Talk:Scientific method/Archive 13

dunno who started adding this. but I don't think it helps or is accurate. it was a major change and wasn't discuss. pls discuss and get agreement before major changes Mccready 15:29, 1 August 2006 (UTC)

see section above re: Removed material. i don't agree with its addition. This idea does not belong in the introduction. David D. (Talk) 17:53, 1 August 2006 (UTC)

Hi David. A valid equation in physics must be homogeneous. Homogeneity is the the quality of having all properties independent of the position. Scientific method assumes that the world is homogeneous. There is no example I can give about it, its an axiom. Scientists, in order for their inductive and deductive reasoning to stand , they are bound to believe that the natural world is homogeneous. They cannot deal with the possibility that the laws of physics could be dependant of space-time position, because in that case their reasoning turns irrational. Faaaa 09:42, 2 August 2006 (UTC)

I assume that I am talking to scientists, who are well known for their effort to be reasonable. My questions are.

1. Is scientific method capable to deal with a natural world that is not homogeneous. yes or no ?
2. If no, then why this is not mentioned into the article? Is it reasonable and scientific to hide the fact that we can understand and predict only in case the natural world is homogeneous ?

The reason we named our natural world homogeneous is because we want to be able to understand it and predict it. Due to our appetence to understand and predict, we repulsed the possibility that the natural world may not be homogeneous, although we have plenty of evidences of that possibility (see Physical paradoxes, black holes e.t.c.) Faaaa 09:42, 2 August 2006 (UTC)

As with a many words and phrases, homogeneous may not translate into a word in Korean which has the same connotations as it does in English. Yes, a physical principle or law that applies in London is also expected to apply in New York and Seoul (and even in Pyongyang). But it should not be necessary to explain this in the article on scientific method. This concept is currently dealt with in a number of other articles such as theory, Physical law, and commentary on the idea in Falsifiability#Physics. The article on scientific method is not limited to physics but must also accommodate many other disciplines where certain "universals" may in fact be location-dependent and must be defined accordingly, which is one of many reasons why the article is written the way it presently is (I'm not saying the article can't be improved). But "homogeneous" is definitely not the term used to express the concept of "universal" appicability of physical principles. ... Kenosis 13:01, 2 August 2006 (UTC)

That current physical theories, quantities do indeed exhibit symmetries, invariances, etc under transformations in space, time has in fact been observed and inferred. For example, conservation of mass was once thought to be the case. Once observations started breaking the conservation of mass, it's accuracy was put into question and, like many older laws, is now seen only true as some low-velocity limit (to use relativity as an example), or when mass-defect is negligible/irrelevant.

The fact that science would crumble if Reality(TM) were merely a jumble of uncorrelated, randomly changing rules and laws coupled with the fact that science is hugely successful all the way from the mundane to the esoteric lends scientific evidence that the universe *does* exhibit this kind of homogeneity you're alluding to. Basically, this homogeneity is tested, like every other aspect of science, with every single observation and experiment. It is not an assumption. No one is claiming that there aren't any exceptions to current theories, nor is anyone suggesting that all of Reality(TM) will be amenable to all-encompassing laws (though many are trying to formulate those laws). Rather, we can observe that it's tending that way.

To answer your question (1), of course it could. Being a bit loose with definitions, if there were an appartment somewhere in Nevada where a mass moving with velocity v actually had energy 10mv^2, then we would have to accept that moving masses have kinetic energy 0.5mv^2, except in Flat 24, Weirdness Drive, Las Vegas, NV, where it will have kinetic energy 10mv^2. It wouldn't be pretty, but it would be coped with. Tez 13:34, 2 August 2006 (UTC)

Tez, I sure as hell hope that's not an argument for the application of the word "homogeneous" to explain "universal generalizations" ;-) Thank you for a superlative explanation. ... Kenosis 15:20, 2 August 2006 (UTC)

Isn't the correct term symmetry? Stephen B Streater 09:14, 3 August 2006 (UTC).

And a symmetry is an invariant under a transformation. (This topic is not part of scientific method. It is discussed extensively elsewhere.) --Ancheta Wis 09:35, 3 August 2006 (UTC)

I agree. The Scientific Method doesn't make assumptions about the nature of laws of the universe. Many things are non-scientific, and these are not covered by science. Stephen B Streater 09:49, 3 August 2006 (UTC)

JA: There was a lot of discussion in the 19th century about the role of homogeneity and regularity assumptions in science. Whether homegeneity and regularity are same thing or not is another good question. Based on a fundamental theme out of Aristotle, J.S. Mill and others derived a few simple inferences about the "regularity of nature" and its salience in science. But the analysis of C.S. Peirce, as usual, showed that the real situation was not as simple as it had been supposed. The issues involved here go to the root of scientific inquiry, and so I will take them up under the more general heading below. Jon Awbrey 14:06, 3 August 2006 (UTC)

Some limitations are to be found in the discussion of Eddington's net: a scientist casts a net with a mesh of 2 inches. He catches no fish smaller than ... guess how big? --Ancheta Wis 18:29, 3 August 2006 (UTC)

JA: Being, I'm guessing, the only person or even quasi-person here with a degree in "Mathematical and Philosophical Method" (BA MPM, JMC, MSU, 1976), I feel quasi-qualified to say something on the nature of method, scientific method in particular.

