Five Popular Posts Of The Month

Tuesday, October 8, 2019

On A Definition Of Science

On A Definition Of Science

Part I: a general discussion
This is a common description of science:

In simple words, science is the practice of thinking about things and producing new systematized knowledge.
This understanding of science is thousands of years old.
It is long overdue to rethink what science is, what science is for, and how science is done. Nowadays, too many people can claim that they are "thinking about things and producing new systematized knowledge" - for example, any bloger, or even an active Facebook user.

Science, any science, every science proves one point, a point that is common to every science and accepted by every scientist. This point is very clear and simple –
things change.
Our knowledge about the universe changes – it grows.
Today we know much more about the universe outside and inside of us – humans, than thousands years ago, or even a hundred, or even a dozen.
But the fact of the matter is that the knowledge does not just accumulates, grows.

It also evolves.

What was common knowledge some time ago may be replaced with completely new.

Of course, the majority of our new knowledge just covers the areas where in the past knowledge simply did not exist – in those areas the know knowledge replaces the absence of past knowledge. However, some new knowledge may clarify, rectify, or negate and replace some of the past knowledge.

Some elements of science that used be considered “knowledge” have been removed from science. Some of the examples are a philosopher’s stone, perpetuum mobile, the Ptolemaic model, eugenics, and many many more.

This evolution of science should also involve our understanding of science - what science is, how does this practice happen, what does this practice include.

In particular, there are areas that in a past were considered as science, but nowadays is not really treated as such anymore. For example, botany and zoology both eventually became a part of biology. Botany is not really a science anymore, but a field of practice that has some elements of science, starting from classification.

Why is the change?

At the birth of science, the mission of science was developing new knowledge – any kind of knowledge. The main method for that was using language for describing observed events/phenomena, including objects, their properties, processes happening to those objects, and properties of those processes. The result and the product of science was – a description (in form of texts, basically, if you write something off your mind about things you see - your are a scientist, or a poet).

With the development of mathematics, that description could employ some abstract elements, but for a very long time mathematics was evolving on its own and was not considered as a tool to study the nature (despite the fact that the origins of mathematics were deeply rooted in naturalistic observations).

The idea of “predictability” was not really a part of a scientific development until about 18th century (however, it was assumed that a good shaman, oracle, wiser, astrologists could predict important events).

With the advances in physics and chemistry, scientific community slowly adapted a new mission of science – making reliable predictions based on developing new knowledge. However, even now a scientific community does not have a unified view on the mission of science. A simple search demonstrates that; try “mission of science” and you will see missions of science departments, scientific magazines, institutions, but not the mission of science as a human practice.

This fact may have several interpretations, including that scientists do not really know what they do (“What do you do? Science. What is it? I don’t know.”).

The central feature of the human practice called "science" is predictability - that is what separates science from anything/everything else.

From this point forward, I define science as a human practice with the mission of making reliable and testable predictions based on previously collected and/or developed knowledge.

Hence, developing and collecting knowledge, clarifying and rectifying it – is a part of scientific discourse, but not sufficient yet on its own to make that practice to be science.

If not science – then what?


A scientific practice.

By employing this definition of science, we utilize one deeply scientific action – classification, part of which involves different names: science v. scientific field – to giving different objects (physical or abstract).

However, all other scientists, including the NSF still use the old definition of science –as a practice of development new knowledge, omitting the necessity of making reliable predictions. And for them a scientific research is an equivalent of making any type of a description – nowadays, that description has to include some data, but no one checks if that data leads to any testable predictions. The quality (or as says the NSF the merit) of research is assessed via peer review. A peer review process helps to eliminate low quality papers. But is also eliminates papers with un-ordinary ideas because they also do not belong the common views of reviewers. The #1 quality of a good paper is the list of references – if one does not have in the list “important names” why bother reading the paper nothing good can come out from an author who does not know “the establishement” (Einstein would not be published theses days).

Part II: specific scientific fields

Now we can make a statement about a specific scientific area of practice that represents a scientific field but not a science.


