Interesting discussion of the philosophy Of science. On a slightly different tack - to attempt to answer the original question, One reason science students don't get taught scientific method earlier is that most people, including science teachers at high school, do not really understand what it means to 'do' science, except in simplistic superficial terms. Discussions of the philosophy of science support this view, as even those who have given it serious thought disagree.

Because despite calls for reform (at least in the US) for the past few decades and quite thorough treatments of the problems and possible solutions in publications like the National Research Council's National Science Education Standards, the 19th century epistemology still seems to have a death grip on educational practices. First, a large number of science teachers have no background in research. Second, it is very easy to introduce The Scientific Method as it existed a century ago through experiments done before that, but very difficult to explain pre-college students debates about underdetermination of theory by data and in general the nuanced ways in which hypotheses emerge from the theories they "test" not to mention have students go out and review the current state of research before planning a pilot study that will lead to their actual experiments. The necessary middle ground is not unexplored, but no particular approach has been particularly successful. Third, because most of pre-college education (at least in the later years) is no geared towards getting students into universities, they are taught the language of the sciences (mathematics) through rote application of memorized rules and are taught little to nothing of the foundations of the sciences (analytic reasoning, logic, argumentation, etc.). Fourth, it took several centuries for scientists to reach the point at which two simple results of simple experiments yielded incompatible results (the work on light by Young and Einstein), thus showing that perhaps the oldest and most venerated of the sciences had failed to realize how theory-laden hypotheses are and so had advanced a view of reality for some 200+ years that garnered ever more evidence for an ever more nuanced and developed account of reality only to find the foundations ripped out from underneath. While reeling from the aftermath of a new physics that referred to mathematical functions as physical systems, Lorentz was kind enough to show how even simple systems could be exhibit incredibly complex behavior, adding epistemological indeterminism to the ontological indeterminacy of quantum physics.

In short, science is still largely taught as it was practiced in the 19th century, science teachers often have little or no research experience, and nobody quite knows how to communicate even to undergraduates the nature of what has become such an interdisciplinary and vast endeavor that a singular term "science" seems inadequate.

Of course, if we focused more on communicating the language and foundations of the sciences (i.e., conceptual mathematics/formal languages, analytic reasoning, logic, etc.), a student with no background in chemistry, physics, or biology could outperform most students with such backgrounds entering universities today. Most of what is learned, from systems of equations to what "theory" or "law" means has to be relearned anyway, and as few students actually know what they intend to do the foundations for being able to do any of the sciences or known have great utility. Years spent memorizing outdated Newtonian mechanics and "calculus" rather than linear algebra or real calculus (i.e., calculus as the beginning of analysis) don't help anybody, but we still require all too many college students to waste countless hours not learning things they won't use or teaching them how to use tools without the background to understand them (multivariate statistics courses are too often a "here's how to use SPSS/SAS/etc. to run this test" because the underlying linear algebra is somehow too much to expect even graduate students in many sciences to learn, apparently).

Perhaps, like experiment about grammar, when it was supposed that it would be mastered by students when they would read and write. May be it is supposed that they would learn scientific method when they would go through others researches. It is why research students are more dependent on their supervisor than on themselves.

Dear Foreman, From my own experience teaching philosophy of science to American undergraduates, I would say that a chief difficulty is closely connected with the fact that the students I have encountered--typical undergraduates-- often do not know enough science. (You first have to explain various scientific theories and results.) It stands to reason that before you can reflect much on science and its methods, you first have to know sufficient science to be in a position to understand its development. This point seems a key to understanding the close relationship which has developed between philosophy of science and history of science.

Recall that even Newton famously said that he "stood on the shoulders of giants." This is to suggest that he couldn't have developed his own physics or his methods, if he had not first understood the scientific work of those who went before him. More generally, if methodology (and even epistemology) arise from reflection on the better and worse of first-order subject-matters in the sciences--successful theory and experimental practice--, then it should be expected that people won't make much headway on methods of the sciences without some overview of the development of the sciences --or of some particular science, as the case may be.

More could be said here, but perhaps this is enough for now --to suggest that delay on instruction on scientific methods makes a considerable amount of good sense. If we suppose on the contrary, that epistemology and methodology of science were somehow known a priori, then the opposite course would perhaps recommend itself.

H.G. Callaway

Very true. Even those social scientists who deal with the philosophy of science are primarily interested in the critieria for "scientific status," rather than research practice int he advanced natural scientist. When they speak of "Scientific Method," that is the methodology taught in social science methodology courses, they are referring to a particular view of science, philosophy of science in the tradition of Logical Positivism. This view has almost nothing in common with the research practice of the advanced natural sciences. Here is an interesting comment by Popper on this state of affairs:

http://www.the-rathouse.com/Pop-Schol/Popper-scientific-method.html

It was interesting when Johannes commented that "Maybe inductive reasoning and falsification does not match very well with the current systems of academic science and publishing?".

I would hold the view that inductive reasoning and falsification are two extremes, I think that the latter is a far better way to do science as it is self correcting but I know that a lot of people are using a mixture of these two methods in their work.

Much undergrad and school education seems only to ever be an attempt to communicate a series of facts which are learnt by rote.

There are exceptions, my phyics teacher told me and the rest of the class that we needed to know less than 10 equations, but we needed to be able to recreate all the other equations in the book from these key equations.

Dear Foreman, It seems now that your interest is a bit different from that of your initial question--more a matter of Why aren't advanced students in the sciences methodologically sophisticated?, rather than, Why is methodology taught so late? These strike me as quite different questions. As for teaching facts, in science, this would seem to be unavoidable, if students are to make sense of the ways in which theory can encompass or conflict with those facts. So teaching of facts in the sciences must be accompanies by the teaching of theory in relation to relevant facts. It is by reference to theory that relevance is more fully brought out. Too, often, however, students just want to get by the exams. I have certainly heard this complaint about lack of methodological/scientific sophistication in advanced students. Here I sympathize.

I have a little story of an encounter with an advanced student (of philosophy) in a European university setting. I showed the fellow a then recent book which I took to be highly relevant to his specific interests. (I was by far the more senior person in the encounter.) What the fellow did was to simply turn to the bibliography in the book, look through the listing of authors, and on that basis decided that the book didn't interest him. (It was typical that the book didn't have an index.) This frankly amazed me, since I was so accustomed to judging of the interest and relevancy of material on the basis of its subject-matter, instead of the author's referenced.

But on reflection this made sense to me, since I saw a pattern in which the students simply studied the perspectives and references of their professors--who were going to be present at the oral exams in any case. This is, or can be, a version of rote learning. The typical result was a very deep knowledge of a fairly narrow slice of the field, and little interest in others writing and publishing on the same or closely related topics. (Very little of "compare and contrast.")

I also recall situations in which different students in the same department (this was at another European university), each following a different professor in the same department, ended up not understanding each other, and unable to discuss their differences, because their professors where in a long-standing conflict of some sort or other--obscure to all except those directly involved. The students seemed to simply recapitulate the standing conflict, and yet this counted as education. If your professor approves your,o.k., otherwise not.

Now, ask yourself why students would concern themselves with complicated questions of methodology in such circumstances. If the students already know the cast of their final exams, then they will likely listen, not at all, to any doubts or alternatives. The impulse of the professors to "reproduce" themselves via their students, and thus show their influence, is in standing tension or conflict with gaining a more encompassing view of the subject.

I should say that I also encountered very considerable intellectual independence among students, who rarely asked for or needed any help within the confines of courses I taught --logic, say. But in a course with unusual or controversial content, I'd see the students of some sitting professor or other, who would come in as self-appointed critics and defenders of the established line. (This seemed crucial to the academic politics of their professors.) I can imagine that your professor wanted you to think through those 10 equations yourself --in relation to the rest. That's a good exercise, though challenging. Beyond that challenge would be to encompassing the divergence between differing text books and approaches. It is by challenging ourselves with differing and divergent approaches and views, theories and perspectives that our methodological scruples are best sharpened.

"Compare and contrast." It may be, however, that this is not often the most direct career route.

H.G. Callaway

Dear Callaway:

Regarding the statement "I would say that a chief difficulty is closely connected with the fact that the students I have encountered--typical undergraduates-- often do not know enough science. (You first have to explain various scientific theories and results.)" This is no doubt true in a certain sense (how can one explain methods used in biophysics to students lacking knowledge of both biology and physics?). However, "science" is not something one knows, but is simultaneously an endeavor, and epistemological framework, and a worldview altogether (but not exactly any of these either). It is not something one knows, learns, or is taught, and as Johannes Shied pointed out, teaching "science" as consisting of facts simply reinforces the notion that science is about knowledge rather than a very particular approach to how knowledge is acquired. Students who take English, history, etc., are learning things about these subjects because these subjects are indeed that: subjects. English is a language. Literature, however diverse, ill-defined (or defying definition), illusive, indescribable, etc., is still something one learns about (and some few produce). Science as it is taught is like this- a subject consisting mostly of facts one learns that concern the dynamics of (mostly physical) phenomena. However, science for scientists is far more something one DOES than something one learns.

But we don't teach it as such. Not only that, we frequently fail to provide a framework for enabling students to transition from learning what previous scientists have discovered (as an aside, despite Newton's famous quote he was an arrogant and spiteful individual who hid his accomplishments from his peers only to find to his extreme displeasure that Leibniz had independently developed his jealously guarded calculus; he was not a humble individual and that particular quote is about the most misleading available for any understanding of the individual). When students study literature they are given writing assignments. When they learn about history they are given research projects. When the learn about science they are told to recreate work already done and carried out according to a method a century outdated. This information is imparted without the fundamental tools underlying the sciences or knowledge that these exist. Instead, students are told of a mythical entity called The Scientific Method. To the extent such a thing exists, it does so in the application of analytical reasoning to empirical findings garnered through particular methods that vary widely according to one's field. Students are given nothing from which they can build off of to understand this. Often, they are systematically misled by an education which reinforces the idea the science consists of coming up with some question independent of theory/framework, testing it with the confidence expressed so thoroughly and so misguidedly by Laplace and Kelvin, and then (if confirmed) magically we get a theory. The analytical skills they might have gained in previous decades from everything from calculus done right through proofs in geometry have steadily but consistently been eroded, abandoned, or replaced. So, after years of schooling, would-be scientists learn that science is like history or declensions in Greek and Latin: memorized facts.

