Home » Library » Modern Library » Richard Carrier Scilit

Richard Carrier Scilit

Test Your Scientific Literacy! (2001)

Richard Carrier


Do you think you know what science is? You may be surprised. Scientific literacy is hard to acquire and is not widespread. Science is, after all, a very complex and nuanced affair that can only be truly understood with wide experience and deep thought. It took the whole of human civilization thousands of years to hit upon it, and thousands more to master it. Recent books have explored the unnatural, counter-intuitive, and difficult nature of science, refuting the naive Enlightenment view that science is nothing more than disciplined common sense.[1] We freethinkers and secularists should be keen above all to become fully literate in the nature of science and to encourage the spread of this literacy throughout the general public. Indeed, ignorance of science is a leading cause of support for superstitions, cults, and those religions that shackle the mind, or foster bigotry or worse.

A recent study of the effect of “history of science” courses on improving a student’s understanding of the nature of science sets the groundwork for a simple test of scientific literacy which I have created here.[2] Though there were several flaws in the study that make its conclusions somewhat questionable, the researchers had the right idea, and their results suggested an astonishingly poor understanding of science even among high school science teachers. By “scientific literacy” in this study, and in the following test, what is meant is the nature of science, not its content. People are awash with scientific content. Many scientific facts are common knowledge: the average person on the street knows more scientific facts than the most educated men of antiquity. “Scientific literacy” in the more prosaic sense of simply knowing what scientists have concluded on any given subject is a secondary concern. It does no good to know all the products of science and yet not understand science itself. For it is the nature of science that sets it apart from all other sources of authority, and which is most helpful in instructing our own lives and our own personal pursuit of the truth. To the scientific illiterate, scientific facts can be little more than the oracular pronouncements of a priestly caste of prophets whom we call “scientists,” an analogy I have heard sincerely trumpeted by the religious right more than once (betraying their ignorance of the nature and history of science). We can know all sorts of facts about geology or astronomy or biology, but if we don’t understand how those facts were ascertained and demonstrated, our minds will remain linked with those of our primitive ancestors who let themselves be led by common sense into all manner of errors and superstitions. Scientific literacy is the key to enlightenment.

The Internet Infidels Test of Scientific Literacy

Answer each question with ‘true’ if what the sentence most normally means is typically true and ‘false’ if it is typically false.

 1. Scientists usually expect an experiment to turn out a certain way.

 2. Science only produces tentative conclusions that can change.

 3. Science has one uniform way of conducting research called “the scientific method.”

 4. Scientific theories are explanations and not facts.

 5. When being scientific one must have faith only in what is justified by empirical evidence.

 6.Science is just about the facts, not human interpretations of them.

 7. To be scientific one must conduct experiments.

 8. Scientific theories only change when new information becomes available.

 9. Scientists manipulate their experiments to produce particular results.

10. Science proves facts true in a way that is definitive and final.

11. An experiment can prove a theory true.

12. Science is partly based on beliefs, assumptions, and the nonobservable.

13. Imagination and creativity are used in all stages of scientific investigations.

14. Scientific theories are just ideas about how something works.

15. A scientific law is a theory that has been extensively and thoroughly confirmed.

16. Scientists’ education, background, opinions, disciplinary focus, and basic guiding assumptions and philosophies influence their perception and interpretation of the available data.

17. A scientific law will not change because it has been proven true.

18. An accepted scientific theory is an hypothesis that has been confirmed by considerable evidence and has endured all attempts to disprove it.

19. A scientific law describes relationships among observable phenomena but does not explain them.

20. Science relies on deduction (x entails y) more than induction (x implies y).

21. Scientists invent explanations, models or theoretical entities.

22. Scientists construct theories to guide further research.

23. Scientists accept the existence of theoretical entities that have never been directly observed.

24. Scientific laws are absolute or certain.

Answer Key:

After each correct answer [(T)rue or (F)alse] is a number in parentheses
that identifies the point of discussion below that explains the answers.

