More Recent Comments

Tuesday, March 04, 2008

Ten Things Everyone Should Know About Science

SciBarCamp is starting in less than two weeks.
In the tradition of BarCamps, otherwise known as "unconferences", (see for more information), the program is decided by the participants at the beginning of the meeting, in the opening reception. Presentations and discussion topics can be proposed here or on the opening night. SciBarCamp will require active participation; while not everybody will present or lead a discussion, everybody will be expected to contribute substantially - this will help make it a really creative event.
Eva Amsen has suggested an interesting topic Ten Things Everyone Should Know About Science. Here's my quick list to get the discussion going ...

  1. Science must adhere to methodological naturalism. Supernatural explanations are not allowed in science.

  2. All scientific models and theories are provisional in the sense that they might be overturned tomorrow. This does not mean that there's a high probability that well-established theories are wrong. It only means that nothing is absolutely proven in science.

  3. Scientists must be skeptics. They must weigh all new data in the light of their current understanding of science. New data must not be accepted unquestionably.

  4. Scientists must never lie about science or deliberately misrepresent it to the general public. No exceptions are allowed, even if a little white lying might be for the common good.

  5. Scientists must never be afraid to criticize other scientists and they must have the freedom to do so without suffering retribution or penalties. Real science only thrives in an atmosphere of freedom of speech.

  6. Theories are the best thing we have in science. A theory is a general explanation of particular phenomena that has withstood many attempts to disprove it. Because of the evidence supporting the explanation and because it hasn't been refuted, a theory will be widely accepted as provisionally correct within the science community.

  7. Science is evidence based and the practice of science follows simple rules of logic and rationality.

  8. Scientific facts must be reproducible or capable of independent confirmation by other scientists.

  9. Science is a way of knowing about the universe. It may be the only epistemologically valid way of knowing. Technology is not the same as science and medicine is not the same as science.

  10. There is no such thing as a rigidly defined scientific method.


  1. A major deficiency is that there's no discussion of what constitutes a "scientific fact" or what a "law" might be.

    If it's aimed at "everyone," it seems to me that these are essential.

  2. "Technology is not the same as science and medicine is not the same as science."

    Are you making the distinction because you view these fields as consumers of science?

  3. You also seem to be making a distinction between data and evidence in that you say that science is evidence based but the all data is provisional.

  4. Don, I don't get that impression at all. He said the models are provisional, the data are decisive. However, the data have to be viewed with a skeptical eye because they do not always mean what they are said to mean, because of errors in interpretation, practice, or experimental design.

  5. I think the concept of fact in Science should be expanded on extensively. A good place to start is:
    Stephen J. Gould, " Evolution as Fact and Theory";
    Discover, May 1981

    "In science "fact" can only mean "confirmed to such a degree that it would be perverse to withhold provisional consent.""

  6. The ten items seem like a good start to a definition of science - at the very least they are a necessary condition. Do you think they are sufficient?

    Or, put another way, if we strive to adehere to naturalism, be skeptical, not lie, be critical, follow the rules of logic, etc., are we therefore doing science?

    I know people who would claim to do all these things, and yet they are happy to tell me that "science is not the answer". Examples would be (some) sociologists, or (some)religious apologists. They only think they are not doing science, but in fact they buy in to all the conditions.

    (First post, been lurking for a while, go easy!)

  7. I see two sentences:
    "New data must not be accepted unquestionably."
    "Science is evidence based..."

    These imply that evidence is more than data. That's all I was saying.

  8. I don't think there is much difference between data and evidence, except perhaps that data can be seen as the raw material from which evidence is constructed. In so far as data needs to be sort of fleshed out into evidence, evidence is "more" than data.

  9. I see those sentences as well. I'm not sure why you believe there is some divide between them. (3) is just a way of saying "all data are created equal". The old data don't go away just because we collected new data. That we interpret new data in light of existing data is a consequence of being evidence-based, not a contradiction to it.