JA: The first thing to know about any method is that it involves a voluntary limitation. So when you say that a method is limited this is just a tautology. It is the limit that makes it a method. The question is whether the limit that you are talking about is the one that is built into the method from the start, or perhaps some other limit that was not intended.

JA: When it comes to scientific method, the limit that defines the method was identified some time ago by Aristotle. I will comment on this after I finish my coffee. Jon Awbrey 12:16, 3 August 2006 (UTC)

JA: I have corrected a slight overstatement of the status of causality in science. I think that my qualifications are a fairer summary of the literature, and it's easy to come up with many citations from the likes of anybody from Peirce to Heisenberg if anybody feels the need to do so. Jon Awbrey 20:45, 4 August 2006 (UTC)

In my estimation, this is an apt correction you made in the callout box, which has my support. I appreciated your qualification of falsifiability which covers the appropriate base about Popper and modus tollens. ... Kenosis 21:13, 4 August 2006 (UTC)

I'm surprised so little is mentioned on the role of probability, in several areas:

1. Precision of measurement and importance of including it in calculations and results.

2. Weighing of the (estimated) probability of the truth of a hypothesis with the effort involved in resolving it further. Failing to do this well is one of the main hindrances to scientific progress. For example, there was a time when the vast majority of doctors chose not to spend 30 seconds washing their hands between patients because they assigned an unreasonably low probability to the validity of the microbe hypothesis, or were simply too lazy, or both. It seems a true scientist would at least be willing to try it for a few months to conduct his own experiment. We must be careful to consider ourselves part of the "enlightened age of enlightened thinking"--those doctors probably felt the same way.

3. Philosophy: nothing can be totally proven and nothing can be totally disproven (only seat-of-the-pants probabilities can be assigned) because the laws of a universe cannot be proven consistent from within that universe. Yes, we've even learned "logic" by observing our universe.

-QZ27, Aug 13, 2006

Thank you for your contribution. This is covered by the references to Bayesian inference. --Ancheta Wis 19:34, 13 August 2006 (UTC)

Just posting to say my first impression of a quick look through the article is a really good one, and I especially like the bit unorthodox way of merging the DNA examples into the article to exemplify the topics that could otherwise be a bit abstract. I haven't seen the method in much use in Wikipedia, but I think it works really well. :-) -- Northgrove 22:02, 23 August 2006 (UTC)

Thank you for your kind words. --Ancheta Wis 23:24, 23 August 2006 (UTC)

Kenosis, I advise you to revert your removal of observation from the callout box. I understand your concern, but I point out that the author(s) of the text have been careful to introduce observation as a 'facet' of scientific method and not as an initial step. This could be clarified without removing the text on observation.

By the way, in the article 'theory ladenness' is mentioned without reference to Popper. Isn't it Popper who speaks of 'falsifying hypotheses' rather than 'falsifying facts'. And isn't it Popper who first argued that observation can't be the first step (Yeah, I know Lakatos argued the same, but I expect he got there later). Anyhow I always thought theory ladenness was Popper's baby. --Chris 15:56, 6 September 2006 (UTC)

I agree. The box says "The scientific method involves the following basic facets". Observation is one of the basic facets of science present at every step. It is not the first step as the initial addition claimed, but it still needs to be in the box.--Roland Deschain 16:43, 6 September 2006 (UTC)

This particular callout box was largely derived from the text Research Methods in Psychology (don't recall which author, there are several such texts by this title) and is presented as one of several ways scientific method is delineated or summarized. Originally it was summarized as "description", "prediction", "control (where possible and appropriate)" and "understanding", along with the three-point addition about "time-order relationships", "covariation of events" and "identification of causes to the best achievable extent". Someone has since replaced "understanding" with "falsifiability" and attached to "falsifiability" the supplementary content about increasing confidence levels in the results of a hypothesis or theory as observations are confirmed or contradicted in accordance with the research. That's Wikipedia for ya'.

As I stated in the edit summary, observation runs throughout the entire process. Please read the already existing text-- although much of it is currently lacking in explicit citations, this article has had ongoing participation by experts in the subject, including several professional scientists. The article amply describes how observation fits into the method at every stage of the process. Two paragraphs down from the second callout box is a paragraph stating that observation is inevitably "theory laden", that is, driven by the concepts one intends to pursue at any given stage of observation. An entire section is devoted to abductive reasoning, in addition to several additional mentions of the concept. A mere reference to "observation" is meaningless here-- everyone observes, including scientists, non-scientists, junk-scientists and pseudoscientists alike. What the article describes is how, in summary, scientific method involves specific kinds of observation, the relevant analysis, further focused observation, adaptation of the focus of the research, further analysis and testing, documentation of all data so others can similarly direct their testing and analysis, etc. ... Kenosis 16:45, 6 September 2006 (UTC)

As you said, observation runs throughout the entire process: therefore it should be mentioned right away in a prominent place. Especially as it is in the rest of the article, it should be included in the box. "Everybody observes" is a true statement, but that doesn't change the argument at all. Critical observation is still part of the scientific method, even though my mother uses it when she does pancakes and my father uses it when fixing the car. A concept should not be excluded just because it is present in other ways of life. Creationist observe that life is complex, yet they are not scientific as the rest of their method is radically different. Observation is such a huge and varied field in science that it needs a mention but cannot be fully exposed in this article as it is such a huge and varied topic. Here are a couple of websites that reflect the predominance of observation , , , (these are all quick and dirty google searches, and it was that easy finding supporting evidence) --Roland Deschain 17:03, 6 September 2006 (UTC)

Point taken, and accepted. Thank you kindly. ... Kenosis 04:47, 7 September 2006 (UTC)

I would like to challenge the statement that we have documented the standard scientific method. From the 'Elements...' section I quote,

"While this schema is currently accepted as standard scientific method..."

