Philosophy of Science
What is experiment? theory? explanation? natural law?
This discussion is based on the book Philosophy of Science, The Natural Sciences in Christian Perspective, by Del Ratzsch. (See book list at the end of ch. 5.) I am grateful for what I learned from his book. Any faults in the following are of course my own.
The philosophy of science is a vast, well-developed field. A few pages here can only give a bare introduction, and hopefully an awareness that such a field exists for further study if interested. It is of course a field in which the practitioners hold a wide variety of incompatible opinions, but what follows is mostly a common foundation on which they agree.
If you find yourself getting lost here, never mind. Go on to ch. 3, and refer back if necessary later.
If asked “What is science?” most people will respond with something resembling “observation, deduction, prediction, experiment, confirmation, modification.” This series of steps is taught numerous times in the course of elementary and secondary education, with a wide variety in the details but the same basic content. This is the scientific method, but you must have some sort of basic concepts, a goal and purpose, before there can be a method. A method must have a goal, or program. These concepts are overlooked in the educational process, even though they are simple and could easily be taught in elementary school. They are implied in the method, but it is a rare individual who will realize this on his own, or question it. You have to take a philosophy of science course in college before anyone will mention them explicitly.
Natural science requires two basic assumptions: that nature is
1 uniform or consistent in time and space, andFor those with a philosophical frame of mind, other assumptions may also be included. An even more basic one, which on this reckoning must be labeled the 0th, is that the physical world really exists and is not a figment of our imagination. For most of us this need not be said; we constantly bump into an objective environment which stubbornly refuses to conform to some of our expectations, let alone wishes. This discrepancy between our expectations and reality results in what we call accidents. Only people with advanced education in philosophy have difficulty comprehending this.
2 intelligible to human minds.
Another added assumption would be labeled the 3rd, that our senses give us reliable information about the world. This of course does not ignore the many aspects of physical reality which are not directly detected by our senses: high- and low-frequency sounds, electromagnetic waves outside the visible range, and so on. And there are the countless fascinating illusions and misperceptions with which magicians earn their living, and some psychologists earn their living by analyzing such perceptual effects. But these can be guarded against and double-checked, so that they do not constitute sufficient cause to make us doubt that any perception is ever meaningful.
Notice that these points are assumptions. That means you cannot ask “Why?” To ask reveals that you do not understand what an assumption is. It is a starting point. If it has a reason, then it is not an assumption, but a conclusion from something else, and that something else is perhaps an assumption. Somewhere there must be a starting point, which can only be believed. To some people this is a startling realization, because they have understood science to be an edifice founded only on verifiable fact, and considered unproven assumptions as a characteristic feature of religious thought. One important point of this entire book is that there is no such clear dichotomy between science and faith.
In our modern scientific age these two (or more) basic assumptions are usually taken for granted, considered obvious, never even stated explicitly, just swept under the rug of “common sense.” But they are not obvious or common at all. If they are, then why did science not begin some other time and place than 16th-century Europe? This question is answered in ch. 5, I and II.
The first assumption basically says that natural laws exist, which of course must be assumed before you construct a method and spend your time and money attempting to find those laws. It cannot be taken for granted that nature has laws. Our world often appears very disorderly, and we must make a leap of sizable faith to believe that behind the apparent chaos are some basic “laws.” If nature has laws, why isn’t the weather forecast more reliable?
The very use of the term “law” in reference to inanimate nature has profound implications and a fascinating history, which is beyond our scope in this book.
This assumption is just that - an assumption - and the best that we can ever achieve is to say that it has succeeded amazingly well so far, for several centuries. But we cannot prove it will succeed when research treads into uncharted territory tomorrow.
The second assumption is similarly uncommon sense. The human brain is a mysterious blob of tangled nerve cells imprisoned in the darkness of our cranium, with complex electrical and chemical interactions. Its information about the outside world depends on signals from sense organs elsewhere in the body. What connection might there be between this inner world and the outer one? Could there, or should there, be a connection? The more you think about it…
Natural science is defined as study and research which is based on these two assumptions, leading to three important characteristics:
1 empirical (experimental if possible),The first and second characteristics apply assumption 1, and the third applies assumption 2.
2 objective (public), and
3 rational (often mathematical).
These requirements are somewhat flexible. Consider the first requirement. Astronomy is undeniably a part of natural science, but heavenly bodies are not accessible for experimentation. The best we can do is refer observations of their properties to the observed properties of materials and processes in our laboratories. Scientific analysis of historical artifacts is another borderline area, where the actual past of these objects cannot be reproduced.
The second requirement is fairly straightforward; the observation under consideration should be accessible to anyone with the requisite education and equipment. If it still seems to be apparent to one observer and not to another, then perhaps it is a subject for research by psychology rather than natural science.