  10. The list looks like an attempt to define science. If so it is way too long. Science is in many ways defined by current culture and that it works. This would revolve around the part of point 9 that observes on ways to validate belief (which is knowledge).

    Supernatural explanations are not allowed in science.

    Yes and no. Willful agents, miracles and hearsay gets kicked out by the requirement of repeatable observations. They don't work.

    nothing is absolutely proven in science.

    Neither is anything entirely certain, since observations and theories have uncertainties.

    Scientists must never lie about science

    Oh, they can and will lie ("culture"). But (free) science is self correcting.

    A theory is a general explanation

    and it owns its facts. So it is better than a single fact, in fact better than the set of facts it contains.

    Technology is not the same as science and medicine is not the same as science.

    If technology or EBM isn't as good, it is mostly because the distributions of characteristics aren't as clearly separable between different realized instantiations as in idealized science.

    There is no such thing as a rigidly defined scientific method.

    But most everyone agree that testability is a necessary requirement. OTOH people can't agree on what it means... uh oh.

  11. #8 ...presumably "confirmation" rather than "conformation" (some typos are more dangerous than others!)

  12. I went back through the list and constructed the "anti-science" way to knowledge:
    1- Appeal to supernatural or occult explanations
    2- Rigid, enforced and unquestioned adherence to received explanations
    3- Lying and dishonesty when presenting these explanations to the public

    Of course, we will all say this is no way to make any progress towards useful knowledge.

    My question is, if you drop the third item, could you sincerely believe that the anti-science approach would work? If so, would the only reason to reject it would be for the pragmatic reasons that we tried it before (till 400 years ago) and it just didn't work?

    Another question: What does this say about our world that the science approach works and the anti-science approach doesn't?

  13. Number 10 contradicts number 1. Number 1 is a rigidly defined requirement for doing science. If it is not, why is the word “must” there? Now if number 1 is a tentative working heuristic that can be changed should the evidence lead in that direction then change the wording to so indicate. If not, then delete number 10.

    I agree with number 2.

    Number 3 is problematic with respect to number 1. What if the questioning of new data leads a scientist to suspect that there is no naturalistic explanation for those data? Can that scientist pursue that line of questioning, or must he devote his efforts to make the data fit the constraints imposed by number 1. Or must he resign as a “scientist” since he will no longer be doing science?

    I wholeheartedly agree with statement 4 that scientists must never lie about science or deliberately misrepresent it to the general public. NO EXCEPTIONS ARE ALLOWED – except when it comes to the theory of evolution. So what is it called when scientists join forces to direct school boards to implement science standards that don’t allow the presentation of evidence that question certain aspects of the theory? Allowing presentation of only supporting evidence is at best a distortion of the truth if it is not outright lying. And what is it called when scientists claim that the theory of evolution is the fundamental unifying principle of biology when the fundamental claim of the theory – that ALL of life’s diversity and complexity is the result of random processes – has not yet been demonstrated?

    Regarding statement 5, the problem is not that scientists are afraid to criticize other scientists; it’s the fact that they do so with a vengeance when it comes to scientists who question the theory of evolution. Do I need to list the scientists who have been denied tenure, lost their jobs, had their teaching assignments changed, or have otherwise been harassed for questioning the theory of evolution or offering alternative explanations such as intelligent design? Yes indeed! Real science only thrives in an atmosphere of freedom of speech. Scientists should be allowed to question reigning paradigms without fear of reprisal!

    If a theory is general explanation of particular phenomena that has withstood the many attempts to disprove it (the word “disprove” is not appropriate since it was stated in number 2 that nothing is absolutely proven in science), then I would question if the “theory” of evolution has even risen to the status of a theory. As I have said, the theory has not explained what it purports to explain – that all of life’s diversity and complexity is the result of random variation and natural selection. It is at best an interesting hypothesis.