1. T (#4  9. T (#4 17. F (#6)     0 wrong = A+
2. T (#1 10. F (#1 18. T (#5     1 wrong = A
3. F (#3 11. F (#4 19. T (#6     2 wrong = A-
4. T (#5 12. T (#2 20. F (#7     3 wrong = B+
5. T (#2 13. T (#7 21. T (#7     4 wrong = B
6. F (#2 14. F (#5 22. T (#5     5 wrong = B-
7. F (#4 15. F (#6 23. T (#7     6 wrong = C
8. F (#5 16. T (#2 24. F (#6     7 wrong = D
8 or more wrong = F


The Sixteen Features of the Nature of Science

Abd-El-Khalick and Lederman defined the nature of science as comprised primarily of sixteen particular features. These were expanded from seven more basic features that they believed were “noncontroversial” and sufficiently general that they could be taught to all students during a K-12 course of education, whether focussed on a future in science or not. The seven basic elements of scientic knowledge are that it is (a) tentative (subject to change); (b) empirically-based (based on and/or derived from observations of the natural world); (c) subjective (theory-laden)[3]; (d) partially based on human imagination, creativity, and inference (i.e. inductive vs. deductive reasoning); and (e) socially and culturally embedded.  In addition, they added “two additional important aspects: the distinction between observation and inference, and the functions of, and relationship between scientific theories and laws.,” These seven factors will become clearer in their meaning as we examine the sixteen more specific features derived from them.  I now turn to seven aspects of the nature of science that, together, explain (without exactly corresponding to) the seven factors defined by Abd-El-Khalick and Lederman above.

1. Science is a Tentative Enterprise

Abd-El-Khalick and Lederman classified many statements made by their test subjects as denying the tentative nature of science that we might suspect as not suggesting such a conclusion. For instance, one subject responded that “What makes science different from other disciplines of inquiry is the fact that it holds universal truths rather than a view of the truth according to certain individuals.” In a sense, this statement is correct, but it was ruled incorrect because it overstates the case. First, science is the product of the judgements of individuals, a fact that must never be forgotten: the special virtue of science is not that its results are decisive, superior to all others by some fiat, but that it requires individuals to defend their conclusions by extensive appeals to evidence and reason. There is commonly a confusion between the terms “universal” and “objective,” as well as about what it means for something to be “objectively true.”  The reality is that objectivity entails arriving at conclusions based on premises whose truth can at least in principle be agreed upon by all parties, rather than importing premises which are only ever true for individuals (such as intuitions and emotional truths), and an “objective” truth is one that can be agreed upon by all fully-informed observers. Second, the truths of science are only universal in the sense of being applicable to everyone, but they are not absolutes. They could be in error, and are thus subject to refinement and alteration. As one respondent put it, “one thing that is certain about science is that it will always change.”

This is one element of science that most people have a hard time appreciating. Common sense tells us that we need absolute truths independent of individual judgement, because tentative truths are risky, not worth the effort, somehow “not really true,” and individuals are biased, fallible and limited in their abilities. Indeed, many conclude that if there are no absolute truths, then there is no knowledge, that if we can’t know something for certain, we don’t know it at all. Indeed, some critics of science have outright said that science is to be disgarded because it is always changing. Others even say that since all truth comes from individuals, all truth is subjective and therefore equivalent in merit. But these are absurd conclusions. History has shown that tentative knowledge is extremely useful if it turns out correct with a very high frequency. That frequency does not have to be 100% for it to be useful or “correct” in a practical sense. A gambler doesn’t need a sure thing to see a really good bet.