The state of a scientific practice in education is similar to botany, or alchemy.

Of course, some predictions can be made, but they are usually trivial, like “practice makes perfect”. We can call that set of rules – heuristics of education (this link list some of those rules: Fundamental Laws of TeachOlogy: a Handbook For a Science Teacher.).

The vast majority of publications in education are not much different from the letters of an explorer sent back to the academy from an unknown frontier – a simple description of objects and events encountered during practice, “spiced” by some speculations on why would those events occurred in the way they did.

The use of some statistical method doesn't make it more scientific. On the contrary, it covers up the fact that - what is done is not science (i.e.

A mere fact of using math does not make practice scientific (e.g. astrology, numerology). If a mathematical analysis of statistical correlations demonstrates a strong correlation between two parameters, that may have some significance - if the number of important parameters influencing a system is small. However, when the number of parameters governing the evolution of a system is large, a strong correlation between two parameters have no significant meaning. Yes, it exists – and that’s that. There are, or at least may be, many other strong correlations that the analysis does not show. Hence, the model cannot be used to make any reliable predictions. Hence, it is not scientific.

This is a case for any social system including educational systems.

Scientific study/research in education may eventually lead to a transition of this field into science; in that sense, this field now is in a pre-science state; it is a pre-science.  To make a reliable prediction, scientific practice needs to collect vast amount of data – i.e. knowledge presented in a numerical form, and then to analyze the data to establish robust correlations. The difficulty is that social systems are much more complicated and diverse than physical (even quantum or astrophysical).

Take, for example, a simple unit of education – a class. There are almost infinitely many combinations of students with different backgrounds, cultural histories, economic circles, etc. But in theory, all possible states of this system – a class – could be described in terms of the values of specific parameters (age, gender, race, and more). And the following observations could let the development of models robust enough for making reliable predictions about the evolution of this system (and its elements - students). Clearly, this type of research would require vast funds and completely new approach/strategy to educational study, and currently is not even being considered by any governmental of private entity. Some publications on the matter are:

More on this page: Strategies For Teaching Science.

Now, let’s discuss a more specific matter – a research, because this is what people do in science.

“I’m a scientist, I research/study this”.

There are two types of a research - a scientific research, and a generic research.

A genetic research happens when one just describes what one encounters; it is a search combined with some verbal description of events. This is what people usually call a “study”. “I study bacteria (or stars)” used to mean (and often still means) “I am looking in a microscope (or a telescope) at those tiny (huge) objects and describe what is happening to them”.

A scientific research includes a generic one, but also involves a search for patterns and strong correlations. Scientific research cannot be done without collecting data. To test if a research is scientific enough, one checks how many predictions can be done based on its results, and how reliable those predictions are. Those tests are often being called – experiments.

There are three types of a scientific research; they are based on originality and technical difficulty of a research.

1.  Standard research - anyone (in the field) can come up with its idea, and then anyone (with resources) can do it. It is just a matter of who comes up with this idea first. This is the type of a research that one finds in 99.99 % of all science publications. Currently, in physics a popular research is to check if quantum mechanics still works for large objects or at high temperatures. Ideologically that does not represent anything new. For example, here authors write, quote: “Our results show excellent agreement with quantum theory”. But the technologies that allow such experiments have become available only fairly recently; such experiments demonstrate more of engineering power than science. BTW: the popular treatment of quantum mechanics in this publication is a good and common example of how a dilettante who does not understand basics makes simply wrong statements (like “interference means an object exists at two places at the same time” – no, it doesn’t). More on quantum entanglement in "Can An electron Travel through Two Slits At The Same Time?" or "On Entanglement Between SuperFluidity, SuperConductivity and Entanglement".

2. Breakthrough research - everyone has an idea of this research, but no one can do it – until finally someone does. An example is the Fermat's Last Theorem (until recently).