The third is the most difficult to pin down, as will be discussed later. The mathematical formulations of Newton, which compose most of the content of physics and chemistry, have become the standard example of “natural science.” But biology is also natural science, and it consists mostly of diagrams and verbal statements about events and processes. Astronomy too has much that is of a similar nature.
This definition of the characteristics of natural science must not be taken as a value judgment, making such topics in any way more valid or important than other topics. It simply means that topics which do not fit these criteria are not natural science.
And finally, based on all these concepts, we have developed the “scientific method,” as mentioned at the beginning.
These assumptions and characteristics of natural science are historically traced back to the British philosopher Francis Bacon (1561-1626), who said that the way to do scientific research is to:
1 collect data (empirical),These concepts were the beginning of modern science. However, Bacon himself did no scientific research, and his plan is impossible, because:
2 organize it without presuppositions (objective), then
3 form principles by induction (rational).
We must have some presupposed theory to guide the selection and organization of data (but then can we still really claim the results are empirical?).
Induction requires imagination and creativity; it is not pure logic (but then is it really objective?). In modern terms, induction cannot be done by a computer program.
There may be many theories that fit the data, and we must choose one, guided by presupposed standards (but then can we call it rational?).
Does this mean science is not empirical, objective, and rational? It seems to be in trouble. But it has carried on very actively anyway. Bacon’s concepts and plan were hopelessly impractical, yet Bacon is honored as the father of modern science. A wrong start is better than none at all.
Ratzsch discusses the viewpoints of various philosophers of science who tried to be realistic about the difficulties in these concepts and still preserve confidence that science is empirical, objective, and rational. Great philosophers for several centuries struggled with the concepts of reality and rationality. What is real? How do we know it? Descartes, Kant, Hume, etc., refuted others’ answers and proposed various answers, but each in his turn was also found to have flaws. There was no really final solution. In the early 20th century, the general opinion was:
1 The final authority in science is still observation of nature. This is empirical.But this still has some problems. Explanation requires causes, not just the ability to predict. Personal and social factors influence what we observe and what theory we choose. And there are other problems, which is why philosophy of science is still the way some people earn their living.
2 Neutral observers all see the same data. The scientific community can review, repeat if possible, and correct errors. This is objective.
3 We use deduction from general laws and initial conditions. Before the event, this is prediction; after the event, it is explanation. We can observe only a small part, a sample, of all of reality. We notice some patterns, assume this is representative of all of reality (uniformity), and form a theory. More experimentation and observation observes a larger part of reality, and the theory is either confirmed or rejected. This is rational.
Thomas Kuhn in 1962 published The Structure of Scientific Revolutions, an analysis of the history of science. His key concept is the paradigm: an accepted, assumed standard system of thought, theory, and research method, such as Aristotle’s physics, or Newton’s. Unlike the idealistic stereotype, scientists are not usually looking for new theories but for ways to prove that their data fit their present paradigm. After devoting years of intense effort to mastering the current paradigm, they are not anxious to overthrow it all!
Most data fit, but some do not; these are anomalies. Anomalies are at first set aside, considered unimportant. Sometimes they are later solved, by finding a mistake in either the prediction or the experiment. When too many anomalies accumulate, confidence in the paradigm begins to weaken. (How many are too many? Whatever produces a weakening of confidence, which is a very subjective criterion.) This is a crisis. A crisis may result in:
1 a solution that re-establishes the old paradigm with some modification, orKuhn marked a minor revolution in the philosophy of science. He based the study of science on observation of its history, not just on philosophical ideals and abstractions! To put it concisely, he applied the scientific method to science itself. According to his explanation, is science empirical, objective, and rational? Kuhn did not end up saying that all of our scientific accounts of nature and natural laws are produced merely by psychological and sociological factors in the scientific community, but some other people have built on his concepts and extended them to this extreme claim. Such a claim seems inconsistent with the fact that so many experiments do not turn out as expected by the scientific community, which is what Kuhn calls anomalies. There really seems to be an objective world out there beyond our minds, that cares little what some community would like to think about it.
2 abandonment of the anomalies, and return to the old paradigm with a new realization of its limits, leaving the anomalies outside the limits and trusting that they will someday be solved by our grandchildren, or
3 a revolution, in which a new paradigm is proposed and becomes generally accepted (e.g., Copernicus’ heliocentric universe, Galileo’s and Newton’s physics based on experiment and mathematics, biological evolution).
Kuhn of course was not the last word on the philosophy of science. Many deficiencies can be found in his analysis, and other aspects can be raised which are also important. Philosophers of science are not yet in danger of unemployment.
III At present
The scientific community at present admits that “empirical,” “objective,” and “rational” are impossible to define precisely, but it still believes that they are meaningful concepts.