In fact, the tentativeness of science is counter-intuitively one of its greatest virtues: for progress toward truth and away from error would never be possible without change, and change would never be possible without perpetual doubt and skepticism (as I note in my essay Do Religious Life and Critical Thought Need Each Other?). But science does not simply undergo any arbitrary change, as religious ideology or clothing fashions do. When science changes its conclusions, it always does so as a result of further inquiry, and not just any inquiry, but rational, observational, or experimental investigations. And thus, science changes in response to those kinds of discoveries that are more likely to advance us toward truths about the world. Moreover, science’s dependence on individuals is not the handicap that such dependence is in other endeavors, since it requires individuals to persuade others with sound evidence and reasoning, and only when an overwhelming majority of those expert in a field are convinced does science declare something “true.” This truth is not definitive or final, but it is the most reliable bet around. No other means of inquiry is as successful or as trustworthy, as history has shown. Science sets the highest bar, requiring the highest standards of verification, employing the most experienced and well-trained judges, and that is what makes it scientific (and superior to all other endeavors that claim to produce knowledge). Moreover, science does not conceal its evidence or rest its case on mysteries or private revelations, but makes everything public, so that all experts, and even the layman, can examine and weigh the claims and arguments of scientists to an extent not possible in any other field, creating the most effective check against individual bias that humans can devise.

2. Science is an Empirical “Faith”

Abd-El-Khalick and Lederman regarded as naive the claim that science uses facts to prove the truth of scientific claims. Intuitively, we might object to that classification, for surely that is what science does. Here I think they may have interpreted the words of their subjects too literally, for we can certainly say this without meaning, for example, “prove” in a mathematical or deductive sense. But this is just the sort of qualification that we should immediately understand and always express, or at least have in mind, when thinking about or discussing science. Scientific “proofs” are not the same things as logical proofs. Science relies on induction and inference far too much for its results to be equated with those of deductive logic, and scientists are perfectly comfortable with that, as discussed in point #1 above.

But the researchers here wanted to make sure that people understood that other factors besides plain observation determine scientific results. First, science encompasses many unobservables like magnetic fields, which are only observable indirectly, and many basic assumptions that are founded only on general experience (such as that the basic rules of induction lead to useful approximations to the truth). Thus, theoretical entities and background assumptions are an essential feature of science. Second, social or cultural factors can influence science in terms of directing what it studies and how, and can adversely affect it by supporting invalid biases or assumptions, and it can also beneficially affect it by inspiring new, more successful ideas. Since science is fundamentally an interpretive activity, and not merely a collection and presentation of facts, science is never immune to such tainting factors as culture or desire and therefore a constant vigilance against their influence upon any final analysis is an essential component of the nature of science. This is a fact we must never forget, and yet we can easily forget it if we think even for a moment that science is all about “just the facts” without human interpretation.

This is another counter-intuitive feature of science: it is thoroughly empirical, as in based on observation and evidence, yet “empiricism” is not observation and evidence alone, but a view of things that is constructed from observed facts. On the one hand science requires faith, a faith “that certain principles or certain bodies of knowledge are in fact true.” On the other hand, this faith is not blind, for, as one subject put it, “you are going to have to back it up with some empirical evidence.” This is a seeming contradiction that trips up many people. Science builds up faith in its concepts, principles, and conclusions through repeated practice or testing, and when its faith is challenged it returns again to examine the facts and see if its faith is justified by them. This is what makes science an empirical enterprise, the fact that it ultimately grounds and justifies its faith by appeal to observable evidence. The idea of an empirically-based faith is hard for many people to grasp, especially if they have been raised or indoctrinated into believing that “faith” is only a reason for believing something when you don’t have evidence. The term “faith” does have both connotations, meaning “belief” but also “reason to believe.” Science has no use for the latter, but it is not true that a scientist “has no faith” in science: he has faith in it, but a faith that is grounded in empirical evidence and reasoning.  By confusing the two notions of faith, common sense creates a false dichotomy between faith and empirical justification. Science unites them.[4]