Of course facts are, as you say, unavoidable. But equating such teachings with teaching "science" is absolutely avoidable. When I learned Greek, Latin, and graph theory, I learned these as the subjects they were; that is, I learned the foundations for these topics/fields. Newtonian gravity, Maxwell's equations, etc., are not the foundations of science. We teach the "facts" of science to avoid reinventing the wheel, but in the end the foundations of science are now, have always been, and will (I think) always be the use of reasoning/logical analysis applied to empirical findings and these do not require knowing anything about a particular field. Someone who knows nothing of work from Galileo to Hamilton can still show that Aristotelian mechanics was flawed and it would still be science. It would do nothing to advance mechanics, but it would remain science nonetheless. So while the facts are important, it is equally important to impart WHY they are of import, and why they are not to be equated with science but with what science has accomplished.

Following up Andrew's answers, I'll dare say: it is a good thing and, moreover desirable, that science students get little teachings about scientific method.

History and philosophy of science show no examples of real scientists who have made of the scientific method a matter of concern for their science. (With the notable exception of Poincaré [19th Century minded?]). To be sure, certainly not as the 20th century till today runs towards us.

In my university we have come to terms to that. Instead of teaching at any level (from undergraduate to Ph.D. courses and seminars) courses on scientific methods/methodology, it makes much more sense to teach a good course on epistemology, history of science, philosophy of science - well, scientific discussions as they run and happen.

I think that the questions of “why are students not methodologically sophisticated?” and “Why is methodology taught so late?” are very closely linked questions. While they are slightly different.

The second relates to the behavior of the university teachers while the first relates to the state of the students.

If the university / university teachers do not teach students about the reasoning behind science then the students are less likely to consider the subject and thus become more “methodologically sophisticated”.

Pitty the book did not interest the student, to me it seems like an example of “argument from authority”, which is a poor way of deciding if something is good / bad or right / wrong. The name of the author and those in the references should be less important than their ideas and arguments.

You are absolutely spot on for asking this question. I also notice that researchers in the hard sciences are not trained in such domains as the history of science, philosophy of science...etc. which seem to be taken for granted. Whereas in the social sciences, there is a great emphasis on the topic of science in general. For example, in management research training, most universities would have electives on philosophy of science and epistemology. I consider the following references as "must read" for any student undertaking a PhD research project:

What is this thing called science, by Chalmers

Conjectures and Refutations, by Karl Popper

The Structure of the Scientific Revolutions, by Thomas Kuhn

Dear Foreman, I think the two questions are significantly different, if one goes about trying to answer them. If I focus on your first question, then I likely think of things like experimental design and the provision of effective controls in experiments. This is a kind of sophistication, much needed, but strongly related to experimental practice--and unlikely to be effectively taught in isolation from the specifics of experimentation. . From a more theoretical perspective, questions about the consideration and comparison of alternative theoretical approaches become important.

If we ask, "Why is methodology taught so late?" then this is strongly related to the status of methodology. Looking at this from the philosophy of science, I think that you have to consider that philosophy of science depends on the sciences. We can ask all sorts of interesting questions about science, say, "what is explanation?" or "what is confirmation?" or what is the role of crucial experiments?" or "what is a good experiment?"etc. But trying to answer such questions without drawing on the history of science would be pretty empty. The philosophical discussion needs to take up the scientific paradigms, the range of differences, and consider them in some detail. No one will be able to effectively discuss methodology and crucial scientific concepts in general ignorance of scientific developments over decades and centuries. This is to emphasize, again the relationship of philosophy of science to the history of the sciences.

Unfortunately, regarding the book mentioned, philosophers tend to live out their careers in distinct academic tribes of sorts, and communication among them can be quite difficult--like communication across cultural differences generally. There is little to do about this, but it should not be allowed to blur our understanding of the scientific situation. More general study of philosophy of science can help, as Benmeriki suggests. On the other hand, the various sciences teach their own methods to their students by means of their distinctive practices --often after the content of established or accepted theory.

H.G. Callaway

How many instructors, faculty, or professors are qualified to teach the philosophy of science? For that matter, how many could teach a broad based class on research? Many researchers that I have encountered could not define, or even describe what constitutes knowledge. I find this somewhat ironic. In virtually any field, one would be required to be able to judge between a product of high quality and one of low quality; between an acceptable product and an unacceptable one. Many doctoral graduates will on the other hand not be able to do the same for what is ultimately the product of of research; namely knowledge.

I teach a course on research design that includes subjects such as the theory of knowledge, the philosophy of science, epistemology and ontology. We debate questions such as whether knowledge can exist without "observation" of a phenomenon, or if theoretical physics is or is not science. Although the course is extremely well received by students, many faculty do not understand its purpose or value.

Could it be that the philosophy of science is not taught because many of us that are in positions that would make us responsible to include such topics in curricula may not have "encountered" them, and are not in a position to recognize that they are missing?

I would agree that the questions "'why are students not methodologically sophisticated?” and “Why is methodology taught so late?'" are more distinct than we might suppose. In fact, I would say that the reason these are distinct are related to why books like those mentioned already (e.g., by Chalmers or Popper) are very rarely used in research methods courses even as secondary texts. While "late" is a relative term, as far as scientific research is concerned I consider undergraduate level courses to be early. It is only at this point that the student has access to anything remotely resembling what practicing scientists do (e.g., university libraries, online databases, interlibrary loans, and similar tools required just to BEGIN some "research" project). I recently came across a paper titled (I believe) "The neural basis of unconditional love". I will never, I think, read it because 1) it's such a great title & 2) I know how horrific neuroscience research is when it comes to explaining higher level cognitive, evolutionary, and similar functional processes s.t. there's no way the paper could say anything about something like love without lacking any sound foundations. However, were I interested in the neurobiology of love, I would need to look at papers like that and many others ("Love hurts: An fMRI study"; "The neural correlates of maternal and romantic love"; "Love is more than just a kiss- a neurobiological perspective on love and affection"; etc.). Before college, this simply isn't feasible.

However, undergraduate research classes do not teach either the philosophy of science or much more than an algorithmic/procedural approach (i.e., critical thinking-lite or critical thinking-free approaches) to research. In a reply to a critical review of their book The Cult of Statistical Significance, the authors thank the reviewers for being practically alone because, despite the ubiquity of null hypothesis testing the authors' colleagues had been either silent or in agreement. The authors note that "[o]ne eminent econometrican told us with a smirk that he agreed with us, of course...But he remained unwilling to teach the McCloskey-Ziliak point to his students in a leading graduate program because 'they are too stupid to understand it.'"1

I can but trust the authors here, so assuming this is true we have here a teacher whose graduate-level research methods students are (in his words) "too stupid" to understand sound experimental paradigms (or perhaps "too stupid" to be taught the problems with the most frequently used basis for experimental paradigms across social, cognitive, psychological, and other sciences). I disagree with this assessment, but the point is not even whether null-hypothesis significance testing is sound or not.2 The point is that research methods are taught, even at the graduate level, with too little regard to what science is, what its foundations are, what makes research good or bad, and so forth. Like the language of science (mathematics) the foundations are frequently taught as devoid of any conceptual content as is possible. I had to teach and tutor undergraduate and graduate students how to use software like SPSS or SAS to perform statistical analyses the logic of which the students were incapable of appreciating (not because of any intellectual difficulties, simply because the courses they took did not include linear algebra or other sufficiently comprehensive treatments of the mathematics underlying e.g., the GLM). Sophisticated computational methods have become a staple in research across the sciences, but the necessary educational pre-requisites for understanding these are lacking. Textbooks like Combinatorics of Experimental Design or monographs like Probability Theory: The Logic of Science are either intended for post-docs, read almost only by post-docs, or not read at all (or included in any courses).

Part of this is due to the need to impart the basics. One can hardly delve into the nuances of experimental use of inferential statistics without having some knowledge of experimental designs that are intended to use these. A topic as specific as what an "observation" consists of can fill volumes (and has), yet one cannot expect a ~20 year old student to learn anything about empirical methods without first having a vastly simpler understanding of what measurement, observation, etc., mean/consist of. Nor are many of the issues philosophers of science deal with of great import to practicing researchers. Modern physics treats physical systems at the most fundamental level using a statistical mechanics and thus consisting of entities without any known ontological status, yet much of the modern world has advanced through (and indeed depends upon) this treatment. So despite the vast literature on what, exactly, the wavefunction "means" or what the "state" of quantum systems are (ontologically), physicists have done just find producing countless results, developing undreamt of technologies, and so on, regardless.

Yet to strip away so much of scientific practice as to make it appear to be a unified body of knowledge all scientists use rather than a set of practices with an underlying logical/analytical framework (and, to the extent the latter is at some point imparted, to do so in ways that minimize this aspect), is to systematically mischaracterize the nature of science. Students continue to think of theories and laws in science as being independent of hypotheses and experiments despite the fact that perhaps the most well-known scientific "law" (the "law of gravity") is not simply not a physical "law" in any sense, but is rather universally recognized as being wrong. It is approximately right in application, but it is nonetheless wrong. While thousands of experiments are carried out every year, most people (whether they majored in some science or not) could probably name but a dozen or so theories if that: the theory of evolution, the theory/law of gravity, the big bang theory, etc. The fact that thousands of experiments have served to create, extend, and modify, and that entire fields (e.g., evolutionary psychology) rest upon, the framework that is evolutionary theory goes unnoticed by those who are taught that one develops a hypothesis, tests it, and then it becomes a theory (perhaps, per Popper, to be one day falsified). The truth, that research questions are borne of the theories that they "test" (and the results to at least some extent interpreted in light of that theory) is not simply skipped over for simplicity. That would imply a need to skip over it, when in reality the only reason this conception of science exists at all is because it is taught. We need only cease to teach it.