If we mean 'standard' in the sense that a research body has seen fit to institutionalize this version of the method, then we should cite the standard. I have inserted the {{fact}} template. On the other hand, if we merely mean that it is standard in the sense that it is typical of how the scientific method is schematized, then it would be more accutare to say just that. --Chris 19:47, 6 September 2006 (UTC)

Chris, I agree that the statement is too strong. What about something like "While this schema may be typical of scientific method, ...". --Ancheta Wis 20:58, 6 September 2006 (UTC)

I've narrowed the statement and provided one citation for now. I should point out that this section was written with a cautious eye on the perspectives of Popper, Kuhn, Thagard, Lacatos, among others, and it generally applies irrespective of paradigm shifts, ant-colony behavior among scientists, jealous covetry of the next big breakthrough, etc. And nothing escapes Feyerabend. As Hugh Gauch put it in Scientific Method in Practice (2003), commenting on Feyerabend's sociological perspective: "Such critiques are unfamiliar to most scientists, although some may have heard a few distant shots from the so-called science wars. Scientists typically find those objections either silly or aggravating, so rather few engage such controversies or bother to contribute in a sophisticated and influential manner. But in the humanities, those deep critiques of rationality are currently quite influential. Anyway, by that reckoning, Figure 1.1 [which diagrams scientific method with a core set of principles and many specialized branches representing particular fields] should show blank paper."

In addition to the two callout boxes outlining aspects of scientific method generally, the Wikipedia article outlines a seven point description of the hypothesis/testing cycle, as follows:

1. Define the question
2. Gather information and resources
3. Form hypothesis
4. Perform experiment and collect data
5. Analyze data
6. Interpret data and draw conclusions that serve as a starting point for new hypotheses
7. Publish results

If this isn't standard, I don't know what is. ... Kenosis 22:08, 6 September 2006 (UTC)

Kenosis, one difficulty is the level of abstraction at any iteration of scientific method. As an example from the craft of movie production, one director might say "move the field of view of the camera to coordinate (x y)", a low-level command which a robot could execute; another director might say "float the scene", a fairly abstract command which an expert cameraman could then implement. In effect the director has delegated something to the cameraman instead of forcing a situation of micromanagement. (In the article, then, the various versions of the callout boxes basically allow different populations to uphold scientific method with varying degrees of understanding. This is a good thing.) To bring a camera into perfect mathematical focus is not possible. There is always the circle of confusion in a situation where physical optics holds and the robot may have obeyed a command which made no physical difference to the movie, whereas the cameraman just may have 'gone with the flow'. My difficulty with 'standard' is the connotation of 'required'; that is what 'typical' might have avoided. But if we are attempting something else in the article, that might be noted. --Ancheta Wis 22:32, 6 September 2006 (UTC)

Absolutely Ancheta, yes. This is, of course, very much what the general concept of abduction speaks to, and what Peirce was grappling with. Maybe let's go get high sometime and just observe things for awhile; stream of consciousness; maybe give the Feyerabend advocates another validation of their position. I'll go change that to "typical" now, if someone else hasn't already.... Kenosis 02:11, 7 September 2006 (UTC)

Neelsmalan 17:58, 9 May 2007 (UTC)

What is science? Let us simplify and popularize our understanding of the concept 'science': Science is knowledge which was tested against a criterion.

Explanation:

1) Untested knowledge remains knowledge. 'That car is traveling fast' is my perception of the speed of a car. My perception may be 'true' of 'false' depending on, amongst other things, what is accepted as 'true' (e.g. faster that 80 kmh) as what as false.

2) Tested knowledge is scientific knowledge (according to my simple definition), but scientific knowledge is not necessarily true or valid. It is just as valid as the criterion on which it rests. Scientific knowledge is 'honest' or accountable knowledge, because you put all your cards on the table. You declare your criterion(s)/ assumption(s). 'That car is traveling fast' may rest on my assumption that the speed testing apparatus I may be using is giving a valid reading, and, secondly, that traveling above 80 kmp is accepted as fast.

3) Conclusion: We may have 'good' science as well as 'bad' science, but we should keep in mind that 'bad' science' is still science. What makes bad science bad is

 a) the invalid testing of a valid criterion; 
 b) the valid testing of an invalid criterion; or 
 c) the invalid testing of an invalid criterion. 

Am I right?