3. Science is Not a Single Method

Many people assume there is such a thing as a singular “scientific method” that can be talked about as if it were a set of known and established procedures that all the sciences follow. It is in some sense true that science by definition follows a certain process of identifying a problem, gathering relevant data, formulating a hypothesis, and testing the predictions entailed by that hypothesis, or in another formulation “adduction, deduction, induction”: we adduce a hypothesis from some collection of data and questions about that data, then deduce what new facts that hypothesis must entail if it were true, and then employ any of a variety of inductive methods to test that hypothesis by seeing if these new predictions hold up. However, this procedure does not automatically produce “science,” for the devil is in the details: the specific possibilities are endless, and some particular methods are good and others bad. Science seeks those methods that work well, and progressively abandons those that don’t, especially those that are found to produce misleading results. But the methods that work well will vary according to what is being studied, and so “the scientific method” ends up a multifarious hodge podge of “methods,” some shared across fields, some specific to certain subjects of study, and these methods are themselves always subject to scrutiny and change over time, as their efficacy, or lack thereof, is discovered, and as new effective methods are found out. The exact same problem can even be explored using several different methods, and in fact scientists often seek to use several, as there is even greater certainty in the results if those same results are arrived at by completely different means.

Thus, those who talk about “the scientific method” are often unwittingly expressing a naive view of the nature of science. It seems unintuitive to say that there is no scientific method when we know there is a common general process in deploying the methods of science for getting at the truth. We talk about “the scientific method” all the time without stopping to consider just what it means, or even if it is what we should be saying, because we naturally assume there is a single method shared by all scientists that makes what they do so successful at producing reliable and confirmable results. What most scientific authors, including me, mean by the phrase is the whole gamut of means and methods employed in science, not a single uniform methodology, but many people don’t know that. Thus, apart from the reality of a tacit “metamethod” that has inspired a search for an endless variety of particular methods in science, “the scientific method” can be as much a misnomer as “the artistic method,” for though one can say a painter typically begins with an inspiration, then a tracing, then paints the background, then the details, then applies the finish, there is in fact no one way to paint a picture, and knowing these obvious steps tells us nothing about how any actual painting was done, or how we could paint a picture ourselves. Likewise, knowing the steps adduction-deduction-induction does not make us a scientist, for we still don’t know what actual methods to follow if we want to carry out these three steps successfully for any given problem. This is why it is often difficult for a scientist to cross fields. A biochemist’s methods do not resemble a psychologist’s methods anywhere near enough for one to guess just how the other should best pursue his research, unless they each had some prior acquaintance with the other’s field. Even some common concepts, like the use of control groups or double-blind experiments, cannot be applied to all scientific problems, so that we cannot define the scientific as that which is discovered through the use of control groups or double-blind experiments (we would in one sweep toss astronomy out as unscientific if we did so), or any other subset of methods. But this is another strength of science: it is not only testing facts for truth, but testing methods for accuracy, and thus is the only endeavor we have that is constantly devoted to finding the best means of ascertaining the truth. This is one of the reasons why science is so successful, and its results so authoritative.

4. Experiments are a Goal-Oriented Form of Scientific Observation

Science is closely associated by everyone with the idea of conducting experiments, but the role of the experiment in science is not well understood. Three common naive views of science are: (1) science requires the use of experiments; (2) scientists have no idea how those experiments will turn out; and (3) an experiment can prove a theory true. All are false.

(1) Science is based on observations. Observing experiments is merely one variety of this activity. Zoology, for instance, relies heavily on observation alone, and in fact often seeks to avoid any hint of an experimenter’s influence on the animal being observed or its environment. Thus, experiments are merely one tool in a scientist’s arsenal for collecting observations. The utility of experiments lies in their controllability, not their repeatability per se, since observations can be repeatable without experiments, as again the astronomer well knows. By setting up an experiment, a scientist can control what affects the process being observed and thus can see exactly which factors have which effects. This is clearly useful, but it is not essential. Many scientific facts can be established, and theories validated, without using experiments, a fact that many a creationist fails to appreciate. Geologists cannot exactly create mountains or continental drift in a laboratory. Yet they can confirm with extensive and overwhelmingly-convincing evidence how these things come about, simply by observing nature.