1 McCloskey, D. N., & Ziliak, S. T. (2008). Signifying nothing: reply to Hoover and Siegler. Journal of Economic Methodology, 15(1), 39-55. (http://www.deirdremccloskey.com/articles/stats/sig.php)

2 “In a recent article, Armstrong (2007) points out that, contrary to popular belief, “there is no empirical evidence supporting the use of statistical significance tests. Despite repeated calls for evidence, no one has shown that the applications of tests of statistical significance improve decision making or advance scientific knowledge” (p. 335). He is by no means alone in arguing this. Many prominent researchers have now for decades protested NHST, arguing that it often results in the publication of peer-reviewed and journal endorsed pseudo-science. Indeed, this history of criticism now extends back more than 90 years (e.g., Armstrong, 2007; Bakan, 1966, 1974; Berkson, 1938; Boring, 1919; Campbell, 1982; Carver, 1978, 1993; Cohen, 1990, 1994; Edwards, 1965; Falk, 1998; Fidler, Thomason, Cumming, Finch, & Leeman, 2004; Fisher, 1955, 1956; Gigerenzer, 1987, 1993, 2004; Gigerenzer et al., 1989; Gill, 1999; Granaas, 2002; Greenwald, 1975; Hubbard & Armstrong, 2006; Hubbard & Lindsay, 2008; Hubbard & Ryan, 2000; Jones & Tukey, 2000; Kirk, 1996, 2003; Lindsay, 1995; Lykken, 1968, 1991; Meehl, 1967, 1978, 1990; Nunnally, 1960; Rosnow & Rosenthal, 1989; Rozeboom, 1960; Schmidt, 1992, 1996; Schmidt & Hunter, 1997; Sedlmeier & Gigerenzer, 1989; Skinner, 1972; Thompson, 1996, 1999, 2002, 2006, 2007; Tukey, 1991)… In his outgoing comments as editor of the Journal of Applied Psychology, Campbell (1982) wrote:

It is almost impossible to drag authors away from their p values, and the more zeros after the decimal point, the harder people cling to them. It is almost as if all the statistics courses in the world stopped after introducing Type I error. …Perhaps p values are like mosquitos. They have an evolutionary niche somewhere and no amount of scratching, swatting, or spraying will dislodge them. Whereas it may be necessary to discount a sampling error explanation for results of a study, investigators must learn to argue for the significance of their results without reference to inferential statistics. (p. 698)

Nineteen years later, Finch, Cumming, and Thomason (2001) noted that little had changed in almost two decades...Thus, to make the same points now that have been made repeatedly for decades is not, by any means, to beat a dead horse. A horse that is winning the race is not dead. The fact that this horse is still in the lead unfortunately suggests that the long history of authors making such observations have largely been ranting to the wind….It is certainly a “significant” problem for the social sciences that significance tests do not actually tell researchers what the overwhelming majority of them think they do (Bakan, 1966). Bakan thought that “everybody knows this” and that to say it out loud is to be like the child who pointed out that the emperor is wearing no clothes. He argues that if we pull out the strand of NHST much of the tapestry of psychology would fall apart. Indeed, NHST, Gerrig and Zimbardo (2002) state, is the “backbone of psychological research” (p. 46). So, instead of abandoning it, which could be very embarrassing, we make the adjustment of simply misinterpreting what it actually tells us, which is…not much.”

Lambdin, C. (2012). Significance tests as sorcery: Science is empirical—significance tests are not. Theory & Psychology, 22(1), 67-90. (http://psychology.okstate.edu/faculty/jgrice/psyc5314/SignificanceSorceryLambdin2012.pdf)

That´s why I studied Philosophy and History of Science before starting doing anysome science. Now I´m in the Neurosciences and keep on studing Psychology. It helps me very to reflect all the methods and different perspectives. I´m always staggered about how blind and unreflective my fellow student keep practicing the scientific method they learned and they are not even open to critics because the´re afraid of belonging not to the mainstream and beeing a maverick at the end! They don´t even see the enrichment of puttting oneself into other shoes and switching between views.

I thin: Every student of science, indifferent wich science, has to fullfill a course in "Philosophy of Science" in "Epistemology" and in "History of Science" before starting to work and think!!!

Speaking from the perspective of a philosopher who has taught philosophy of science in higher secondary education (i.e. sixth form, gymnasium, or lycee), on under and postgraduate level, and to PhD students from a range of disciplines (IT, Health and social science), as well as research methods in the health and social sciences, then I can only say I agree completely that it is a complete puzzle why students in the sciences do not get more training in critical scientific thinking (which should be a combination of "this is how we do it" and "this is why we do it like that"). In most cases students are just fostered into the paradigmatic ways that researchers in their particular discipline conduct their research, with no particlar heed given to anything else than just various measurement and interpretation biases of the particular experimental set-up.

I think the reason for this is that the natural sciences seem to think that their research is philosophically unproblematic (they are, after all, the "exact" sciences) and of course often it is. But it is often overlooked that the exactness and certainty is confined to research that works within given parameters, using tested and validated methods. Such research (what Kuhn called "normal science") includes probably 90% (or more) of all research in all disciplines. The certainty does not hold for research that really is pushing the boundaries of our knowledge.

The idea that natural science is unproblematic combined with the fact that most researchers only do relatively unproblematic research, gives a false sense of certainty about research in the natural sciences generally. The social and health sciences on the other hand have to fight to justify the validity of their research, wherefore they think it is important to have a good understanding of scientific thinking generally. The result is that in interdisciplinary research, students with a background in natural science often find it more difficult to adapt to an interdisciplinary milieu.

I would like to push the question even further, notably why the philosophy of science isn't a compulsory subject already in secondary school, since science now has such a central role in human society?

Dear Ingthorsson, In the sense that philosophy of science draws upon results and practices of the sciences and goes on to characterize them or attempt generalization, science must, plausibly, be treated as unproblematic --in general. This is not to say that particular developments or proposals can't be treated as problematic: maybe "cold fusion." Again, philosophy of science may have a strong interest in failed scientific theories--in order to understand how they fail. Insofar as philosophy of science may correctly generalize about the sciences, then it may be of interest to the sciences, too, as a kind of feed back loop, but frequently it goes off in its own directions, and scientists may sometimes justly feel that there is too much speculation afoot.

I suspect that the idea of "normal science" has sometimes been over-emphasized, and though there may be such a thing, it often sounds to me like a representation of the experimental side of things, put in contrast with the development of theory. I'm not convinced that there is any great danger in the various sciences teaching their methods to their students together with their usual practices, theories and results.

It is sometimes said that a chief problem of the social sciences is that if you take a byte of data in the social sciences, you may get bitten back! We recognize important restrictions on experimentation with human subjects, too. The human relations and moral issues involved are more direct and more complex than what is usually encountered in the natural sciences. So, it may be that the problem you see in the students of natural science adapting to the interdisciplinary has this root in differing experience and practice. But I believe that there will be somewhat similar problems in moving from one specialized natural science to another--in the sense that the experimental problems and conditions are likely to be so distinctive that not much useful carries over --just analogies, or something like J.S. Mill's "methods," of induction. BTW: Mill used to be taught quite regularly toward the end of introductory logic courses in the universities here, and likely this is still quite prevalent.

I have no view of teaching philosophy of science in the secondary schools.

H.G. Callaway

Dear Callaway. If I understand your view correctly, you think students should get training in scientific methodology/scientific thinking late in their education. Your reason for thinking this is that you found that the students you taught philosophy of science didn't know enough science to really understand the philosophical arguments. Could it be that you were teaching philosophy students whose background educations was primarily in the social sciences and humanities, and that you were teaching them general philosophy of science using as literature books like "What is this thing called science?" or perhaps Hempel's classic? These are all good works, but they primarily use the natural sciences to illustrate the philosophy, wherefore it is not surprising that students who haven't studied physics or chemistry to any great extent struggle with the examples. This is, unfortunately, how philosophy of science is typically taught in philosophy departments around the world. That was how I was taught it, and it is how I taught it initially. This changed when I ended up teaching research methods in food and nutrition, social work, physiotherapy, nursing and occupational therapy, and had to learn how to weave together the particular methodology, say, of dietetic research, and the philosophical ideas about the strengths and weaknesses of that methodology. When you do that, the problem of students lacking understanding of the facts of the discipline disappears, they suddenly realise the relevance of it, and find that their understanding of the subject moves to a whole new level. The same is the cease if you teach philosophy of science in the sixth form to a class of all natural science students. They have no problem of understanding everything Popper, Kuhn, and Chalmers say, except perhaps that they detect (like Messing points out) that their views of 'scientific method' is a little old-fashioned and outdated. But they easily adapt that to their more advanced knowledge, say, of physics and chemistry. In essence, I think the problem of teaching research methodology/Scientific thinking to young students is a teacher oriented problem, not a student oriented problem. It is the failure of the teacher to address the philosophical problem from the perspective of the science (or academic discipline) that the students have had their training in. This failure is of course not the teachers fault. It is the fault of the educational system, which trains philosophers of science mainly within the paradigm of natural science. Of course, 'science' in English is used mainly for natural science, but in German and all the Scandinavian countries, the equivalent word (wissenschaft, vetenskap, vísindi) includes all academic disciplines involved in research of some form. It is the latter sense which is more relevant for the philosophical subject called 'philosophy of science', because it isn't confined to the natural sciences (although it is often taught as if it was).

Dear Ingthorsson, It is not that I'm advocating teaching methods later in the education of science students, it is more that I am inclined to explain the later teaching by the student's need to master the science or sciences involved.

I think there should be a place for teaching general philosophy of science, including its focus on natural science, and such courses need not be adapted in detail to the particular scientific aims or interests of students. There would be some danger, in that direction, of an excessively narrow approach. Again, I think that the details of methods are reasonably taught in the various scientific disciplines themselves.

When I recently taught philosophy of science, the students were generally lower level undergraduates, some with thoughts or aims for science-related careers, but who mostly had little developed, or specialized training. There are differences between American and European universities along related lines, in my experience, and American undergraduates tend to be less specialized in their early course of studies. I was not teaching classic twentieth-century material, or only partly, and that was combined with historical materials. This included a goodly emphasis on early modern science and its contrast with ancient treatments of related topics. (The textbooks were suggested by the department chair, a specialist in philosophy of science. I think he wanted to see how the materials would go over.) I believe that these students, given their range of career aims would likely have been less interested in philosophy of social sciences & etc.

H.G. Callaway

The previous discussants undoubtedly raised a number of interesting points. However, I cannot resist the temptation to observe that it is an undisputable fact of experience that people are usually quite uninterested in meta-analyses, even when they themselves are more or less directly involved. Consider, for instance, the average people who typically are not very well versed in psychology and have only an amateurish knowledge of the studies on sexuality, or religious people who show the same lack of concern towards the philosophy, history or sociology of religion. Scientists’ indifference to any kind of “science theory” is, in many respects, no exception to the rule.

Dear Salanti, I think you are right about more general lack of interest in meta-analysis of various sorts, though philosophers, or particular directions of philosophy, constitute an important exception. Still, on the other hand, it is important to be able to get down to specifics, in any case where meta-analysis may prove itself useful. Otherwise, the meta-analysis may create generalized expectations without fruitful application and/or without valid testing/evaluation. The problem for students of philosophy of science who know little of the sciences, is not however, merely that they may have little interest. It is more that they lack the background knowledge needed to make much sense of plausible or standardized generalizations.