Well, the criterion has historically been observation. If the criterion were mythical or unobservable or a matter of faith, that would not qualify as science. As always, the truth matters to scientific method. --Ancheta Wis 18:05, 9 May 2007 (UTC)

Stanislaw Ulam in his autobiography Adventures of a Mathematician recommended Poincaré's Science and Hypothesis to all. A link to an e-print of this classic is listed in the bib. of the article. --Ancheta Wis 21:17, 6 September 2006 (UTC) But the font of the e-print is pretty strong. Maybe it was meant to be read to a class of students in a classroom.

I've removed the following sentence from the first paragraph of the intro, for further consideration as to its accuracy and relevance in the intro. Perhaps it was an attempt to explain the concept of a theory?, which is already introduced elsewhere. Also, there were previously some concern about whether the statement "All such evidence is collectively called scientific evidence" belonged in that paragraph, so I've removed that pending a clarification on whether it's needed in the lead paragraph. ... Kenosis 16:54, 12 September 2006 (UTC)

• All such evidence is collectively called scientific evidence and is used as a means of structural cohesion to that new information. This new information or scientific evidence, becomes a building block to that structure or particular body of knowledge. ... 16:54, 12 September 2006 (UTC)

Good call I think. --Chris 22:13, 12 September 2006 (UTC)

I've removed the large section on Peirce's theory. There does not appear to be a good reason to single him out for a special place in this article. Banno 21:32, 17 September 2006 (UTC)

In my interpretation www.wikinfo.org has a much more clear and succinct introduction for "scientific method":

"The scientific method is a multi-step continuous and iterative process to expand knowledge by observation of phenomenon, formulation of hypotheses, experimentation, verification and unification. The details of the process might vary from one field of inquiry to another but share the same core abstract principles. In addition to the above, other principles include requiring research to be objective, logical, open to refinement and accepting of external criticism."

What do you think? I think the current version of this article is needlessly wordy and has confusing sentence construction and its introduction misses the abstract essence of the scientific method. zen apprentice T 18:53, 7 October 2006 (UTC)

Here are a couple of defenitions from the internet:

• "The scientific method is the process by which scientists, collectively and over time, endeavor to construct an accurate (that is, reliable, consistent and non-arbitrary) representation of the world."
• "The principles and empirical processes of discovery and demonstration considered characteristic of or necessary for scientific investigation, generally involving the observation of phenomena, the formulation of a hypothesis concerning the phenomena, experimentation to demonstrate the truth or falseness of the hypothesis, and a conclusion that validates or modifies the hypothesis."
• "The scientific method is the experimental testing of a hypothesis formulated after the systematic, objective collection of data."
• "The scientific method is a process used to systematically investigate observations, solve problems, and test hypotheses."

The current into, I totally agree, has to be rewritten. It has no citations, which shouldn't be really hard to get for such a prevelant topic.--Roland Deschain 19:38, 7 October 2006 (UTC)

Added a citation for Isaac Newton's Principia to give proper credit to the central figure of the scientific revolution. --Ancheta Wis 21:54, 7 October 2006 (UTC)

Just so you know, the 4 rules of reasoning can be read at History of scientific method. 21:58, 7 October 2006 (UTC)

Why do you want me to know them? One person's list of rules is incomplete compared to a clear and succinct summary that points toward the abstract essence of the subject. zen apprentice T 05:31, 10 October 2006 (UTC)

I agree that the intro could be better. I'm less certain that the article should be reduced to an essentialist vision of scientific method. However, such a view might make a worthwhile addition to the article if citations can be provided. Chris 11:38, 10 October 2006 (UTC)

Please expand on what you mean by "essentialist vision"? I mostly mean "succinct summary" using simple sentence construction and unambiguous language. In your interpretation how is my proposed summary introduction above at odds with the rest of the article or the subject such that citations would be required? zen apprentice T 16:31, 10 October 2006 (UTC)

What I mean by "essentialist" is the view that although scientific method appears to be a diversity of things, it is nevertheless a cohesive idea to the extent that a common thread, or essence as you put it, maintains throughout. What's more, the essentialist would consider this essence of highest import. In opposition to this, we might think a methodology is important in that it differs from other methodologies. So, for example, Popper's method shares some characteristics with Bacon's (observation, experiment, emphasis on falsifying instances) but the similarities, although perhaps interesting, do not highlight what is important in either method. I asked for citations since I thought the essentialist view unusual, not to mention unlikely. I reserve this if you meant something different. --Chris 21:02, 10 October 2006 (UTC)

I am not saying my proposed hopefully succinct introduction above is perfect, just that it is better than the current hodgepodge needlessly wordy and unclear introduction. I would say the scientific method is not a cohesive idea but an abstract concept or a non-concrete methodology, the only really "essential" part separate from observation, hypothesizing, experimentation and verification is that it is an iterative process which does not produce an absolute conclusion [much confusion exists on this point]. I still find it odd that you are asking me for citations for something that is mostly a condensed/summary version of the current introduction, how exactly is what you are labelling the "essentialist" view "unusual" or "unlikely" compared with the current introduction? zen apprentice T 01:08, 11 October 2006 (UTC)

Your criticism might be more helpful if you explain in detail, word for word if necessary, what specifically you see as wordy or unclear. Since I should practice what I preach, it is the text "core abstract principles" I object to in the introduction you propose for reasons already mentioned. In the current introduction we speak of "identifiable features that distinguish scientific inquiry" which leaves open, to my mind, the question of whether all those features are always present. The second paragraph of the introduction talks of repetition rather than iteration. There’s not much to choose between 'repetition' and 'iteration', but I prefer the former since iteration suggests to me something almost mechanical.