(2) Experiments are goal-oriented procedures. As one subject put it, “if you are going to organize the experiment you sort of need to know what you are looking for.” As the researchers note, prior expectations play a crucial role in designing and conducting scientific experiments. Rarely are experiments set up just to see whatever happens. Rather, the typical procedure is for a scientist to have a hypothesis or theory already in hand, and thus already to expect a certain outcome of a certain arrangement of events. Indeed, that is the point of the experiment: to see if the intended outcome materializes or not. The experiment may still surprise the scientist by not doing what it was predicted to do, but even then that possibility was usually already anticipated in the form of a null-hypothesis, i.e. the result the scientist would expect if the theory being tested is false. In fact, the very purpose of most experiments is to create an opportunity for an observer to control the outcome by manipulating the variables that come into play, including their environment. However, contrary to our common sense reaction, this does not mean that the scientist proves himself right by manipulating the data. The data will confirm or disconfirm the theory being tested whether the scientist likes it or not. Nor does it mean that the manipulation of the experiment invalidates the results. To the contrary, it is the manipulation of the experiment that validates the results, by eliminating uncertain causal factors and limiting the observation to a controlled set of causes that the observer can clearly identify and trace. This is yet another counter-intuitive feature of science: that an artificial set of circumstances can teach us about the natural world seems illogical, yet we have discovered it is true.

(3) No single experiment can prove a theory true. Indeed, it is only in a colloquial and tentative sense that we can say a theory is ever “proven.” As one subject said, “an experiment is a controlled approach to test the validity of a theory or hypothesis. No experiment can ever fully validate a theory as fact.” As a result, “experiments are constantly refined in an attempt to substantiate the implications of a theory.” A well-established theory, which we would normally call “true” in casual conversation, is one whose every notable implication has been repeatedly tested by numerous different means. This falls back to point #1 above, the tentativeness of science. But here the implications of that tentativeness must be drawn out: many experiments and observations are necessary to prove a point scientifically, not just one, and most preferably by completely different scientists or scientific teams. This is an important check against possible random errors or biases, and also serves to distinguish theories that make many of the same predictions: by checking what the theories predict differently, we can eliminate one theory in favor of another. But most importantly, we should never jump the gun and hail the truth of a theory that has passed merely one test, or a fact that has been established by merely one observation. That would be fundamentally unscientific behavior.

5. Scientific Theories are Explanations of Scientific Facts

Many people in the study claimed that theories could not be tested, or that they were just ideas or hunches. Only a few understood that in science, as one subject put it, “the word theory is used differently than in the general population. It does not mean someone’s idea that cannot be proven. It is a concept that has considerable evidence behind it and has endured the attempts to disprove it.” This is one of the most commonly misunderstood aspects of the nature of science, largely because the word “theory” has a different colloquial meaning. We commonly hear a scientific claim being dismissed because it is “just a theory,” expressing a belief that theories are unprovable or unproven concepts, when in fact for an accepted concept to be given the rubric “theory” in formal scientific literature it must be far from unproven. To the contrary, it must be extensively tested or confirmed and it must continue to survive attempts to disconfirm its predictions in order to win and keep such a title. Otherwise, it is more properly called a hypothesis. Therefore, every currently-accepted theory is well-supported and not “just a theory,” though all theories will vary in how well-supported they are at any given time, and some theories have been refuted and are no longer accepted.

Related to this misunderstanding is a confusion between facts and theories. Facts are what have been carefully observed to be the case. Theories, in contrast, are explanations of those facts. It is possible for a theory to be so well confirmed that it is nearly as certain to be true as the facts it explains, but that would still not make it a fact: theories are not observed, they are only confirmed by what is observed. This suggests a paradox to our common sense, since here we have something that can be empirically tested yet never observed, a counter-intuitive notion that has led to many an ignorant polemic against science. The correct understanding lies in the role the theory plays in explaining and thus predicting facts, for such an explanation must always include one or more unobservable features (the idea of causation, for example) that account for what is observed, since humans cannot observe everything. In the end, a theory’s success in explaining the past and predicting future facts is the basis for believing it is a correct account of things, and that is more than sufficient as a justification. Most people would be astonished to know how frequently they completely rely on theoretical notions their whole lives, constructing explanations for everything they see or even take for granted. Science merely does this more rigorously. For example, the moment you say your car is in your garage even though you aren’t there to see it, that is a theory, not a fact. Of course, it is so certain to be true that we readily treat it as a fact and thus forget the important difference–so one can see here how futile and ridiculous it is to dismiss something by saying it’s “just a theory.”