What comes to mind is the use of the idea that observation is "theory-laden." I have no doubt that there is a sense in which this is true--observation is often and usefully theory-laden. Yet this idea has often been interpreted to mean that scientists can only see what they expect, and that the advocates or defenders of particular theories will be unable to observe counter-evidence. On the contrary, it seems clear that the Newtonians could also observe and measure the bending of starlight in the vicinity of an eclipse of the sun, as predicted by Einstein --the observational displacement in relation to others stars further from the eclipse. This is a firm point for anyone who knows a bit of the history of physics. But for someone introduced to the science, merely to explain the meta-analytical claim, the entire business may look ad hoc. Prior education in physics may be expected to render the point of confirmation firm--and thus facilitate a more adequate understanding of the claim that observation is theory-laden.

One reason for a general lack of interest in meta-analysis is that it is very difficult to evaluate for anyone not familiar with the relevant object-level results. In consequence, meta-analysis lacking an appropriate context of knowledge plays too easily into bandwagon effects and doubtful speculation. At this point the scientist will likely think that it all leads off into useless speculation and doctrinaire or ideological dogma --not without reason, as I see it.

H.G. Callaway

Coming from a rather mixed background, I learned Philosophy of Science in the context of a joint honours degree in Philosophy and Psychology - so I was learning experimental methods and statistics in my Psychology practicals, at the same time as learning Philosophy of Science as a course within Philosophy - and at that time I found it very frustrating that the philosophers seemed to lean so heavily on made-up or trivial examples. I became a convinced Popperian partly because Popper seemed to be the only one whose thinking related to actual science. Since that time, however, Philosophy of Science has engaged much more closely with scientific practice - and I think perhaps we have the opposite problem - if you read Hacking, for example, the philosophy seems to have disappeared into the History of Science.

I worked in computing, both in industry and academically, and for several years I taught a Research Methods course for MSc students in computing and computer networking. Here I encountered the problem that has been alluded to above, of the stage at which students know enough about actual research to be able to appreciate philosophical issues. Whereas as a Psychology undergraduate (for half of my time) I had been expected to read primary research literature from the very outset, Computing students had no such previous exposure to real research. I remember discussing this with a colleague in Physics, and he said there was no way that undergraduates could be expected to understand Physics research papers at all. This then makes it hard for me to understand what their education consists in!

It was certainly difficult for my computing/networking MSc students to produce a Dissertation - modelled on a scientific paper - without the sort of experience of both reading published scientific papers and regularly writing up lab reports in the form of a scientific paper, that had constituted a central part of my undergraduate education in Psychology. (By the way, the "Significance Test Controversy" was already a matter of live debate in Psychology when I was an undergraduate, and we certainly would not have mistaken the significance of an experimental result for the size of an effect.)

I've now gone back to Philosophy as a research student, and am trying to get to grips with the relationship between Philosophy of Science and what I would call "the disciplines of computing" (as opposed to "Computer Science" which seems to be rather a question-begging appellation!). The structure of research papers in much of the field departs very markedly from that of the classical experimental paper, and one thing I'm trying to understand is whether this reflects different types of knowledge claim and different forms of argumentation, or is it just a stylistic preference or even a lack of basic education in scientific writing?

Ingthorsson wrote: "Of course, 'science' in English is used mainly for natural science, but in German and all the Scandinavian countries, the equivalent word (wissenschaft, vetenskap, vísindi) includes all academic disciplines involved in research of some form. It is the latter sense which is more relevant for the philosophical subject called 'philosophy of science', because it isn't confined to the natural sciences (although it is often taught as if it was). " This raises two queries in my mind:-

(1) I have the impression that in English "Science" (unqualified) is usually taken to mean "Natural Science", while "Social Science" has two different connotations - some think it means applying the methods of natural science to society and social problems, while others think it is a completely different field, and that "scientific method" is therefore inapplicable. I also have the impression that "wissenschaft" in German is wider than both of these.

(2) Popper put forward a criterion for discriminating between science and metaphysics - but what word did he use when presenting this criterion in German?

Julian, I too have wondered about the structure of philosophical papers, which, as far as I can tell, follow no established conventions at all. Most papers are of course just fine. Authors do tend to find a good way of presenting the argument, and I do think most of the heterogeneity can be ascribed to variation in the structure of arguments. However, too often one does come upon a paper that is so ill-structured that it would have been better for it to have been rejected simply on that basis. And sometimes you find papers that do not come to any conclusion at all and you are left wondering what the point really was.

I think there are two causes to the situation. One is that there is no convention about the structure of philosophical papers other than those decided by individual journals, and I don't think philosophers in general see a need for such a convention. The other is that until very recently philosophy students didn't get much of a training in methodological issues. The idea was that each student learns to write papers by writing essays in the various modules, and pick up other things as you go along. But, it is then likely that they will pick up good and bad habits on the basis of the good and bad habits of the teachers they happen to encounter. We are leaving all of this a little bit to chance, I feel.

The worst thing about this is perhaps that philosophy becomes guilty of the very sin they complain other sciences commit, lack of self-reflection. We think all the special sciences must have courses in scientific method and scientific thinking, and that this must include philosophy of science, but there haven't been much in the way of philosophy of philosophy courses for philosophy students. But what is there to teach, if there is no consensus about what good philosophical practice is? I am not altogether decided on whether this is good or bad in the end. It just bothers me that philosophy seems to think that philosophy is a special case, without being able to say why.

Rognavuldur: I regret my post was ambiguous. When I referred to "the structure of research papers in the field" I meant research papers in the computing disciplines, not research papers in Philosophy! I guess I take it for granted that Philosophy is not an empirical science and so I don't expect philosophical papers to follow the established norms of a scientific paper, but I do find it strange that computing papers don't do so (except perhaps those which are concerned with purely mathematical issues).

To return to your point ""what is there to teach if there is no consensus about what good philosophical practice is?" - this seems to be a general problem of the humanities: before I studied Philosophy and Psychology, I took a degree in History, and was quite frustrated that nobody addressed what it was that constituted a good historical explanation. Recently a former fellow student of mine gave a valedictory lecture, in the course of which he said that Prof Trevor Roper had taken his duties as Regius Professor very seriously, in particular guiding PhD students on better writing style. I was astonished at the thought that good style rather than sound argument constituted "seriousness" in historical work. I certainly think clear presentation is important but what value is there in a clear presentation of muddled thought?

Julian: I greatly appreciate your comments.

Going back to the original question. The study of science itself, or for that matter the scientific method that emanates from this discipline, is a philosophical study. It is a study that focuses on the nature of knowledge and how it is acquired.

Many of us were taught the rudiments of the scientific method in a high-school physical or biology class. To study the method, however, requires a deeper appreciation of concepts such as the theory of knowledge, the theories of truth, or even the definition of knowledge.

Teaching the method is not particularly difficult. Without a firmer grasp of the underlying fundamentals, this becomes a mechanistic or rote understanding of the method. There is a difference in being told that hypotheses are set up to falsify a claim and consequently executing according to instruction, vice understanding Popper's arguments on falsifiability.

Understanding the nature of science raises a broad spectrum of questions. Many of these underscore that science is itself just a method; in essence a way to execute a discipline or belief structure. Science is also extremely empiricist in nature, relying on the notion that the requisite justification of a claim to establish it as knowledge requires some form of observation. This brings into question whether there is knowledge in mathematics, or for that matter in theoretical physics. Einstein was undoubtedly a physicist. Was he however a scientist?

I believe that both the nature of science, as well as its mechanics can be taught at quite an early age (e.g. high school). The topics however need to be taught in context, using appropriate approaches.

I think that the YouTube video is a great example of children debating the nature of knowledge, which actually has tremendous bearing on the understanding of research, and ultimately science. https://www.youtube.com/watch?v=_ldGT2R-pJM

Of course, this does not directly answer why science is not taught more broadly.

Hmmm. "The study of science itself ... is a philosophical study". Well, about 99% of existing disciplines seem to have been "spin offs" from philosophy. But we tend to suppose that those disciplines we call "sciences" are distinguished by some special characteristics and often this is thought to be a generic "scientific method" - with the corollary that if we want to put knowledge of a new field on a sound footing then we ought to apply that method to the new field. Philosophy of Science looked like the way to identify the essential characteristics of that method and thus make it applicable to the new emergent fields. But there's now a contrary tendency to emphasise the situatedness and specificity of the methods of the special sciences, and to undermine the idea of a generic scientific method. Well, if Psychologists can do without the Mind, I guess Philosophers of Science can do without the Scientific Method ;-)

We have reached a point where the use of science in the common vernacular is about as distinctive as the title of Vice President in corporate America. It is no longer a good indication of what is being done. Science has been appended to just about everything because of the extremely positive connotations that it brings. It is only science in name.

Science is quite distinctive, and although certain fields may emphasize some facet over another, the principles are not particularly 'negotiable'. I believe that where there is an emphasis on the specificity of an application, and the argument that a 'special' science is being applied is a cop out. If what is being applied does not comply with at least some of the basic principles on which science is grounded, it is not science.

That is not particularly a bad thing unless, of course, one insists on having the titular form of science, AND retain the belief that science is being practiced. This is often due to the misconception that science is the only means by which knowledge is acquired. It would not be of great consequence if it did not result in flawed research through the dilution of foundational canons. It results in bad science, and for that matter in flaws in methods other than science that are pushed to be 'more scientific'.

Interesting discussion of the philosophy Of science. On a slightly different tack - to attempt to answer the original question, One reason science students don't get taught scientific method earlier is that most people, including science teachers at high school, do not really understand what it means to 'do' science, except in simplistic superficial terms. Discussions of the philosophy of science support this view, as even those who have given it serious thought disagree.

OK, so we have the "prestige" factor of wanting to be called a science - the use of the term "Computer Science" to describe what is basically a technological or engineering discipline would IMHO be an example of that. But when we look at technological disciplines they do generally give rise to applied scientific issues (what some have called "engineering sciences") and in order to research these I suppose we should observe the same norms as are appropriate to "pure" science. We also have a range of problems and issues which are tending to undermine confidence in scientific method. Some of these are mainly logical, e.g. the Quine/Duhem/Lakatos position that there cannot be a crucial experiment that conclusively refutes a theory - this tends to undermine the pure Popperian view. Some are more sociological - studies of how scientific communities actually behave don't show them conforming to the Popperian norms. Some are technological - the increasing dependence upon computational science, big data, simulations, so-called in silico experiments etc. Thus we have Paul Humphreys arguing that scientific epistemology should no longer be anthropocentric.

I would argue that Popper is still fundamentally correct on two points:

(1) there is no valid logic of induction from particular observations to general laws or theories (as Hume pointed out), only from general laws or theories plus initial conditions to expected particular observations;

(2) we ought to subject our theories to the most rigorous tests that we can devise, rather than (for example) abandoning our judgement in the face of simulated results.