    Number 8 is ambiguously worded. I’m not sure what a reproducible fact is. I can understand what is meant by a reproducible experiment from which facts about nature are subsequently derived. Having said that, I am always troubled by the fact that many things that are of scientific interest are not reproducible and rarely does the person who talks about reproducibility expand on this point -- the Big Bang, the origin of life, and the origin of new body plans, complex biological structures, and new organs, for example. Origins of things are obviously not reproducible, so something needs to be said about the methodologies of origins or historical science.

    And finally, if science is the only way of knowing about the universe and science is restricted by rule number 1, how do we know that science has given us the truth, the whole truth, and nothing but the truth about the universe? My definition of science is that it is that enterprise that generates truth statements about the universe and the natural world. I submit that your definition of science will not do that.

  14. I question number 9. An engineer or an artist often doesn't know what is possible to exist in the universe until they try all sorts of creative approaches. It's not knowledge on the same low level as science, but still knowledge nevertheless.

  15. doublee seems to have some difficulty in understanding evolution. It seems that doublee would like some "evidence" to be presented against evolution. This must mean some form of creationist "evidence" (YEC, OEC, IDC) as they seem to be the only ones who are scared out their underhosen by evolution. Sorry doublee, but they have no evidence to present (see the Dover trial results where Behe was forced to admit that he does absolutely no scientific research to support his IDC claims, and that according to how he would like to redefine science, astrology is right on par with IDC). Here is a short summary of creationist science doctrine so that you may better appreciate their methods.

    The Creation “Science” 10 Commandments (with apologies to Larry Moran):

    1.Creationist “scientists” and proponents MUST adhere to faith-based credulous speculation. Scientific research and explanations are absolutely NOT allowed in any form of creationism (YEC, OEC, IDC).
    2.All creationist ideas and notions are final and not open to revision or debate under any circumstances (especially from those nasty real scientists). This means that there is no probability that their dogmatic ideas can be considered wrong. It means that everything is absolutely “proven” in creationist “science”.
    3.Creation “scientists” and proponents must be credulous believers. They are required to be excessively critical of real science. No scientific data is produced by creation “science” so this greatly simplifies their acceptance of creationism. New data from real science must not be accepted at all, and can only be utilized for misrepresentation and quote mining to support creationism.
    4.Creation “scientists” must always lie about science and deliberately misrepresent it to the general public. No exceptions are allowed; the bigger the lie the more it furthers the cause.
    5.Creation “scientists” must never be afraid to criticize real scientists and they must have the freedom to do so without suffering retribution or penalties. Creationist proponents must NEVER criticize creation “scientists”. Creation “science” only thrives in an atmosphere of ignorance, repression of real science and free speech, and poor quality education.
    6.Ignorance, repression and misrepresentation of real science and a poorly educated public are the best things creationists have in bag of tricks. A “theory” according to creationists is a vague guess about some aspect of science (evolution in particular) they do not accept; it does not matter that it has withstood many attempts to disprove it, the theory must be continually criticized, even though creation “science” cannot offer even the slightest amount of alternative evidence. In spite of the complete lack evidence supporting creation “science” and that it has been completely refuted by real science, creation “science” must be widely accepted by everyone as totally, completely and indisputably correct. [The exception for the definition of the word theory is when creationists compare their irrational ideas and notions to real science, in which case they demand that their “theories” should somehow be considered equal to real scientific theories].
    7.Creation “science” is faith based and the practice of creation “science” follows the most irrational rules of logic possible.
    8.Creation “scientific facts” do not exist and are therefore not reproducible or capable of independent conformation by any scientists, but that does not matter.
    9.Creation “science” is a way of NOT knowing about the universe. Although only one of many inferior “ways of NOT knowing” it is the the only faith-based way of NOT knowing that creationists are allowed to accept. Technology and medicine are OK if they can be misrepresented and warped to provide “evidence” for creation “science”.
    10.There is no such thing as scientific method in creation “science”.