Just to explain one point, scientific method can be considered a cohesive idea without recourse to essentialism. Wittgenstein's family of usages suggests one way and I imagine there are others. Maybe you could clarify your last question. Are you asking why I believe the essentialist view unlikely? Or are you asking to what extent I see the current intro meets or diverges from an essentialist conception of scientific method? --Chris 05:59, 11 October 2006 (UTC)

The phrase "body of techniques" portrays too concrete a concept given that specific rules might vary from one field of inquiry to another. Only "core [abstract] principles" or at least less concrete "rules of thumb" can be common among something that is variable...? The first sentence of the current introduction reads like a definition which I consider to be incomplete because a key point, iteration/repetition, is not mentioned until later and mentioned insufficiently explicitly. I chose "iterative process" to hopefully convey that the process repeats forever or at least does not produce an absolute conclusion in and of itself [much confusion exists on this point]. The word "repetition" by itself to me means repeating the exact same concrete process over and over again which I consider to be incomplete/misleading, it's better to note the scientific method's repetitive nature more explicitly but also less specifically, something to the effect of: iteration is inherent in the scientific method. "Recursive" might actually work better than either "repetition" or "iterative" now that I think about it.

My primary issues with the current introduction is its phraseology, it has little passion and is difficult to parse and comprehend and seems written by committee (all over the place). Though, the current introduction is not as bad as I first thought after reading it a few more times, but I repeat it is difficult to parse, especially for someone with apparently poor or slow reading comprehension (me). Other issues with the current introduction: the word "taint" is better than the word "bias" in my interpretation, everyone has some sort of personal bias, "taint" in this context applies unambiguously to research/results, it's more important to note that research might be tainted (which indicates the research is potentially worthless). I asked why you labeled the above proposed definition the "essentialist view" because I still have no idea what that label means or how it is relevant. Now that I understand your other counter criticisms more I will try to come up with a hopefully better new compromise/superset introduction proposal soon. zen apprentice T 20:19, 11 October 2006 (UTC)

I know that we never seem to get further than the introduction, but here's a thing. We say in the second sentence, "It is based on observable, empirical, measurable evidence, and subject to laws of reasoning." In the citation, Newton uses the work 'rules' rather than 'laws'. I notice that this is normally the case as in Descartes' Rules for the Direction of the Mind, Popper's "rule of thumb". Bacon, despite being a lawyer, calls his New Organon, "true directions concerning the interpretation of nature". I suggest we change the sentence to read "rules of reasoning". --Chris 21:34, 10 October 2006 (UTC)

I concur. --Ancheta Wis 21:51, 10 October 2006 (UTC)

The scientific method is a logical, multiple step, and recursive process to expand and improve knowledge using general principles and procedures commonly including the observation of phenomenon, formulation of hypotheses, investigation, experimentation, verification and unification. Specific procedures often vary between different fields of scientific inquiry. Other principles include requiring research to be open to both refinement and independent scrutiny.

Comments welcome. zen apprentice T 01:00, 12 October 2006 (UTC)

Actually, guessing or abduction is as much a part of it as logic; see also hypothesis. Observation need not be the first step; a question was the basis for relativity; (What would I see if I were riding on a beam of light? --Einstein). And you didn't mention prediction, which is the logical deduction from the guessing stage. That's one reason for the length of the article; scientific method is like an onion. One person's guess can be another person's starting point, which is the reason for the Heisenberg quote deep in the article. All of that is pretty tough to jam into an introductory paragraph.

I listed the common principles and procedures in sentence form, rather than as a bullet pointed list, to convey precisely that "step" order is variable. The process of hypothesizing creates predictions. A layered onion still can have a succinct introduction that gives a general overview rather than being all over the place. Note: when I use the word succinct I mean clarity not necessarily brevity. zen apprentice T 00:50, 13 October 2006 (UTC)

If you are looking for a quick intro, read the material in the yellow boxes. There are 3 versions of the box in the article. But they are not equivalent; they are like layers of an onion. --Ancheta Wis 03:12, 12 October 2006 (UTC)

The word hypothesis is within the proposed introduction above, the process of hypothesizing creates predictions. Feel free to update the proposal above with anything you consider to be missing. Do you like the current introduction? Do you agree or disagree whether the current introduction is phrased unclearly? zen apprentice T 07:30, 12 October 2006 (UTC)

Once you start saying 'A' includes 'B' then you cannot justify disliking the current intro, which includes all of the words in the successive paragraphs in the entire article. And if you admit that truth is layered then the current intro suffices, even in the face of its stylistic deficiencies. The archives show that the proposed intro had its roots in previous versions of Wikipedia's article. "What I cannot create, I do not understand": even Feynman had this urge. That does not justify attempting to nail down the entire article in the intro paragraph. The intro is designed to lead the reader onward. It ought not give the reader a false impression of comprehension or comprehensiveness.