What about Bayesian statistics? Bayes does not refute Hume, he simply produces a mathematical model for revising a judgement of probability in the light of new evidence - and while this model may or may not be an empirically adequate representation of how we think about probability, it does not and cannot show that our psychological propensity to think inductively (which Hume recognised) has any logical validity.

So why did the discussion of this question apparently die off? Does this perhaps show that philosophy of science reaches an impasse which makes it distasteful to science students? I still have hopes of making computing students think more like experimentalists ...

I don't want to get involved in a protracted discussion of "philosophy of science", but as a practicing scientist/educator/mentor (US university-level) and a student of the history & philosophy of science, I can make a few general observations in response to the original question.

* First: Why don't professors teach more about the practice of science? (Here I mean experimental "natural science".)

i) Students who are not going to be professional scientists want to know the "facts" obtained by science because that's what they'll be asked for on their medical school (and other) entrance exams.

ii) It's easier for a professor to explain how things work than how we gained that understanding. Explaining the historical background behind important discoveries is boring to students who want "just the facts".

iii) Most importantly, science cannot be taught in a lecture hall. Science can be learned only by doing it! This is why research universities so strongly encourage undergraduate science majors to become involved in independent laboratory or field research under the guidance of a faculty researcher. In practice, students who are interested in careers in research -- whether or not they go to grad school -- need to have undergraduate research experience. It is also valued by the health professions!

* Second, in the famous phrase of Richard Feynman, "Philosophy of science is about as useful to scientists as ornithology is to birds" -- in the strict sense that "philosophy of science" does not inform the practice of science. Philosophy of science is an attempt to explain to laypeople what science is, or how scientists work, or what knowledge is, all of which is irrelevant to the practice of research (research results either work or don't work). Unfortunately, this has had the unfortunate side effect that too many working scientists, being generally ignorant of the history and philosophy of science, make public statements about science that are uninformed, ignorant, or arrogant.

* Third, there _is_ no (one) "scientific method". Paul Feyerabend came close when he said, "Scientists do not solve problems because they possess a magic wand - methodology, or a theory of rationality - but because they have studied a problem for a long time..". (Feyerabend was _almost_ right in "Against Method"; he came up short partly because he was overly-influenced by theoretical physical sciences, and partly because, as he said in his memoirs, he was young and arrogant.)

Incidentally, Karl Popper's notion of "falsification", proposed as a means of distinguishing "science" from "pseudoscience", is no longer credible. ((Feyerabend apparently disproved it on purely logical grounds.) Scientists don't actually work like this; instead, they'll say that a hypothesis isn't useful if it cannot (in principle) be tested experimentally. Moreover--and a point that most scientists aren't aware of--a hypothesis or theory *cannot* be "falsified". Only statements of fact (data) can be called "true" or "false". Because a hypothesis is a statement about relationships among data, it--like a logical inference--is said to be "valid" or "invalid". In practice, the validity of any hypothesis depends upon the truth of its underlying data: this is why we say that theories come and go, but good data--solid data--are forever. In other words, incorrect data can drive misleading hypotheses forever, but correct (and important) data will sooner or later lead to a valid hypothesis.

* Finally, my crankiest complaint: The illusion that there is a "method" or neat formulaic process which, when applied properly, will lead to "scientific discoveries" is, I argue, promulgated by K-12 teachers who have no first-hand experience with basic research, have themselves been mis-taught about it, or think it's "too complicated" for grade schoolers. (This attitude results in what I call the "science fair" approach to science -- but that is a separate rant.)

Peter, thank you. I think you have provided a very clear statement of a popular view, some of which I think is true and some not so true, but it gives a really good basis for discussion.

I think you are spot on about the attitude of a lot of students, especially those who are not going to be professional researchers. However, we should know better than to give them what they want; we should give them what they need. A medical student who is allowed to go through education by memorising facts and repeating them like a parrot, never learning how we got to know these facts, will become a bad doctor, no matter whether they do not have the intention to do science. Not the least because in most medical professions you are now expected to do some kind of research alongside your clinical practice. Anyhow, a lot of students don’t realise they want to be professional scientists until they’ve finished their degree. As a result, a large proportion of research done in medicine is done by medical professionals who really struggle because they haven’t got the training in doing science because they did their education believing it was all about the facts and their professors only encouraged that attitude.

It is exactly right that it is easier for professors to just explain how things work than how we got to know how it works, which is why too many professors only give the students what they want and not what they need. But, it isn’t really too much of a struggle. I have taught research methods in dietetics, catering management, physiotherapy, nursing, occupational therapy, social work, home economics, and they all accept why they have to learn the philosophy of science when you explain it to them.

You are exactly right that science cannot be taught in a lecture hall, which is why all undergraduates should do a research project as a part of their final examination, with the guidance of a lecturer. In Sweden this is standard practice. Final theses in the empirical sciences are all ‘by resarch’, although there is great variation in the quality of that research because there is great variation in the quality of the supervision and training in scientific thinking at various departments.

However, I have to disagree when you say that "philosophy of science does not inform the practice of science”. Aren’t you guilty of the ignorance of the history and philosophy of science that you yourself bring up? Aren’t you thinking that because you don’t see philosophy of science telling you anything today about your science, then it does not inform it?

Philosophy, Science, and the Philosophy of Science have a common history. Galileo Galilei was not just a scientist, but a philosopher of science too, as was Ibn al Haytham (Alhazen) before him, and Bacon, Newton, Descartes and Leibniz later. Philosophy of science didn’t start in the 20th Century. All these people contributed to science as well as philosophy of science. You shouldn’t confuse the fact that active researchers in physics have contributed to the development of ideas about how to do science, with the fact that such development is philosophical no matter what degree the contributors had.

Ideas about how one should do science simply is not a scientific discovery similar to how the neutrino was discovered. I suspect you overlook that the ways of thinking that have now become obvious, weren’t always so, but have only become obvious through centuries of ‘indoctrination’. You may not have received your understanding of science from a philosopher, but your discipline and the thinking deployed in it, sure as hell has been influenced by philosophy. This influence has become such an ingrained part of each scientific discipline that they no longer treat them as philosophical; they have become the rational backbone of the discipline.

As for Feyerabend getting it almost right, you may be right. However, I think one of the dimensions of ‘almost’ is that he didn’t recognise that the kind of thinking he himself deployed in the criticism of the idea that there is a single scientific method, is arguably the kind of thinking we really are looking for when we are looking for the “scientific method”; i.e. the method is not really a method of empirical investigation but a mode of thinking.

Let me put it like this. First, exactly like you yourself say, ‘Science’ is often associated with a belief in some “neat formulaic process which, when applied properly, will lead to ‘scientific discoveries’. I agree that this is obviously wrong. If there is a common element in all research it must be some extremely abstract model of thinking, because it can’t be identified with any particular procedure deployed in any particular discipline; it must be a manner of thinking that is supposed to be present in all the different procedures used, be it the use of Dual X-ray absorbtiometry to determine bone mineral density, surveys to determine psychosocial wellbeing in some population, or interview studies of how the fatally ill cope with the knowledge of death. So, being scientific in all these situations must refer to the way we attempt to produce a valid conclusion in a particular situation, but it must be abstract enough to fit them all. So, I think Feyerabend is right to say that there is no one procedure, but overlooked there is a kind of thinking that we all employ, and that has to do with rationally scrutinising the data, the method… everything, until we can think of only one reasonable solution.

Being scientific is to constantly asking the question “could we be wrong” and to do something about answering this question and minimalize the risk of having made mistakes or somehow wrong. I think Feyerabend’s criticism of the “one procedure” is good, but I tend to think he completely lost his marbles when attacking also the idea that there is one kind of rationality. The decisive difference between science and pseudoscience is simply that science asks and is willing to confess mistakes and to do something to correct them, while pseudoscience does not accept fallibility and does nothing to either undermine or justify its position. The procedure for Astrology was determined millennia ago, and since then no astrologer has asked, AND acted on the question, could we be wrong and how could we find out if we are wrong. If they did, they would no longer be doing astrology but science; just as a scientist when asking ‘how should we do our science’, is no longer doing his science but is doing philosophy of science. Science is all about finding the truth while admitting that we are fallible and therefore go to great lengths at minimising the risk of mistakes. A valid conclusion is not really, not in practice, a true conclusion, it is the conclusion to which we can find no reasonable alternative, and therefore accept it as true until further notice.

That was my rant about science.

Feynman's 'ornithology' jibe might have been justified by the vacuity of much philosophical writing in mid-20th century, but if we take it seriously for a moment we can see that "is this a bird?" is a very different kind of question from "is this a science?". Unlike biological classification, the criteria for accepting that a knowledge claim is "scientific" include a normative element, and attempts to define "scientific method" are simply attempts to codify those norms.

In the context of a specific experiment, the hypothesis is a predicted fact deduced from the combination of theory and initial conditions, and if the observed fact differs from the predicted fact then the hypothesis is indeed falsified. Of course, this may point to a failure on the part of the experimenter to produce the intended initial conditions, or it may point to the theory not being a true universal statement. But if a theory is so vague and flexible that it can't produce a falsifiable hypothesis, I think we are justified in regarding that theory as unscientific - and I fear that much of what today passes for science does not meet that "normative" criterion - notwithstanding which its practitioners appeal to their own authority as "scientists".

Those students who only wanted to know the authorised facts in order to pass their exams are likely to learn, in the course of a working life, that a great deal of what they had been taught now turns out not to be true - or not quite true - or true only subject to special conditions - or whatever.

I would like to add now a mini-rant about philosophy: Every philosopher should be exposed to doing experimental work in at least one science, and to writing up a scientific investigation in a way that obliges him/her to engage with the primary literature of that science. Otherwise their philosophy will indeed be as useful - to laymen as well as to scientists - as ornithology is to birds.

I think the reason why students get taught so little about the scientific method is because there's no strong consensus on what's important about it that we want to convey.  This makes it harder to teach teachers about what it is about the scientific method we think should be worth communicating. 

For example, the Edge question of the year 2011 was, "WHAT SCIENTIFIC CONCEPT WOULD IMPROVE EVERYBODY'S COGNITIVE TOOLKIT?".  Out of 167 respondents, there were 167 different answers.  Really.

So, if the question was "what is the scientific method?" we could talk about different things.  Peirce and others have argued that the thing that makes it important is not the knowledge produced but the method of knowing because it precedes the knowledge.  What exactly is the normative method of knowing is, itself, contestable. 