Because of a theory’s role in generating predictions, scientists employ theories to guide further research, which is another feature of science that goes underappreciated. The amount of new scientific knowledge that the theory of evolution has led us to in just the past century and a half is vast in scope and depth and not to be sneezed at. It has been tremendously successful as a research programme, and has not lost its steam, for it has survived incredible efforts at refutation and become more and more confirmed from more and more angles. In contrast, creation theory offers no guidance for further research and has produced no new scientific discoveries. But this is not to say that only new information, or new technology allowing more accurate observations, will lead to a new theory or the modification of an old one. Theories might change, as one subject said, “due to the reinterpretation of extant data” or “as perspectives and values change.” This is most frequently the case in history, where the same data is looked at in a different way or from a fresh perspective, or connected with other already-known data, and a new theory is proposed that is a superior explanation of the facts at hand. This reiterates the fact that science is as much a rational and logical enterprise as an empirical and investigative one.

6. Scientific Laws are Descriptions of Nature’s Behavior

Many people think that scientific laws are absolute or certain, but this represents the same lack of acquaintance with the tentativeness of science discussed in point #1 above. As one subject claimed, “a scientific law” is “something that has been proven, and therefore will not change because it is true.” This is not what a scientific law is, nor is it even a property of a scientific law. Laws can change, as we discover past errors or new facts. The study’s subjects also betrayed ignorance of the conditional nature of most scientific laws. Laws only apply when certain conditions are present. A common example is the creationist who forgets that the second law of thermodynamics only applies to closed thermal systems. A less common case is the green physics student who thinks the speed of light is a single value, when in fact it varies according to the medium through which light passes. This makes science difficult, because it is complicated, and people expect the world to be simple and easy to grasp.

But more telling, a kind of popular folklore has it that there is, as the researchers’ say, a “hierarchical view of the relationship between scientific theories and laws” such that, as one subject said, “a scientific law is a theory that has been accepted by all scientists and has been proven again and again over time to be true.” This idea that a law is just a theory with a privileged status is frequently taught in high school science courses (I recall being taught it myself), but this is a very naive view that completely misses the function of scientific laws and their relationship to theories. One teacher correctly stated the case: “A scientific law states, identifies or describes relationships among observable phenomena” while “scientific theories are inferred explanations for observable phenomena.” In other words, theories explain, laws describe. These are very different functions. Newton’s laws are often thought to have been overthrown by Einstein’s theory of relativity, but in fact Einstein’s theory explained Newton’s laws. It did lead to some modifications of those laws, but it incorporated them rather than casting them out. For Newton never proposed any theory of gravity, shying away from attempting to know why gravity behaves like it does, and contented himself with merely describing gravity’s behavior in precise mathematical terms, by collecting and analyzing a large amount of relevant data. Laws, unlike theories, are proven directly by the data: the relationships or behaviors found in the data entail some mathematical or logical relationship that is categorized a “law.” Once this is understood, it will be easier to see how scientific laws are not something that had to be decreed (as a common sense reading of “law” would suggest), but are simply patterns of behavior inherent in nature: laws describe how nature behaves, while theories seek to explain that behavior.