  16. Anonymous:

    It is only within the past year that I have started posting responses to blogs relating to evolution and intelligent design. As I have observed responses it seems that more often than not the respondent rarely addresses the specific issues raised in the post he is criticizing. The respondent makes unwarranted assumptions about the poster and posts a response based on faulty assumptions. Your response follows this pattern. (I’m curious. By your definition, is a creationist anyone who doubts the theory of evolution?)

    I was hoping for a discussion regarding what science is verses what I think science should be. A diatribe against Creationists is not helpful in this regard. In fact, I see a parallel between the Creationists approach to science and the evolutionists approach to science. Much of what you say about Creationists, with adept word substitution, I could say about evolutionists. One thing I will say, however. Both begin with an a priori commitment to what kind of answer is acceptable. Creationists must find an answer that supports their interpretation of the Bible. Evolutionists must find an answer that is consistent with methodological naturalism. Will either approach give us the truth about nature? What about an a priori commitment to following the evidence wherever it leads? Nevertheless I offer my response to your post.

    As a matter of fact, I do have a lot of trouble understanding evolution. The evolution I am talking about is not the evolution that can be observed in the laboratory or in the field, but the evolution of major body plans, new organs, and coordinated biological systems. Apparently scientists are having trouble understanding this evolution as well.

    At least one scientist admits as much. As long as you opened the door with a reference to the Dover trial, I offer the following exchange during the cross examination of Dr. Kenneth Miller. (Sept. 26, 2005, p. 47, line 12)

    "Q. Is it true that scientists do not know enough about all structures in the cell to describe how they all work or how describe how evolution could have produced each of them by step-by-step Darwinian processes?

    "A. Well, you ask a very interesting question. And I, first of all, am going to enthusiastically agree with the first part, which is that scientists certainly do not understand enough about all of the structures in the living cell to understand how they work. That really is the business, my business and the business of Dr. Behe. Because the answers to that questions are going to come out of genetics -- sorry. They're going to come out of biochemistry. They're going to come out of cell biology and maybe molecular biology and genetics as well. I'll answer the second part of your question this way. Until we understand the first part, which is how everything works, we can't even begin to understand how things evolved. So we will have to have an absolute and complete and total understanding of how everything in the cell works before we can even begin to put together an understanding of how it evolved."

    To repeat for emphasis:

    “So we will have to have an absolute and complete and total understanding of how everything in the cell works before we can even begin to put together an understanding of how it evolved."

    (Yet, Dr. Miller claims to have rebutted Dr. Behe’s claims that the bacterial flagellum is irreducibly complex.)

    So if science truly does not know how evolution works, how can it go before the public and claim that it does? (This is not a rhetorical question. I would appreciate an answer.)

    According to number 4:

    Scientists must never lie about science or deliberately misrepresent it to the general public. No exceptions are allowed, even if a little white lying might be for the common good.

    Just as science does recognize the problems with developing a naturalistic theory of the origin of life, so should science recognize publicly that it does not yet have the answers about the evolution of life. Yes, if you read the fine print scientists do admit that they are still investigating the mechanisms of evolution. But the fine print is not what the public hears. The public hears things like “Evolution is a solid theory that has withstood the test of time and many attempts to disprove it.”

    But this is not fine print to me. This is the crux of the problem. Can the mechanism of random undirected processes account for the diversity and complexity of life or not?

    Anyone investigating a process that appears to be random needs to consider the possibility that the process may not be random. How can anyone claim to be following the scientific method if that person rules out a possible result of that investigation because he won’t like that result?

    Is it possible to come to the conclusion that random, undirected processes cannot explain the complexity and diversity of life and that intelligent agency is required? If it is not possible, how does science know that? By an appeal to the definition of science?

    So it seems that we have truth by definition and not truth by investigation. And this is the fundamental point of my post.