The steps of scientific method are currently our best approach to understanding, and they are rooted in our history, but it has taken the best minds of our civilization to come up with the steps. Most people can only hope to engage in only one or two of the steps at a time. Scientists work in communities and specialize in one or two of the steps themselves; since unusual people like Newton or Galileo showed the breadth and depth of comprehension to be able to engage in all the steps (see the yellow boxes), then it may help to characterize what one or two of them has done, before proposing a succinct intro paragraph. That of course limits the scope of generality of any proposed intro paragraph. --Ancheta Wis 10:18, 12 October 2006 (UTC)

The current introduction is the result of a great deal of thought and discussion by at least seven or eight editors. It takes into account many aspects of this broad subject, including but not limited to such issues as the demarcation problem and use of scientific method by the social sciences. ... Kenosis 15:29, 12 October 2006 (UTC)

Responding to everyone above, in my interpretation my proposed new introduction at the top of this section is better than the current introduction but since allegedly "seven" or "eight" user accounts disagree I suppose I will continue to discuss and work toward a larger consensus with more and more people. I would not characterize the current introduction's deficiencies as being merely "stylistic" but instead I interpret them to be fundamental deficiencies. Though I am not saying necessarily throw away the current introduction completely, just that the top proposal is a good first introductory paragraph (there can be more than one introductory paragraph, feel free to add on to it if you think it is incomplete). The way the current intro is written does not lead the reader onward in my interpretation and seems almost as if it was fabricated for the purpose of befuddling readers subtly. Please explain how you think the proposed new top introduction gives readers a "false impression of comprehension of comprehesiveness"? If an introduction is just a tease readers are more likely to keep reading. Note: I added the word "investigation" to the proposal at the top of this section. zen apprentice T 00:44, 13 October 2006 (UTC)

Actually, for the moment I'd like to amend my statement so as to make that "allegedly 'six' or 'seven' user accounts", per the immediately above, because one of the "alleged" users might have been a sockpuppet, and an additional two dozen or more might be derivative sockpuppets, but there's no way to WP:VER this at the moment. ... Kenosis 04:45, 13 October 2006 (UTC)

John Rennie wrote:

• A central tenet of modern science is methodological naturalism: it seeks to explain the universe purely in terms of observed or testable natural mechanisms.

Is there anyone who has ever sought to generalize science beyond the self-imposed limitation of naturalism? That is, to extend the range of science beyond the physical, material world of nature and to explore the inner realm of thought or the spiritual world (afterlife) spoken of by religions? --Uncle Ed 20:40, 13 October 2006 (UTC)

Here is the full quote:

"Creation science" is a contradiction in terms. A central tenet of modern science is methodological naturalism--it seeks to explain the universe purely in terms of observed or testable natural mechanisms. Thus, physics describes the atomic nucleus with specific concepts governing matter and energy, and it tests those descriptions experimentally. Physicists introduce new particles, such as quarks, to flesh out their theories only when data show that the previous descriptions cannot adequately explain observed phenomena. The new particles do not have arbitrary properties, moreover--their definitions are tightly constrained, because the new particles must fit within the existing framework of physics.

In contrast, intelligent-design theorists invoke shadowy entities that conveniently have whatever unconstrained abilities are needed to solve the mystery at hand. Rather than expanding scientific inquiry, such answers shut it down. (How does one disprove the existence of omnipotent intelligences?)

So the quote applies to the creationism movement in the U.S. I'm just trying to put the quote in context.

Thought is very much a scientific endeavor. Due to it's complexity (the major obstacle being a agreed upon definition of thought), most of the work is done in psychology, but the molecular basis is being slowly worked out as well. The most important recent advance is the human/chimp genomes and large scale micro array studies that allow scientists to pinpoint differences in the genomes as the pertain to brain function and development. Be prepared for a lot of scientific research on that topic in the coming decade. Afterlife cannot be investigated using the scientific method.

I also think the phrase "self-imposed limitation of naturalism" is wrong. Nobody, to my knowledge, has found a way to investigate beyond the observable natural phenomena. So it's not a self imposed limitation, but rather a limitation imposed upon science by the reality we live in.

The above quote should definitely be in the article. It gives a nice parameter of what science can and cannot approach. Most importantly, this article should discuss why this principle must be a part of science (the historical, philosophical, and (of course) scientific reasons). --Roland Deschain 01:53, 14 October 2006 (UTC)

1) One never proves the hypothesis to be true--they fail to disprove it.

2)What about mentioning the null hypothesis. In some circumstances the null hypothesis is tested to prove good research methods/techniques.

I just hacked away at the intro. We really need to keep this as simple as possible. I have tried to rewrite it in a way that is more crisp and simple. I hope this is OK. Comments welcomed. David D. (Talk) 16:02, 20 October 2006 (UTC)

Paragraph two contains the text,

"Any hypothesis that cannot be subjected to a test is not considered to be scientific."

For sure, we need to say something about testability. Testability has been a feature of scientific method, in various guises, since Bacon. I would prefer to avoid the suggestion that it is the determining factor in deciding whether an argument is scientific. If we could be certain of this then we would have a solution to the demarcation problem and I don’t think we have. In any case, it is a tangential philosophical issue and it would be good if we could keep every sentence of the introduction focused on method. Here are a couple of examples how we could do that:

"This method is efficacious to the extent that experimental results are able to unambiguously affirm or refute the hypothesis investigated."

Or perhaps,

"The efficacy of this method is dependent on our ability to unambiguously affirm or refute hypotheses using experiment."

Both of these sentences relate the importance of testability to scientific method without going so far as to define science in terms of testability.

--Chris 22:07, 20 October 2006 (UTC)

Another point, from para. 2,

"Theories that encompass whole domains of inquiry serve to bind more specific hypotheses together into a logically coherent structure."

What do we mean by "logically coherent"? Possibly we mean that the hypotheses cohere in a structure according to some form of logic. Wouldn't it be easier to say "structurally coherent"? I'll change this to read "structurally coherent system of knowledge".

--Chris 22:31, 20 October 2006 (UTC)

By-the-way, thanks David for your edits, I think on the whole they were good. --Chris 22:44, 20 October 2006 (UTC)

One more thing, this time from the last paragraph,

"Another facet shared by the various fields of scientific inquiry is that the process must be objective..."

I find this sentence confusing. Yes scientists strive for objectivity. Are we saying (or in this case implying) that scientific method is one of the means by which scientists attain objectivity? That is perhaps what we should be saying. The sentence could just as easily read that the objectivity in science must transcend scientific method. That is to say, we must avoid bias even when we employ the methods of science. This might well be true; double-blind tests spring to mind. I'm tempted to say that this second interpretation should be dealt with in the body of the article rather than in the introduction. For now I propose the replacement text,

"Scientists strive for objectivity in their results, and it is believed that scientific method helps them achieve this. That is not to imply that scientific method is infallible in this respect. Scientists employ a peer review process and double blind trials among other methods to further minimize the effect of their own personal bias."

--Chris 23:28, 20 October 2006 (UTC)

The introduction to this article had been quite stable for about the past five months or so. Recently it would appear that a few modifications were made bit by bit. Then some major changes were made today. I've returned it to the earlier consensused form, and adapted the sentence about theories encompassing wider domains of inquiry in keeping with the obvious quirk noticed by both David D. and ChrisSteinbach. Given the breadth of the proposed changes, I do not see adequate indication of good cause to override the earlier consensus about the introduction which has remained in place over these months. ... Kenosis 00:34, 21 October 2006 (UTC)

I agree. I should have waited for some agreement before committing my changes. David: I guess likewise you will have to argue your points one by one if you want to see your changes remain in the article. --Chris 17:47, 21 October 2006 (UTC)

Without David's changes I think only my last criticism is still relevant where I suggest the replacement text "Scientists strive for objectivity..." (See above). --Chris 18:04, 21 October 2006 (UTC)

Personally, I'm not opposed to cautiously rephrasing this sentence in keeping with the earlier intent. It is stated as an imperative ("must be objective..."). To say that scientists do strive for objectivity is perhaps not giving adequate credit to junk science and the like. To say that science somehow makes them be objective or helps them be objective opens up a conceptual can of worms, I think. We actually had arguments with a couple of people before over such issues as insistence on presenting Paul Feyerabend's viewpoint ("anything goes" in scientific method), and other such contra views (Feyerabend remains, but not in the introduction). But if there's a better way of saying it, I'm all for it, on a cautious point-by-point basis. Other than myself, Banno and Ancheta Wis were involved in this article before, and perhaps several others that haven't recently made their thoughts visible on the talk page. But I'd support a clearer way of saying it, understanding that we may need to account to those and perhaps other editors as well.

Having said that, I think that sentence should be simple and straightforward, avoiding qualifications such as "that is not to imply..." or other attempts to get excessively detailed about it. Underneath this assertion in the introduction, however it is phrased, are issues such as countering subject and observer bias (double blind studies with human subjects, but other methods must be used to minimize or eliminate other forms of observer bias), the Hawthorne effect (if it in fact exists), fudging of data (called fraud but under the same rubrick of non-objectivity), junk science (not quite fraud, but involving experimental designs and interpretations designed to make one's backers willing to hire you again), etc. Whatever it is that is said should be said very quickly and move right on to quickly touching the next necessary basic introductory point. ... Kenosis 21:12, 21 October 2006 (UTC)

First I want to thank Banno for his recent edits which I believe were an improvement. It's only a shame that the demarcation section had to go, although I understand why it had to. A lot of work is required on the demarcation article itself. Once things start to shape up there, maybe we can re-introduce the section. --Chris 20:53, 21 October 2006 (UTC)

I've just reverted an anon user's contrubution from the intro, specifically one which replaced the longstanding second paragraph of the lead (http://en.wikipedia.org/w/index.php?title=Scientific_method&diff=84756426&oldid=84514984 implemented here). Because it appears to be an explanation that could be of potential future value in the future, I'm placing it immediately below for futher reference. The removed material is as follows: ... Kenosis 04:21, 31 October 2006 (UTC)

• The scientific method follows a regular sequence of describing and explaining observations of nature. First there are observations that are generated by descriptive studies. For example, one might ask how many trees are found in a tract of woodland. The next step involves explaining why these observations were found. This causal question leads to the generation and testing of hypotheses by way of well-designed and controlled tests. If the hypothesis is true (hypothetically speaking) and the test were completed, then a prediction can be made. Real tests generate actual results that can be compared to the predictions. If they match, the hypothesis gains support and eventually may be accepted as likely to be true. ... 04:21, 31 October 2006 (UTC)

I'm surprised that this is not a featured article yet. What's the holdup? Sr13 07:27, 3 November 2006 (UTC)

These ideas were skipped over by Isaac Newton with, "I do not define time, space, place and motion, as being well known to all."