3. The original research - not anyone can come up with that idea, but when the idea is out there, anyone could do it as well. An example of this research is Einstein’s explanation of the photoelectric effect. The original research is the one that may lead to a change in a scientific paradigm.

Now a note on the importance of language in science.

Scientists do not talk to each other like: “Hey, Jim, look at that thing with a thing doing this thing”. They develop a specific language to communicate. Some of the words in that language may sound/look like regular words from an every-day vocabulary, but in fact they usually have a very narrow, specific meaning; and there are also words (and even symbols) invented specifically for scientific communication. Everyone who study science must learn that language.

Nowadays there are many pseudo-scientific writers who write about scientific discoveries but are too lazy to learn scientific language. Many do not understand the difference between an actual object and an abstract description of its properties. This happens a lot when people write about physics, especially about quantum mechanics. Without getting into more details (follow to this page for the details: Fundamentals of Quantum Physics), I just want to note that many of those writers do not know the meaning of even such fundamental terms like “an object”, “a field”, “a wave”, “a particle”.

An object is something that represents the focus of our attention. This is what we are talking about. An object can be physical (e.g. we can touch it; often we also call it a system) or abstract (e.g. a symbol). A small physical object localized at a specific place in space is called a particle (usually, we say a system when we focus on many particles or on a large object). Some people believe that this definition should be broaden to: “a particle is a small physical object that can be localized at several specific places in space at the same time”, but so far, this is a matter of a debate. Using a traditional definition of a particle, we can say that it may occupy some location; locations may change with time; and then we can start developing means for describing that change, and then describing possible interaction with other particles. A particle may exhibit a deterministic behavior, or a probabilistic behavior. For each particle to describe what is happening to it, we can assign a set of parameters, and a specific set of values of those parameters we call “a state of a particle”. That state may evolve, i.e. change in time; when that happens, we call it a process, or a behavior.

A field is a mathematical (i.e. abstract) description used to represent properties of matter in large regions of space. A field is used to assign a specific state to many different locations in space. But at each location in space there is (almost) always an actual physical object – usually described as a particle (an atom, a molecule, an elementary cell) – in a specific physical state. This is what we call a physical substance. In reality, a physical substance always has a structure. A physical substance is composed of many interacting and localized objects – a smallest portion of a substance that repeats itself in space; and all of them can have different states, hence, evolve.

Again, a field is an abstract object used to describe properties of many particles existing simultaneously within a large region of space. In other words, a field is an abstract construct used to describe a distribution of possible states in space and time. Every existing field (with one exception, so far) – even quantum ones – can have such an interpretation. One exception is – a gravitational field. So far, we do not know if gravitons (quantum particles associated with a gravitational field) exist. Most probably they do, and in that case even a gravitational field becomes just an abstract construct. But until we know for sure, we can treat a gravitational field as a mathematical description of the actual properties of space and time at different locations and instants within large regions of space and time. We can think of space-time as a substance. At a classical level, an electromagnetic field also looks like a field with no substance that describes a state of each point in space at a given time, but at the quantum level it becomes a description of light-particles, i.e. photons.

When a state of a substance changes, it is reflected in the changes in the field associated with that substance. That change may affect different parts of a substance in a specific – consecutive-like – manner. This process is called a “transfer” or a “propagation” and can be described in terms of regular (in some way) physical and abstract objects called “waves”. A wave is just a specific shape of a substance, and a specific way for a substance to change its state. A mathematical description of a physical wave is an abstract construct that represents a specific state of a field.

When people who write about physics do not know its language, what they write often does not make any sense. It may sound “scientific”, though, for a person who did not have solid science classes.There is a difference between writing about science and being a scientist (as well as there is some difference between doing science and being a scientists).

The best science to teach science - its way of thinking about nature, its language and principles - is physics.

Because physics is the simplest of all natural sciences, and hence, the most developed and most understood one.

A Full Physics Course


  1. The font is hard to read. Please do not use italics.

  2. Great insights. I look forward to reading what you're planning on next, because your post is a nice read.

  3. 833F7