Empirical data come from an external world of nature that often refuses to do what psychological and sociological factors lead everyone to expect (Galileo’s observations, Michelson-Morley experiment, structure of the atom, solar neutrinos, etc.). So it does not seem reasonable to say that all of science is a product of psychological and sociological influences.
Objective/subjective is a spectrum; nothing is perfectly objective, but some data are the same for all observers, and are almost perfectly objective (instrument readings, positions of planets, etc.).
Rationality cannot be defined, and sometimes the entire scientific community has been wrong. But it still must be right to prefer theories that are accurate, consistent, broad, simple, fruitful, etc.
Is a theory just a game that predicts correctly (operationalism), or is it real (realism)? There seems to be a spectrum of reality. It is difficult to deny that planets, atoms, elementary particles, genes, germs, etc., are real. Fields, energy, force, etc., cannot be seen but seem very real. Quarks cannot be seen directly, but indirect evidence keeps accumulating. Many dimensions, other universes, etc., are suggested theoretically but cannot be seen, and it is at present questionable whether they are real. But in Galileo’s day the same was true of the motion of the earth, and one hundred years ago the same was true of electromagnetic waves. It is dangerous to predict what will never be detectable. Such predictions have a notorious and hilarious history.
Some people emphasize the uncertainty of science. They say that many formerly-certain “facts” have later been disproved, and so they conclude that nothing is certain in science, but everything will someday be disproved. This is not quite true. The theories of science can be placed on a spectrum from minor to major. Minor theories deal with a small range of phenomena, and should be called models (shape of a molecule, steps in a chemical reaction, historical origin of present records and objects). Such models always need improvement and correction, and sometimes are completely rejected and replaced by a better model. But major theories have a broad range of observational confirmation, and should not be quickly rejected when one apparent anomaly is discovered (conservation of energy and momentum, laws of heredity, etc.). The people who talk about doubting these laws are usually trying to win acceptance for some other strange theories of their own, and they consider their theories to be certain beyond any doubt! In ch. 7 we discuss cases in which such tactics are sometimes employed in defense of recent creation.
Scientific revolutions since Newton have not totally rejected major theories, but have made them part of new, broader theories. Einstein’s theory agrees with Newton’s laws for velocities far smaller than the speed of light. Spacecraft are still navigated according to Newton’s laws. The conservation of energy, momentum, etc. have withstood countless tests. We can never say it is impossible that we will someday discover that these laws are wrong, but they are correct for a very broad range of phenomena, there is at present no known reason to doubt them, and it is not probable that any such reason will be discovered soon, if ever. While such things technically never attain the status of absolute truth, we can in practice be certain about them.
This ends the discussion based on Del Ratzsch’s book.
We can summarize: What do we mean by “explain”? When a child asks “Why?” what does it take to satisfy him or her? “It just is,” or “just because,” are not explanations. Given several explanations, which is the “best” one? Science explains a phenomenon by describing a deeper, more general phenomenon. For example, the motions of planets in the sky were explained by describing Ptolemy’s geocentric epicycles, then Kepler’s heliocentric ellipses. These were then explained by describing Newton’s gravitation and laws of motion. Gravitation is now explained by Einstein’s general relativity, which relates mass, force, and space-time. But why is there mass, force, and space-time, and why do they interact as they do? Science will always have a deepest level of explanation which is descriptive and unexplained. When we find a new explanation, that becomes the new deepest level. There will always be a currently unanswered “Why?”, never a final answer with no further questions possible.
This is related to the indefinable concept of rationality, discussed above. A theory is never proved, it is only the “best” of the currently available explanations, the simplest, the most general, has the least apparent exceptions, uses the most acceptable assumptions. Our confidence is proportional to how much better it is than the next-best candidate. Overwhelmingly best explanations are called laws. This can be called the logic of explanation, and is the basis on which science operates, as well as history and criminal investigation.
Science has limits. Within those limits it is undeniably important and correct. Science can be defined as the study of the natural world which is uniform and intelligible (2 assumptions), or the natural world can be defined as that which science can study. Science can tell us about the universe and its laws, and can tell us a lot about the past and future as far as the laws of nature apply. But science cannot tell us the origin of the universe and its laws, or the reason why it exists and has these laws. It will also never be possible to prove scientifically that nature is uniform and intelligible nor explain why it is; we can only say that that assumption has succeeded extremely well until now. It is the basic faith on which science is based.
This limitation can be called methodological naturalism. This means that the scientific method is limited to investigating natural processes, or chains of natural cause and effect. Science is limited to studying such chains. By definition, it is the study of what is consistent, or uniform. This in itself does not assert that nothing exists that is not consistent, only that it is not part of what we call nature, and that natural science cannot study it.
Link to next full chapter: Link to next section: