The Philosophy of Science
Here I would like to mention some things about science in general. Any scientist should reflect now and then over what science, research and technology is all about. The philosophy of science is also a topic I enjoy thinking about in my spare time. I think many "common" people barely understand what scientists do and by that I mean not even in terms of methodology and working principles. (If you want to know more about my research topics in particular, there is a specific page for this in the menu.)
All the things I write here are personal reflections and it may well be that other scientists (and philosophers) do not agree with me. The theory of science is a subject of philosophy and contains somewhat more subjective discussions. I am far from the best person to write about it as I have only taken a few basic courses, but perhaps my texts here can act as a source of inspiration to make more people interested in the topic. I do think it is very important for everyone to reflect upon what science really is.
The texts below remain under construction and will grow with time.
Science is a method
First of all, science is not a belief system that contains certain "truths" abot the world which one is supposed to accept. Science is a method that attempts to evaluate in the best way possible what is true and not. This page is about trying to describe how that method works, i.e. how scientists work (or at least how they should work). The biggest difference between science and belief systems like religions is that they generally simply offer "truths", not methods on how to reach the truth. There are clearly quite many things within science that are regarded as true, but this is simply because there is such an overwhelming amount of previous work supporting that they are true. The results have been obtained by the scientific method of working.
Although I do support science outreach activities I am sometimes disappointed by how science is introduced to the people during activities like science festivals or at science centers. There is nothing wrong about showing off explosions, colorful bubbling liquids or wierd animals to gain public interest. In some cases people might even learn some scientific fact about a phenomenon. But in the end science is about something much bigger and deeper! It is a strive for rational unbiased thinking and experimental design to evaluate reality by the best possible methodology. As imperfect humans we may or may not fulfil the goal of working according to the scientific method, but that does not change that we can all agree to strive for it and agree on common ideals. In my opinoin it would be even more valuable if one could also teach exactly this, the scientific method, to the public.
We cannot believe what we want to believe
Science is about investigating the universe in which we exist without any prejudgement of what to expect. As a scientist one cannot claim that something is true just because one wants it to be true. (It may happen that some researchers do, but if so it is simply an example of bad science.) Scientists develop models for how the universe works. We try to find rules for how nature behaves, whether we like those rules or not. This is perhaps the most basic part of scientific methodology.
Regardless of what view on science one holds on to, it is generally safe to conclude that science is about trying to match theories with empirical observations or controlled experiments. In other words, scientists gather evidence in order to support (or falsify) theories. Evidence can be indirect, i.e. a certain conclusion can be realistic following logical reasoning and already available information. Evidence can also be direct (or "conclusive"), i.e. we can have data that shows directly how one effect or attribute gives rise to something else. For instance, what evidence is there that the Sun will rise tomorrow? We can prove this theory indirectly: Considering that the Earth rotates in relation to the position of the Sun, it should appear the next day. This is essentially an argument of geometry and is based on the assumtion that the Earth does rotate. However, we can also prove the theory by crude empirical observation: We have a lot of gathered information on how the Sun has indeed risen and settled for as long as we have existed. Thus, the hypotheis that "the Sun rises once every 24 h" has very very much empirical support (as long as one is not too far from the equator) and it is therefore rational to regard it as true. This is a type of simple direct (some would call this "inductive") evidence that governs most people in their actions, usually without anyone thinking about it. We can also look at the example of falling objects. Everybody has seen that "stuff falls", which is direct evidence. The theory of gravity is a more general (and perhaps elegant) theory, but it is still indirect in principle as we use it to predict (or perhaps explain) that things fall because masses attract each other.
Scientists produce data in their experiments. A common criticism against the scientific method is that we cannot act as neutral observers of reality. In other words we cannot know that what we observe is the "true" world because our senses are limited. It is correct that the limited and inconsistent perception of humans can lead to problems. This kind of relates to complete relativism, that we can never know anything for sure, all theories are equally worthy because we cannot judge etc. Such extreme viewpoints only leads to madness and nobody lives by such principles in practice. (Almost everybody assume, in general, that what we perceive is true.)
The path of science acknowledges that human perception is not perfect, but that does not imply that trying to collect experimental data is meaningless. We should simply do our best to acquire as neutral data as possible free of subjective interpretation. It is not necessarily so difficult. For one thing, practically all experiments are nowadays run with instruments that provide us with numbers, which are in turn handled by computers. There is little room for human error in such data acquisition. Further, if several scientists worldwide can acquire data which looks the same, independently of each other, with different instruments (or even different methods), it is a very good sign that the data is valid.
I also wish to point out that the interpretation of the data is something different than the data itself, although often presented together with the data set in scientific publications. It is important that scientific records explain exactly how the data was acquired and presented, such that other scientists can take the very same data and use it for other purposes, even if that means coming up with a conclusion which contradicts that of the original reserachers performing the experiments. I am sometimes worried about the trend that less and less focus is put on explaining how data was acquired in scientific literature and how it may have been treated before presentation.
Science is Predictions
This is perhaps the most important subtopic in my opinion. How can we distinguish scientific theories about the universe from those that are not scientific? From religion, say Christianity, we get the concept of a God that rules (or at least created) the world. Without going into a discussion on whether this is correct (which I think is normally a fruitless discussion), let us instead consider whether the existence of God is a scientific theory. This is a much more interesting question to ask. The best way to tackle this issue is to consider whether the existence of God makes predictions about the world. If you think about it, this is not the case, at least not for most modern views on what God is. For instance, assuming a "good" God, when good things happen we say that it is the will of God. However, if bad things happen we also say that it is the will of God: We are just not wise enough to understand how they contribute to the greater good! ("God works in mysterious ways.") In this manner, most modern God concepts offer no predictive power whatsoever. Importantly, as a consequence, they cannot be falsified, because whatever happens can be interpreted as in agreement with Gods will. Following this line of thinking it becomes clear that a theory that can explain any hypothetically possible outcome of an experiment is, somewhat counterintuitively, completely worthless! The reason is that this means that the theory cannot be falsified, not even in a thought experiment, which is equivalent to saying that it cannot make any predictions. Hence it is not a scientific theory!
There are also theories from religious sources that actually are truly falsifiable. Thus they are scientific theories (or at least hypetheses) in principle. It is just that existing evidence strongly suggests they are false. In this view, whether a theory is scientific or not has nothing to do with its truthfulness. For instance, from the Bible we can deduce that it claims that the Earth is 6000 years old (which is still a common belief among conservative Christians). The theory makes quite many predictions that we can test, especially in relation to dating of objects. Thus is it scientific, but false because existing evidence falsifies it (carbon dating etc.). One could also make the argument that the theory is not rational from the start - it is just something found in an old book, why would it be true? I have nothing against this argument but strictly speaking it is not until we test the theory by comparing it with empirical data that we can say it is false. The mere fact that such a comparison is possible means the theory is scientific in principle even though it originates from religious source. I thus claim a theory can be scientific or not scientific and that this has nothing to do with whether it is true or false.
Falsifiability and Popper
Closely realted to the previous part is the topic of Popper and his falsifiability criterion. In brief, Popper said that a scientific theory must be possible to falsify, at least in principle. This is essentially equivalent to saying that the theory makes predictions. A theory that allows for anything to happen cannot be used to predict any outcome. Popper's main contribution was to give us a tool to clarify what is science and to identify pseudoscience. Popper has been heavily criticized throughout the years but I am still waiting to see any criticism which really destroys his main point. Here I would like to briefly comment on a few things which are often claimed to show that Popper was wrong:
"Scientists do not work like that in practice, they do not falsify theories." First, Popper never claimed to say how scientists DO work, he said how they SHOULD work. Second, I am actually doubtful about this. At least in my research, I work like Popper said all the time: (1) I formulate a theory about how that system I am studying works. (2) I think of an experiemnt which has an outcome that will either support or disprove the theory. (3) My group performs experiments (or maybe check already existing data) to see if the data is in agreement with the theory. If not, we declare the theory to be false (at least as a preliminary outcome) and think of something else to test instead. Third, the theories that are falsified are usually not "grand scientific theories", they are small ideas that pop up now and then which you will never see in scientific papers because we do not publish stuff that we concluded is false. So it is completely natural that you do not hear in the media very often about scientific theories being debunked.
"Any experiment has many other implicit assumptions in its design when testing a theory, not the least other scientific theories. If the outcome is not in agreement with the theory being tested, it does not necessarily mean that the theory is the problem." It is true that the testing itself contains assumptions in the experimental design, often even other theories. Let us say we test a theory B, which is based on the theory A being true. If the outcome shows "false", how do we know that is is not A which is the problem? The answer is of course that we need to set up additional experiments to properly verify A before taking on B. In the end we build science step by step starting from simple starting points, axioms, assumptions (or whatever you want to call it). We end up with a whole bunch of experimental data and we try to stitch it together, to see which theories are true and which are false, to establish a consistent view as well as we can. Another way to put it is to say that it is still true that our experiment falsifies B in principle, UNDER THE ASSUMPTION that A is true.
"Nobody sane would ever declare a truly established theory like that of gravity or evolution to be false because one or a few experiments suggest so." This is similar to the other points above. I agree! Nobody would and I really really do not think that is what Popper was suggesting either. Maybe an instrument was malfunctioning and that was the real cause of the "failure" of the particular experiment. This does not mean Popper's idea was wrong. It just means we need to be careful when doing science. The key is that Popper talked about the requirement of falsifiability IN PRINCIPLE. If a theory has been validated over and over in the history of science, the TOTAL evidence is still in favor of regarding it as true. Indeed, because of the fact that any experimental methodology will suffer from human error now and then we should NOT expect single experiment ever performed to not falsify the theory.
Science and technology
In addition to science, most researchers nowadays are also involved in applied research, i.e. technology. This is normally with the intention of improving quality of life in one way or another. When we understand how something works and can predict a system response through our models, this information can be utilized in various devices or, in some cases, governing principles. Nowadays technological impact is a major driving force for research advancements since many funding agencies only provide support for projects with a clear technological goal which should preferably also promote industry. This may sound beneficial: Why give money to wierd people who play around with expensive instruments to test their obscure theories that have no connection with everyday life? This is not how to improve society and does not motivate using up its resources. On the other hand, when one looks back at the history of science and the amazing technological achievements of humankind, it is often so that the major breakthroughs have come through curiosity driven research or mere chance. There has not been any clear application in mind during the work, but it emerged as a logical consequence of understanding the phenomenon under study. Further, a significant delay may be involved before the application comes and knowledge that appear useless now may not be so in the future. A well known example is that the Google search engine uses mathematics which was produced many hundred years ago.
The role of science in society
I beleive the role of science and the scientific method is now more important than ever as we have entered the era of extremely quick and personalized information spread. The role of neutral objective research without political motives is a cornerstone to create a functioning global world. My vision is not that scientists become politicians, because then they are no longer in a position to be objective (and they would not have time to do both). Instead I wish that politicians listen to scientists and use scientific results in their argumentation. Some of the responsibility also lies on the scientists themselves in the sense that they must communicate their findings to the outside world.