion. Once the law is stated in this form it becomes universally applicable. Macroscopic bodies obey Newton's law to this accuracy all over the universe, or so we believe.

    A lot of new discoveries in science are made by uncovering and contemplating what is remaining tacit in our statements/experiments. Once the limitations of size in Newton's world are identified, Quantum Mechanics can be discovered. Once the limitations on applicability of the Second Law of Thermodynamics are identified, a new theory for solar coronas can be discovered.

  5. the term 'scientific' seems quite relative actually...
    anything consistent wid the presently known and (unanimously accepted as)scientific laws and theories is generally termed scientific while others arent....
    whereas you never know when even the most 'unscientific' things may turn out to be the most 'scientific' ones.....

  6. I don't think any or you is getting the point. The idea is not to classify what particular theories or laws are science but to identify what 'method' is science. I'm talking here about the philosophy of science and the scientific method rather than the actual theories and practices. See wikipedia for more details.

    Especially the section of falsifiability or deniability as I call it.

  7. From where I see it, the post is about how would you classify anything under scientific or non scientific. It (read 'the post') just *looks* *scientific*, it really is philosophy ;)

    AND what I was trying to say in my earlier comment was wrt:

    "In the modern world, as soon as you propose a new idea, the first argument that you will be faced with is that the idea is not scientific".

    I was asking did you have *a new idea* in your mind thinking upon whose feasability bought on all this discussion of it being scientific, feasible, accepted, however.

  8. That is right, Nainy. As I say in my previous comment, the post is about the scientific 'method'. And no, I didn't have any particular new idea in mind. So don't worry, I'm not about to get a Nobel anytime soon. :D

  9. The so-called "peer Review journals" have multiple issues. Publication is expensive, either for the author or the reader. Many articles are refused just because of some formal detail that has nothing to do with the factual content, or just because the papers can only publish so many articles per month. And their rules against so-called "redundant publication" mean that they categorically refuse to publish theories just because they happen to have been first published in something that is not officially considered "peer review". Refusing to acknowledge a theory just because of historical contingencies of whereabouts of publication is of course no better than refusing to acknowledge a theory just because it conflicts with some religious scripture. Those who defend that narrow system tend to justify it by claiming something on the lines of "or else any quack could publish anything and call it scientific", but that is just a false dilemma fallacy that ignores the more efficient ways to review articles that the Internet allows. It would be perfectly possible to create an Internet database with reviews that refute bullshit, and then the critical reviewers can simply post links to the relevant articles wherever the nonsense articles pop up. Cheap, efficient and no false dilemmas!