7. Science is a Creative Enterprise

Finally, we must not forget how fundamentally imaginative and creative science is. The popular image of the scientist as unimaginative and blindly conservative does not reflect the reality that science only progresses when humans are at their most creative, plumbing their imaginations for ideas and ways to understand the facts and theories of science. This also stands in the face of the common sense notion that science is destructive, being hopelessly skeptical and overly critical, ending up hostile to the feeling of wonder. This mistaken view of science is so common that several authors have written books attacking it and presenting the way science really is: full of wonder and beauty, a source of belief as well as doubt, and fundamentally constructive.[5] Many people might think that the creative aspect of science is limited to the initial stages of research, but in fact it permeates all scientific activity. Brilliant and original ways of collecting and analyzing data are developed, clever theories proposed, experiments run imaginatively. But this does not undermine the value of scientific results: to the contrary, science needs this inventiveness to make progress. It is one of the virtues of science that it is so versatile and can find and make use of so many different ways to explore and study man and his natural world. Science is not shackled to one set of rituals or procedures.

This does mean that scientific explanations, models or theoretical entities are human constructions, created by human beings, drawn from their imagination. But the proof is in the pudding: what makes science so successful is that it marries this tireless resource of human fantasy to a dedication to test everything thoroughly against observed facts and the rules of reason and logic. Thus science gains the best of both worlds: the benefits of human creativity without the shortfalls of superstition or fancy, plus the benefits of a demand for empirical proof without the limitations inherent in the data taken by themselves. Thus, science certainly relies extensively on the believed existence of theoretical entities that have never been directly observed, but does so only when extensive evidence is offered from which those entities can be inferred. This brings us to a close with the one most fundamental and characteristic feature of science: it constructs an understanding of the world through the logic of inference. Though simple deductive logic plays a role in science, the conclusions and discoveries and theories and laws which comprise nearly the whole of scientific knowledge at any given time are all built on statistical inferences (whether overt or implied), employing inductive reasoning to arrive at the most plausible and most probable interpretations of the observations we make not only in the laboratory or in the field but in daily life. In the end, it is a mark of profound achievement that the human species, lacking any means to be certain about anything, has come to know and master so much about the universe.

Concluding Observations

In the sense science is defined here, the term “science” can be expanded to encompass all rational-critical fields, even mathematics, analytical (vs. speculative) philosophy,[6] and history. These last three fields differ from what we normally classify as “science” in certain respects that are not fundamental to the nature of science as such.

  • In general, math and philosophy rely on analysis far more than observation, but neither is independent of observation, and each have their own form of experimentation and require the same ultimate demands of rational-critical peer review. For example, many a mathematical theorem begins with an observed problem or phenomenon, or was hit upon through diagrams or other empirical tools of discovery, and theorems have to be observable and repeatable. And though mathematical “proofs” are unlike scientific theories in that they are genuine proofs, intended to be final and decisive, they may still be in error and thus remain somewhat tentative. Ultimately, in mathematics empiricism is to be found in its heuristics and applications, not in its proofs as such.
  • Philosophy often calls upon observations and experiments that everyone can perform anywhere (self-examination, for instance), and relies heavily on both the methods and conclusions of all the sciences in arriving at an accurate, unified, and comprehensible worldview.
  • Finally, history rarely involves experiments (though it sometimes does), yet like philosophy it still draws heavily on the conclusions of science, and employs hypothetico-deductive methods that are not fundamentally different from other observational sciences (like geology, though geology also has a strong experimental side, as many other observational sciences do). In the end, the historian must gather facts, test theories against them, and construct logical arguments that are subjected to intensive peer review, and thus in effect his observations must be repeatable just as in science, and like scientific theories, historical theories entail predictions about the possibilities of future evidence.

Abd-El-Khalick’s and Lederman’s study also does not address the role of methodological naturalism in scientific method, perhaps because this is not essential to science, only useful, and has more to do with what science has discovered than with the nature of science as such. But something needs to be said about it here. It has been observed by countless authors that science has over the last several centuries discovered a vast body of diverse knowledge which, in hindsight, we now see supports a unified picture of a single natural world,[7] a picture that can thus be used to make scientific research more efficient by directing studies where results are promising, and that confirms the success of science in no small way as a means of getting at the truth. All the sciences contribute to eachother: biology converges on and informs psychology, as physics chemistry, and chemistry biology. This is one proof that these sciences are getting it right. The only failure in this regard to date is the supposed irreconcilability of relativity and quantum mechanical theory, but with the body of support for a coherent natural world we are confident that this conflict is only the result of either or both theories being incomplete or incorrect.