Rather than going into a polemic about why this over that, I will just agree with Peirce and say that the scientific method is thinking correctly, that is, to apply the sequence of reasoning rigorously, viz., abduction, deduction, induction/inference to the best explanation/repeat.  Moreover, Simon has argued that a scientist is, in fact, a problem solver.  Combined, this would suggest we should know this domain of overlap between scientific method, critical thinking and problem solving clearly and explicitly...but how can we get everyone to agree to such claims for more effective communication?  For example, why this and not observation/hypothesis/examine/analyze/evaluate/etc...?  (There's actually a good reason why but I won't discuss it).

There are also logistical problems, like producing a generation of teachers with enough knowledge of today's technologies with consideration to existing structures so as to effectively communicate the different steps of thinking/reasoning/understanding. 

All people frequently rely on some pieces of tacit/intuitive/practical etc. knowledge without firm theoretical foundations. It is an undisputable fact of experience that people are usually quite uninterested in meta-analyses, even when they themselves are more or less directly involved. Consider, for instance, the men in the street who typically are not very well versed in psychology and have only an amateurish knowledge of the studies on sexuality, or religious people who show the same lack of concern towards the philosophy, history or sociology of religion. Scientists’ indifference to any kind of “science theory” is no exception to the rule.

Dear friends, I arrive late to this discussion (I was traveling, in fact). I can see many valuable and well grounded answers and comments on the question raised by Mark.

If possible, I would like to suggest a different take - a radical one, if any. To begin with, as we know, there is no such a thing as "the" scientific method. From this point of view: a) It is indeed desirable not to teach such fantoms and ghosts as "The scientific method", something that is truly and invention of some philosophers of science (all of them led by the lights of Popper).

As a consequence, it is much more sound (to me), f.i. as H. G. Callaway says, to show real experimental, deductive, thought-experiments and several others in each case by each scientist. Therefore both the history of science and the philosophy of science go hand in hand - something that goes without saying. Close to this, teaching some psychology of the scientific discovery might be very useful. This opens the door, f.i. to the importance of failures in discoveries and inventions, in arguments and demonstrations. I do not know of many places where this is being done, although we all know about this (at least in this post).

Dear Mark,

Dear All,

BSc and MSc trainings generally do not focus on research and methodology. This is because mostly these degree holders are produced for the practice. I teach agricultural entomology and recognised about 20 years ago this training lack in our curricula. We planned and announced a subject: research and evaluation methodology in crop protection. However, students do not chose it because this is a facultative subject and students do not like to increase their requirements. Certainly, the lack of methodical knowledge can be seen in the diploma and degree works.

Hi Mark; Here in India, serious practical training started during eleventh and twelfth class. Though students try to escape these lab classes and teachers also do not give there serious view. It is really true that science students are not being taught seriously on scientific methodologies. For preparing good researchers for doing good science we must take serious view by the teaching community.

Regards

Dear Mark

Do you really think that there is only a single scientific method (Conventional Scientific Method with its empiricist philosophy) that is followed even in a single discipline or even in a research paper. Most theories in science are built upon other theories, accepted by some criteria as true, only through speculation and applying deductive not inductive logic as scientific methods demands. Some researches are descriptive in main part and these descriptions lead to deductive inferences and many times concrete conclusions. No researcher can work only with scientific method's inductive reasoning alone, to interpret results of analysis in the theoretical context, one has to deduce findings and also arrive at conclusions by arranging results in a taxonomy of the sort (description).

However, at school level when students are being taught elements of science, it is better to let them master basics.

Of course, at master/major graduate/ PhD entrants must be taught snot only scientific method but also philosophy of science with theory of knowledge etc.

Mark,

If someone want to become a good runner, he needs to run and to train intelligently.  She may need to get a trainer.  But the runner does not need to learn a running method and try to follow a running method although the trainer may have a training method.  But the trainer do not have to apply a trainer method to get a training method.

Philosophy of science is not intended to train scientists.  No.  It is a self-reflection of some scientist or philosopher trying to reflect on their practice.  The purpose of this self-reflection is not to create a manual about how to train a scientist.  Some scientists will feel the need through their practice to inform themself about these reflections.  It may at this point be of used to them given that they felf this need.  But they are in a position to reject the philosophy of science that do not make sense to them and select those that feel right.  A young scientist in training is not in such position.  Better to train by doing it and later in life he/she may if she/he feel like it get into philosophy of science..  

Thanks Mark; for sharing this important issue. I believe that many pass there scientific life and they do not come to read about things like inductive reasoning and the better ideas of Karl Popper (falsificationism) and never  read  "What is this thing called science" by Alan Chalmers.

But it is true we (Faculty) are not seriously giving the importance on science methods. The syllabus must be improved at school and university level for significant improvement in science in all over the World. Thanks

@Louis

I agree with you that in a state of steady progress one needs not to look over one's shoulder. So much so, Stephen Hawking may afford to declare that “Philosophy is dead”. Is philosophy really dead? In fact, at a point in time of progress scientists have to look over their shoulders whether they are going in right direction, whether they have really made progress or whatever they believe is only their perception and has nothing to do with reality. Should they not evaluate their methods? Sooner they do it, it will be better for both science and society. Hawking in his book, 'A Brief History of Time' has itself acknowledged this possibility.

Dear Mohammad,

Hawking is the proof that one does not need philosophy to succeed in science.  He is a  philosophical moron and still he is bright scientist.  But not all bright scientist are so dim about philosophy.

Carlo Rovelli wrote:

"It's sort of the fashion today to discard philosophy, to say now we have science, we don't need philosophy. I find this attitude very naïve for two reasons. One is historical. Just look back. Heisenberg would have never done quantum mechanics without being full of philosophy. Einstein would have never done relativity without having read all the philosophers and have a head full of philosophy. Galileo would never have done what he had done without having a head full of Plato. Newton thought of himself as a philosopher, and started by discussing this with Descartes, and had strong philosophical ideas.

But even Maxwell, Boltzmann, I mean, all the major steps of science in the past were done by people who were very aware of methodological, fundamental, even metaphysical questions being posed. When Heisenberg does quantum mechanics, he is in a completely philosophical mind. He says in classical mechanics there's something philosophically wrong, there's not enough emphasis on empiricism. It is exactly this philosophical reading of him that allows him to construct this fantastically new physical theory, scientific theory, which is quantum mechanics.

The divorce between this strict dialogue between philosophers and scientists is very recent, and somehow it's after the war, in the second half of the 20th century. It has worked because in the first half of the 20thcentury, people were so smart. Einstein and Heisenberg and Dirac and company put together relativity and quantum theory and did all the conceptual work. The physics of the second half of the century has been, in a sense, a physics of application of the great ideas of the people of the '30s, of the Einsteins and the Heisenbergs.

When you want to apply thes ideas, when you do atomic physics, you need less conceptual thinking. But now we are back to the basics, in a sense. When we do quantum gravity it's not just application. I think that the scientists who say I don't care about philosophy, it's not true they don't care about philosophy, because they have a philosophy. They are using a philosophy of science. They are applying a methodology. They have a head full of ideas about what is the philosophy they're using; just they're not aware of them, and they take them for granted, as if this was obvious and clear. When it's far from obvious and clear. They are just taking a position without knowing that there are many other possibilities around that might work much better, and might be more interesting for them.

I think there is narrow-mindedness, if I might say so, in many of my colleague scientists that don't want to learn what is being said in the philosophy of science. There is also a narrow-mindedness in a lot of probably areas of philosophy and the humanities in which they don't want to learn about science, which is even more narrow-minded. Somehow cultures reach, enlarge. I'm throwing down an open door if I say it here, but restricting our vision of reality today on just the core content of science or the core content of humanities is just being blind to the complexity of reality that we can grasp from a number of points of view, which talk to one another enormously, and which I believe can teach one another enormously. "

Mark, you have asked a first-rate question.  Let me try to offer an equally good answer.  An answer that I don't see here is the lack of high-quality scientific biographies (in various languages), accessible to and readable by students and by their teachers, of Galileo, who put scientific method into its modern form, and of other Renaissance giants.  This lack exemplifies the sometimes great difference between market value based on "wants" versus real value based on "needs", many of which (needs) are unperceived without educators who themselves are educated on the history and philosophy of science.  Even if a good account of the creation of modern science is written in one language, the lack of market value (in relation to real value) leaves authors and publishers with insufficient capital to translate such work into other languages.  How do I know this?  From direct experience.

What’s the purpose and can we get beyond the fact that the label “scientific method” is attached to so many different things?

Is it not more important to show students that there is an honest attitude toward getting at the truth and that there is a most reliable way to get there?

“There are in science three fundamentally different kinds of reasoning…These three kinds of reasoning are Abduction, Induction, and Deduction. [—] All the ideas of science come to it by the way of Abduction.”  ~ Peirce

But how many of us have even heard of Peirce’s philosophy?

What’s a good example of Peircean abduction?  What’s the best example?  Why?  Is this not teachable?

Since I have worked in research labs, for 25 years, I might be able to offer some insight to this question.  It has to do with the way that scientists reproduce themselves.  People with an interest in science take classes like biology, chemistry, microbiology etc..  These classes don't teach you how to "do" science rather they just teach you basic information with, I might add, little connection to each other.  Then you graduate with this degree in microbiology or chemistry etc.  People then go on to some medical field, or maybe find themselves teaching in a high school, or end up in a research lab.  It is in the lab where the individual is taught how - that lab - does science/research.  You get bits and pieces of the scientific method-this is how we run an ELISA- oh yes we have controls and we run our samples in duplicate.  If you are lucky, you help come up with experiments.  If you decide to go on and get a masters and then a PhD then you might take a class in experimental design, which actually is mostly statistics.  No philosophy classes for sure, just more advanced classes, where you begin to read journal papers and begin to learn how to question what is being published.  You have heard or read the words: hypothesis, observation and conclusion by now because, they frame how you present your work,  But no, we are not formally taught "the scientific method."

Stefan,

Today students are far from actual experimentation in real circumstances; They are exposed to   totally pre-plan laboratory procedures where students are just going over the motion and asking each other what I am suppose to observe now, what is the results I should get. I failed the experiment and got bad results.  As if passing a test!!!!  This is a parody of the scientific spirit which is one of curuorisity, observation, groping in the dark, etc.  The modern scientific spirit grew in the 16th century out of gradual dissatisfaction with the learning from books written millena ago by great masters.  The new spirit was to learn directly from the only great master: Nature.  But as you said, mass education has brought back the old efficient method of learning from books only and fake expirimentation.  Science was originally developed by a subgroup of people of the elite class.  They had excellent and wide ranging education provided by great masters and they were doing experimentation first hand.  Those that want to earn money in that class did not go for sciences; they had much better options.  Most people involved in science then were totally motivated and interested and they shared their passions with their students.  Who among you really had such a chance in the modern age?  I envy the few that had that chance.