  17. In response to Doublee:

    Creationism and ID don't add anything to the discussion. They offer nothing to advance our knowledge, but only the sewing together of random facts to justify a position that will never be questioned. And sure, there are many things we can't explain to the nth degree of detail, many gaps in our understanding, but to point them out doesn't make the alternative explanations valid.

    Also, as has been pointed out before, show an evolutionist evidence of rabbits in the pre-Cambrian and he will disavow evolution. What would it take for a creationist or IDist to do the same? If there isn't an answer, then these have no place in science.

  18. Don:

    Dr. Miller says we don’t know enough about how the cell works to even begin to explain how it could have evolved.

    You say there are many gaps in our understanding and to point them out doesn’t make alternative theories valid.

    I agree that pointing out gaps doesn’t make alternative theories valid. (It makes me suspicious that they might be valid, however.) My purpose is to point out gaps and wonder why that, in spite of these gaps, science can claim that the “theory” of evolution itself is valid. Or as Michael Ruse has asserted, “Evolution is a fact, Fact, FACT!”

    As far as I am concerned the theory rises or falls on whether or not it can be demonstrated that random processes are indeed responsible for all of the complexity and diversity observed in living things. The bottom line is that evolution has not explained what it purports to explain. If a theory can’t explain what it purports to explain, then there is no theory (i.e., explanation). So, please tell me. How does science infer from the evidence that it does have that random processes are responsible for the complexity and diversity of life, other than that is the only explanation allowed under rule number 1?

    I’m not quite sure how a pre-Cambrian rabbit would falsify the theory of evolution. Already the fossil record is replete with fossils of animals that have no apparent ancestors. I envision a reaction such as “Boy, evolution is more clever than we had ever thought.” And with the rule of science rigidly in place, would science immediately jump to the conclusion that an intelligent agent was responsible and say you ID guys were right after all?

    So in lieu of the pre-Cambrian rabbit, consider this thought experiment. Someday all the greatest biologists, philosophers, and computer programmers in the world gather around a super computer and run the ultimate evolutionary algorithm. This ultimate algorithm must have no genetic information smuggled in by the programmers, and it must truly emulate the random processes claimed to be responsible for evolution.

    According to the definition of science, this algorithm can generate only one possible result. That is, it must confirm that random processes are indeed responsible for all the complexity and diversity observed in nature.

    Is this really what science is all about? According to rule number 1 it is. Science MUST adhere to methodological naturalism. The word “must” turns the rule into an oxymoron. The rule says that no evidence will or can be discovered that would suggest that there could be influence on nature that comes from outside nature.

    What if the algorithm doesn’t yield the defined result? I seem to recall that Michael Ruse allowed that the explanation for evolution could be found outside of nature. But, he said, such an explanation would not be a scientific one. This raises another interesting question.

    Can the scientific method (as embodied in the super computer experiment) yield an answer that is not scientific?

    What if it is true that intelligent agency is the source of the complexity found in nature? (Even Judge Jones allowed that ID could be true.) What institution would determine that, if not science?

  19. In response to doublee:

    With the pre-Cambrian rabbit, I'm talking about the facts of evolution.

    Here are the facts:
    - Given any species, there was a time that it didn't exist.
    - Sometime after this point, there will be a time that the species doesn't exist.
    - The current species are descendants of the species of the past.

    A rabbit in the pre-Cambrian would definitely fly in the face of the two facts and challenge the third.

    As for the theory of evolution by natural selection, the current debate among the scientists is the contribution this makes to evolution, among the other competing forces (e.g. genetic drift); however, no one is proposing ID because the evidence is better explained without introducing intelligence and introducing supernatural intelligence is untestable.

    If all you are saying is that you find this theory of evolution unsatisfactory because it doesn't answer all the questions of mechanism and origin, then I say fine, let's go find the answers. Just stick with procedures that have been shown to work - methodological naturalism. If there's an oxymoron, it's "outside nature".