This same phenomenon is relevant on a smaller scale, too: scientific results are built from stronger foundations to less certain theories, so that there is more room for change at the cutting edges of science, but the deeper you get into fundamental matters in any scientific field, the greater and more powerful the span of evidence. Indeed, all the other discoveries in a science in some degree confirm the fundamentals, so that to overthrow the fundamentals would require re-interpreting a vast mass of ancillary research. This cumulative nature of science is another factor supporting its truth, and in conjunction with the proven inter-connectedness of all science, it requires positing too great a coincidence to suggest that the fundamental findings of science could be false. But this leads us to discussion and exploration of what science has found out, rather than its nature alone.[8]

I would like to thank Keith Douglas and others for their many critical and helpful remarks in my preparation of this paper for the Secular Web.

[1] This is not to say that science does not employ, and perfect, natural patterns of thought shared by all humans, only that the way people often think about things is contrary to the way science teaches them to, at least until they are taught to think scientifically. See especially Alan Cromer’s excellent book Uncommon Sense: The Heretical Nature of Science (1993). Several books have discussed the natural human tendency toward irrational or unscientific approaches to belief formation, in particular: Stewart Guthrie, Faces in the Clouds (1993); Stuart Vyse, Believing in Magic (1997); Michael Shermer, How We Believe (1999); and Donald Calne Within Reason: Rationality and Human Behavior (1999). The above books naturally touch on the biology of belief formation, which has become a hot research subject of late, now more thoroughly discussed in the following recent works: Joseph Giovannoli, The Biology of Belief (2001); Andrew Newberg, Eugene D’Aquili, and Vince Rause, Why God Won’t Go Away: Brain Science and the Biology of Belief (2001); Eugene D’Aquili and Andrew Newberg, The Mystical Mind: Probing the Biology of Religious Experience (1999). On the nature of doubt and skepticism as studied in psychology, see Bibliography.

[2] Fouad Abd-El-Khalick and Norman G. Lederman, “The Influence of History of Science Courses on Students’ Views of Nature of Science,” Journal of Research in Science Teaching 37:10, pp. 1057-95 (2000).

[3] A note of caution: I don’t think Abd-El-Khalick and Lederman mean subjective here in the sense of non-objective, but in the sense of human-constructed. We don’t see theories, we invent them, and then infer their correctness. Though they can (and when well-supported, probably do) represent objective truths, they are only subjectively perceived, e.g. we do not see electrons orbiting a nucleus, we imagine this in our heads, and then deduce consequences that we then look for in observed facts. More on this follows in the text.

[4] See my A Fish Did Not Write This Essay.

[5] Prominent examples: Carl Sagan’s Demon Haunted World, Richard Dawkins’ Unweaving the Rainbow, and Edward O. Wilson’s Consilience.

[6] I mean by this the distinction between analytical philosophy, which includes such endeavors as analytical ethics and metaphysics (this latter being the analysis of the conclusions of the other sciences for the purpose of constructing a sensible unified worldview), and pure metaphysical speculations of the sort made by Plato or Descartes, or mystical or theological philosophy, which rely in any way on dogmatic assertions or non-intersubjective personal experience.

[7] For more on these issues, see our many essays on Naturalism, as well as my Prima Facie Presumptions vs. The Lessons of History.

[8] For the whole scoop on the philosophy of science, the best recent survey of the subject is provided in two volumes by Mario Bunge, Philosophy of Science: From Problem to Theory and Philosophy of Science: From Explanation to Justification, 1998. Also recommended is Robert Klee’s Introduction to the Philosophy of Science: Cutting Nature at Its Seams, 1996.