 Polanyi contrary to Popper was actually a first rank scientist before he changed carrier and became a philosopher.   Although Popper's clarification of falsification benefits the majority of scientists today, Polanyi's ideas on science are less well known.  Kuhn's idea which borrowed from Polany are well known but Kuhn did not touch on the discovery process itself and concentrated on the social and historical dynamic of the social process as Polanyi had also described.  

Isn’t the problem to deal with what we have now and make it better?

“’The Scientific imagination dreams of explanations and laws.’  CP 1.48

“Charles Sanders Peirce was a restless and passionate man.  His greatest passion was philosophy, especially that branch known as logic.  A person of immense intellectual power, Peirce devoted much of his thinking life to the analysis of reason itself…

The origin of new ideas and scientific hypotheses was always of particular interest to him, and he seems to have had a life-long fascination with the question of genius and creativity…Producing a harmony of creativity and logic was one of Peirce’s underlying philosophic tasks-and in order to do so he had to reformulate logic itself in a radical way...”

~Kathleen Hull, “Why Hanker After Logic?  Mathematical Imagination, Creativity and Perception in Peirce’s Systematic Philosophy”

“Peirce described his thought as “the attempt of a physicist to make such conjecture as to the constitution of the universe as the methods of science may permit, with the aid of all that has been done by previous philosophers” (CP 1.7)” ~ Parker, “The Continuity of Peirce’s Thought”

"Think of twentieth-century philosophers, and a host of polysyllabic European names, by turns guttural and sibiliant, come immediately to mind.  Pronounce the odd-sounding “Peirce,” and many will either not know what you are talking about, think you have sneezed, or direct you to the bearded nineteenth-century logicians conference in the next building.  But it is a name well worth looking into.  Charles Sanders Peirce was a logician and scientist whose thought spawned pragmatism and modern semiotics….

Though neglected for much of the twentieth century, his thought is enjoying a renaissance as researchers from disciplines as diverse as logic, psychology, pedagogy, cognitive science, mathematics, and artificial intelligence are discovering a veritable goldmine of ideas that allow them to obviate the positivistic clockwork conception that has so profoundly informed modern research into the natural and human worlds….

Of the myriad concepts that make up the constellation of Peirce’s thought, there is one that is of special interest to many researchers in cognitive science and artificial intelligence, namely, his notion of hypothesis formation or what he called abduction.  As the available literature attests, the AI community is interested in abduction in hopes that the process can be formalized and implemented in a wide array of intelligent or expert systems….(Costa, “Can Creativity be Formalized?”)

“What is Science? We cannot define the word with the precision and concision with which we define Circle, or Equation, any more than we can so define Money, Government, Stone, Life. The idea, like these, and more than some of them, is too vastly complex and diversified. It embodies the epitome of man's intellectual development...

To most men, including all who are outside of the world of science, the term means a particular kind of knowledge.  Wherein lies the essential peculiarity of this knowledge? 

Some thinkers agree with the ancient Greeks in making it consist in the Method of knowing, the manner in which the truth is laid hold on...

There are in science three fundamentally different kinds of reasoning…These three kinds of reasoning are Abduction, Induction, and Deduction. [—] All the ideas of science come to it by the way of Abduction.  Abduction consists in studying facts and devising a theory to explain them. Its only justification is that if we are ever to understand things at all, it must be in that way…” 

~ Peirce

But how can we recognize a good example of Peircean abduction?  Can we teach students this?  Of what use is a good, simple, correct formalization?

Jerry,

Can you elaborate a bit further how Peirce describes abduction? Michael Polanyi has also described the creative part of science.  For Polanyi, the familiarity with the phenomenal domain, not only a theoretical familiarity but a lot of hand in experimentation gradually created into the scientist nervous system an implicit knowledge of the phenomena. This implicit knowledge then gradually guide the scientist onto new experimentation or new theoretical ideas.  For Polianyi, the creation of the implicit knowledge is similar to the creation of perceptual gestalt.  It is essentially an natural perceptual development.  But the path to find out an explicit theoretical expression although guided by the implicit knowledge is not straighforwards but a gradual process that may take years.  One of Polanyi's famous saying is that we know more than we can say.  Knowing this, how implicit knowledge naturally developed should encourage the scientists to thrust their intuition even though they have no clue where it is leading them.  They should follow their hunch even though they cannot justify initially their intuition.  It is not a method at all but simply an encouragement in not always doubting what we cannot immediatly justify.  This is a big problem in today science where a research proposal has to clearly explain what the scientist is supposed to discover in the next five years.  I have a hunch is not going to convince a lot of people although in reality this is the biggest predictor of success.

Dear Louis,

There are too many things to be said about abduction.  :)

I've been writing about it though! "What is the best example of Peircean abduction?"

In focusing on what you said above, the closest thing that comes to mind is “collateral experience”. 

For example, there are lots of natural phenomena out there but a useful one will display regularity so that you can keep poking at it and testing it to figure out the rules of the system.  If you immerse yourself in enough phenomena and put them next to each other, you can extract deep homologies, which can be used as heuristics for application in novel domains.  Awareness of such rules constitutes a part of one’s collateral experience (but it means more than this because the object being considered is also imbued with collateral information).  I think Goethe has a lot to say about this matter of developing the scientist. 

Abduction is the tool by which to constrain such experience to solve problems/give meaning.  One way it does this is by forcing you to be explicit about what you say is happening at the outset.  Abductions are sometimes ill-structured, that is, either C, A or both are ambiguous but the goal is to make them clearer and make them match.  Also, a good abduction is induced by genuine doubt.  So, you might have a hunch but you should be able to justify why this hunch instead of that one (e.g., your training/skills aren't necessary in that one and you can't suggest an alternative).  It might be genuine to you but you may have to explain enough of the collateral information to others so as to convince them why the hunch is a surprise that can be resolved.  It's a structure that helps to make the argument clearer. 

There are different degrees of creativity.  For example, you can use abduction to figure out where your keys are (once you find them, you don’t keep thinking about it), you can use it to design and execute a novel project or you can use it to do convergence science.  The argumentation (abduction with precisely fitted premises) gives you an idea of the magnitude and one should keep an eye on its transformation in futuro, the vision, to help articulate the justification.  Some justifications are better than others and economy is a huge reason given for further pursuit. 

teaching science properly has  always been a challenge to many teachers. In many classrooms science is taught like one would teach History or Literature. There is more of memorising facts and very little investigations. Some experiments may be done in the laboratory but the approach is merely to show students some classic experiments. Science in the classroom is not about investigations and problem-solving, students being presented with challenging problems at a suitable level which with the help of the teacher they can explore and solve. In this way they will be working like scientists but at an elementary level and beginning to understand what it means to be a scientist. Because of the poor science teaching methods, students end up not knowing the authentic nature of science.

Eunice, your sentiments (joy of discovery) have been echoed by many others, especially Alberts.  So, what's the solution?  Where do we start?  How can we show science teachers that yours is the correct way and ensure that those who have it wrong are able to correct themselves?  If not in some simple formalism that allows expansion into correct complexity, which?

Here is a video that makes best use of what you're talking about:

https://www.youtube.com/watch?v=O1dA0sMSobE

Thanks Jerry, you are spot on ! That is the kind of science teaching I was talking about . I have viewed the video but I can tell you that in some schools  old fashioned conservative principals would describe the class as chaotic just because children are not sitting in rows quietly listening to a teacher. Such principals might not realise that the children are engaged in creative thinking and stimulation of problem-solving skills. Proper training of teachers is essential to try and change their mind set about how science should be taught. Once the teachers get the hang of hands-on, minds-on science teaching, both them and the children enjoy science. It is true that using this method of teaching science needs more time to prepare materials and so you do need teachers who are passionate about teaching science successfully.

So not only do we need to teach teachers but also principals.  But why would they change when they're principals and I'm not?  So goes the world...

You'll notice in the Next Generation of Science Standards do not use the scientific method wording, but rather the practices of science, for good reasons. Those reasons are articulated in the Framework better than I could probably paraphrase them. It is a free download from the Academy (http://sites.nationalacademies.org/dbasse/bose/framework_k12_science/index.htm). Multiple scientists have spoken into the need for focusing on the practices of investigation and engineering, rather than a didactic scientific method on a worksheet, as many teachers still do. 

Dear Mark:

A very interesting and important question.  Perhaps general education students from K -16 should be taught the history, philosophy and sociology of science directly and science itself indirectly as a kind of science and engineering appreciation course.  Let's face it, very few students will actually end of doing science and engineering, and the problem with many students who do so on the university level, is that they really do not understand what they are doing and are in STEM disciplines because that's where the jobs are and not because they appreciate what they are studying.   

Mike

Dear Manuel:

McLuhan said that "art was what you can get away with."  An eminent anarchist philosopher of science, Paul Feyerabend, said the about same thing in regard to science (Against Method).  He, along with Thomas S. Kuhn (The Structure of Scientific Revolutions) and Karl Popper, known for his falsification theory, were considered the "hated triad" among scientists because of their questioning of traditional interpretations of the nature of science and the scientific method.  You might go back and look at  formal logic again as it really is an art form taking minimalism as far as it can go.   

Mike

You talk of formal logic as a true thing but I doubt I agree with your definition. 

What do you mean by formal logic?  Your explanations can be helpful for science students studying the scientific method.   

According to Karl Popper: "the only method we have is trial and error". 

What is the sicentific method? does any

In trying to give reasons for why students get taught so little about the scientific method,

I don’t see how your burglar example demonstrates formal logic as an art form taking minimalism as far as it can go.  It just shows that we make mistakes in reasoning, contents of arguments matter and we don’t always care to analyze the form of the argument.  The particle collider example is too esoteric for me to understand and I consider myself a reasonably educated person.  It just makes me think of physicists as arrogant folk who thinks everyone should care about what they think about in the way they think about them. 

I am also skeptical that we have a common view of what the traditional nature of science and the scientific method consists in.  However, if you can make that clear, it would be a good starting point.

It appears, then, that there are many reasons for why students get taught so little about the scientific method, one of which is that there’s too much disorganization. 

I often wonder if the confusion is intentional.  Why do we even want an informed citizenry?

Then how is that formal logic?

in either situation in which omission appears to be the defining feature of making a correct judgment.  If omission is the important bit, then what does scientific method have to say on omissions? 

I don't want to lose sight of the original question.  The goal is to determine why students are not taught enough about the scientific method. 

Have you seen what is happening in medicine with regards experimental replication?  How are students supposed to make sense of the scientific method if your model is so sensitive to contradiction?

This is a very good question, and such an omission appears to be very wide spread. I apparently recognised this and did 'Knowledge and the Sciences' run by the Philosophy Department and this gave me knowledge of induction, deduction, Popper, and other Philosophers of Science and so on, and I later completed a degree in philosophy. However, in my opinion, all science students should automatically be taught something about such subjects in their science degrees.

Abduction is neglected, again.  Never fails.

Alright Jerry, add abduction too if you wish. I have no objections.

Much that are returned for a google search on “problems with replication in medical research” will serve as the things in particular to which I am referring, but it has a lot to do with the complicatedness of biological data.  How much do you, as a physicist, know about how medical data is collected to support conclusions?  Importantly, how much do you care to know?  Yes, experience matters.

This notion of causality, for example, is grounded in experience, as Goethe put it.  He replaces “causality” with “circumstances/conditions”.  So, without knowing context, it’s difficult for a reader to know what you’re talking about, much less a concept/generality of "causality".

I believe causality has to do with valuations for the connections within a context but this varies a LOT.  What caused the Greek debt crisis?  What caused World War II?  The big problem is that when two people with different Weltanschauung (worldviews) discuss a context, it is precisely the causality they are arguing over. 

A method grounded in notions of causality may be too controversial to serve as a way to teach about the concept of scientific method, unless it is to bring attention to the different attitudes and discuss the relations. 

Manuel, I think our conversation adds to reasons why students are taught so little about scientific method.  So, how to make things better?

A defeasible argument has been presented, the Neglected Argument...but it's neglected.  People give different reasons for why; Peirce's ideas are too complicated, they lack coherence, he was a cad, didn't play well with others...but I think he was a thinker ahead of his time and realized this, so decided to leave us with a coherent formalism for the modern age that values computation. 

For example, he anticipated our conversations and left us with his replies, which I will repost:

"What is Science? We cannot define the word with the precision and concision with which we define Circle, or Equation, any more than we can so define Money, Government, Stone, Life. The idea, like these, and more than some of them, is too vastly complex and diversified. It embodies the epitome of man's intellectual development...

To most men, including all who are outside of the world of science, the term means a particular kind of knowledge.  Wherein lies the essential peculiarity of this knowledge?

Some thinkers agree with the ancient Greeks in making it consist in the Method of knowing, the manner in which the truth is laid hold on...

There are in science three fundamentally different kinds of reasoning…These three kinds of reasoning are Abduction, Induction, and Deduction. [—] All the ideas of science come to it by the way of Abduction.  Abduction consists in studying facts and devising a theory to explain them. Its only justification is that if we are ever to understand things at all, it must be in that way…” 

A demonstration: 

http://www.researchgate.net/post/What_is_the_best_example_of_Peircean_abduction

A good answer Jerry, but first we propose a hypothesis based on what we know so far, (inductive reasoning) and then devise ways to test the hypothesis. If the hypothesis passes many tests it can be elevated to a theory, but it is not a law, and can still abandoned if contrary data are found, and perhaps a modification of the theory, or a new hypothesis can be proposed, and the testing goes on. This is my, perhaps simplistic, understanding of the core of the scientific method,

Regards,

Keith   

Thank you for the conversation, Keith.  But what goes into choosing that hypothesis?  Peirce wrote extensively on this matter and formalized it in abduction:

“It may seem obvious that, before we can begin to verify a hypothesis, we must somehow “acquire” one. Yet, until Peirce began working on his theory of abduction, little thought had been given to the issue of HYPOTHESIS ACQUISITION and its everyday equivalent goal acquisition. Even today, most people seem satisfied with the idea that goals and hypotheses arise “somehow,” and that the primary purpose of scientific inquiry is to verify a hypothesis; and, of ordinary life, to achieve goals. The idea of a normative method by which hypotheses should be formed (abduction) belongs to Peirce.”

~Chiasson "Peirce and the Continuum of Means and Ends"

and more from Peirce:

“…what does it matter how the work of abduction is performed? It matters much, for the reason that it originates every proposition. It is true that, however carelessly the abduction is performed, the true hypothesis will get suggested at last. But the aid which a correct logic can afford to science consists in enabling that to be done at small expenditure of every kind which, at any rate, is bound to get done somehow. The whole service of logic to science, whatever the nature of its services to individuals may be, is of the nature of an economy.”

~ Peirce, CP 7.220n18

It seems to me that if we were to attempt spread of a curriculum for students about the philosophy of the scientific method, then decision makers should have consensus on which philosophy is correct, complete and relevant. 

Is there such a philosophy that makes that claim that is directly relevant to modern students other than pragmatism, by which I mean the logic of abduction (pragmaticism)? 

How many teachers know its complete structure?  How can we make them know it?  How can we show that it is correct and complete?  What are the criteria?  What is the best example?  What is the best worked example?

Peirce was very conscious of his own development as a philosopher. He constantly refers to what he read and by whom he was influenced. He continually returned to what he had written to annotate and correct his opinions. Therefore, the most reliable and fruitful source for an appreciation of what he is trying to do is Peirce's remarks about himself.

In a letter to William James, dated November 25, 1902, Peirce remarks that many philosophers who call themselves pragmatists "miss the very point of it," and he tells us why:

But I seem to myself to be the sole depositary at present of the completely developed system, which all hangs together and cannot receive any proper presentation in fragments. My own view in 1877 was crude. Even when I gave my Cambridge lectures I had not really got to the bottom of it or seen the unity of the whole thing. It was not until after that that I obtained the proof that logic must be founded on ethics, of which it is a higher development. Even then, I was for some time so stupid as not to see that ethics rests in the same manner on a foundation of esthetics..

~Potter, Charles S. Peirce on Norms & Ideals

Rationale

In the past century, school science has been dominated by the educational requirements of our future scientists….However, during the past two decades, the growing concern about the relationship between science and society has led to a concern to improve the quality of formal education about science – in short, to ask what kind of school science education is required for citizenship in a participatory democracy?…

Yet, part of the difficulty of determining what should be taught about science is the failure to agree on an acceptable account of science within the scientific community, or amongst philosophers and sociologists, let alone between the various communities. Hence, the lack of any consensus makes the task of defining that aspect of the formal science curriculum which might portray ‘ideas about science’ problematic for policy makers – the more so as there is, in contrast, a well-established consensus about the content aspect of science curricula.

This study…sought to determine the characteristics of scientific enquiry and those aspects of the nature of scientific knowledge that should form an essential component of the school science curriculum.

Methods

The study reported here sought to explore this issue by undertaking a Delphi study with a group consisting of 23 individuals drawn from 5 groups – scientists, philosophers, sociologists of science, science educators, and science teachers.

~ Osborne, Ratcliffe, Collins, Millar and Duschl, “What should we teach about science? A Delphi study

Yet, Peirce wasn’t invited as a panel member.  

So, pragmaticism or results of Delphi study? 

To what extent are issues the same; rather not organized as economically?

Thus, which is more correct, complete and relevant?

How many times will this conversation repeat in the future due to neglect of the Neglected Argument?

Jerry,

There is no concensus of most philosophical issues and what is science is a philosophical question that has many answers as all complex philosophical questions.  This should not be an obstable since we teached history and there too there is no universal consensus.  I do not even think that teaching a philosophy of science should be done early in the scientific formation. I believe into learning through doing. Peirce learned it this way anyway and it should be valid not only for the genius but for all of us.

Louis,

I’ve heard similar arguments by philosophizers in the past; that philosophical issues will have no consensus.  But in some situations, I’m left with the impression that philosophizers are not seeking solutions.  Rather, they enjoy the process of philosophizing because they think they’re good at it, it’s fun and also provides justification for investing in their favorite philosopher.  But a good answer will annul this feeling, so it’s not surprising that some will resist. 

Besides, what’s so hard to understand about presenting a defeasible argument, that is, the best argument for what the scientific method should consist in (i.e., if not this, which…and for what reason)?  Moreover, it’s clear from reading Peirce that he was convinced he found a satisficed solution about the structure of scientific inquiry.  Moreover, he agreed with you that one learns by doing, since the Neglected Argument is a demonstration of how to use that structure. 

“…the Neglected Argument, greatly cut down to bring it within the limits assigned to this article. Next should come the discussion of its logicality; but nothing readable at a sitting could possibly bring home to readers my full proof of the principal points of such an examination. I can only hope to make the residue of this paper a sort of table of contents, from which some may possibly guess what I have to say; or to lay down a series of plausible points through which the reader will have to construct the continuous line of reasoning for himself. In my own mind the proof is elaborated, and I am exerting my energies to getting it submitted to public censure.”  ~Peirce

Let me ask you, do YOU know this structure well enough to confidently state that he was wrong?  That is, do you, and by this I mean a common representative of a well-educated scientific inquirer, know of the history of Peirce’s thought?  How many of us do?  Yet, there are growing accounts that say that Peirce’s philosophy is spreading.  I imagine from this that his ideas are worth knowing. 

Finally, the process of finding a solution is extremely difficult because of the uncertainty but  “All truths are easy to understand once they are discovered; the point is to discover them."  ~Wiki

That is, once one finds a solution, talking about that solution becomes simpler and the structure of abduction is simple enough that it can be spread (a syllogism, just three lines!!).  The challenge is in learning how to use it and to use it well.  In fact, the following indicates that composing a simple structure was Peirce’s intention:

"[I intend] to make a philosophy like that of Aristotle, that is to say, to outline a theory so comprehensive that, for a long time to come, the entire work of human reason, in philosophy of every school and kind, in mathematics, in psychology, in the physical sciences, in history, in sociology, and in whatever other department there may be, shall appear as the filling up of its details. The first step toward this is to find simple concepts applicable to every subject [for example: one, two, three; sign, object, interpretant; chance, law, habit-taking or continuity”

I also agree that organizing a curriculum around his ideas are not something you present directly to a fifth grader.  Yet, the structure is something that can be understood well enough by teachers so as to facilitate presentation of content.  For example, why not use it in accordance with spiral curricular concepts?

The surprising fact (C scientific method) is observed.

But if (A “it consist in the Method of knowing, the manner in which the truth is laid hold on”, ie., abduction/deduction/induction) is true,

then C would be a matter of course.

Hence, there is reason to suspect that A is true.

The reasoning may appear circular but it is not.  It is hierarchical.  That is, think of the next situation.