This is not right. He doesn't skip over them. Newton is saying that he is not defining what is popularly understood as these things and immediately goes on to define each of them in detail in absolute and relative terms. The text is at http://pratt.edu/~arch543p/readings/Newton.html

-Pepper 150.203.227.130 05:41, 13 November 2006 (UTC)

I made some changes to the first 3 paras. I find that in most of this,that too much the "results of" and/or "the management of" science, it's experiments and it' "useful paradigms and expectations" (e.g. that hypotheses will get broader, more accurate, etc., or that they must be made by the "rules of logic" which themselves have yet to have a fully acceptable foundation, as is the same with math, that measurement is indirect "obervations usually based on a theory", etc., etc., is all only an observation that hypotheses agglomerate into theories, that these so very often are expressed in math terms, that so much of physics can be based on Lagrangians, etc., etc. I vote to keep it to the bare 3 steps. Any more is as yet untested, unhypothesized "observations" on our proceedings to date with the three steps. We may call much of it meta scientific method or emerging philosophy and separate, but not all that is put forth here is really needed, or I believe really is in sci meth. 129.24.141.64 23:32, 25 November 2006 LekLiberty

I changed it back to the earlier approach, arrived at by consensus of a number of editors in the Spring of 2006. The consensus is not inflexible, but should be discussed interactively before making any major changes to the approach. ... Kenosis 03:01, 26 November 2006 (UTC)

Experiment is part of scientific method. That is the reason I reverted the latest contribution. Although I agree with an edit which champions scholarship and research, there is a problem with not designing research with an experiment in mind. The issue is that the concepts in play need to be out of the hands of the researcher. That is the role of the experiment in the protocol of the method. Otherwise, it is too easy to play games with the steps and to confuse what is known with what is unknown. When an experiment is designed, the outcome is unknown; the researcher may have hopes, but no control over the outcome. That is what keeps the method free of games. --Ancheta Wis 02:47, 8 December 2006 (UTC)

Experiment is part of the methods used by experimental sciences, not by science in general. Science in general uses something like "real-world corrective feedback" of which experiment can often be a part. Perhaps physicists would like to define "science" as meaning "experimental sciences such as physics." Such things are disingenuous. Paleontology is not an experimental science. Neither is Astronomy. Both are sciences, since they make and break theories by feedback from the real world. But there are more forms of feedback besides performing an active experiment with control of variables, etc. --Wjbeaty 03:21, 21 December 2006 (UTC)

I did not know that laughter (hahahaha), which I have never done before, explained the scientific method. Hey everyone! Laughter is the best way to solve and conduct an experiment! 67.86.24.40 00:13, 11 December 2006 (UTC)

To the anon. contributors: When we update the article, we typically give rationales. So if you have something to contribute, please say it here first. --Ancheta Wis 23:42, 11 December 2006 (UTC)

1. Experiment is part of "Observation" and/of of "Testing of Hypotheses" which is part of SM. Experimentless observation/testing is possible and allowed, required some might say when a discipline is very new, cuz at that time there is not enuf "known" to even think of hypothesizing or testing! So Experimentation is not the top level part of the SM definition; Observation and/or Testing is and they includes Experimentation as subsets. 2. The problem mentioned is better stated "there is a problem with desining experiments without a Hypothesis in mind". And research is a synonym for either SM or literature review/search! 3. The research is necessarily in the hands of the researchers, with errors, ignorance and even game playing if "so unethical". Replication of any of the three steps of SM I suggest is it's definition, Observation, Hypothesis and Test, catches all these: errors, ignorance and even game playing and unethical practices. LekLibertyLekLiberty

Well, if anyone is able to arrive at a definition for "The Scientific Method" which the experts can agree upon, then finally we also can define the word "science," and the long standing demarcation problem is solved. But it's far more likely that these problems will attract false "solutions" which try to make the problem appear simpler than it actually is. --Wjbeaty 03:10, 21 December 2006 (UTC)

To Wjbeaty and anyone else out there:---

Your comments and suggestions appear to me to be an infinite regression:

How do you define, and get a list of cooperating members for us to querry, experts in science, without a definition of science? And how do you get them without a definition, and get a list of cooperating cooperating memebers for us to query, of experts? This progression is essentially an infinite regression and won't terminate.

Agree, getting a definition of SM gets us to a definition of science. Still I submit the majority of practitioners (experts if you must) will ultimately agree that the SM is only the Three Steps of Observe, Hypothesize, and Test. All other suggested components are essenially only management and dissemination and "commercialization", I submit!

There will be discussion, because, today there are few Renaissance science experts to get at due to extreme specialization. And there will be a need to distinguish the active practice of science, the verb, and the body of knowledge that presently considered true (which only means it has not yet been disproved), the noun(s).

I'd like to see my definition presented to your experts, etc., cuz I think most all of the stuff I've read misses the mark. And in so doing, ya'll are allowong the very distortions you speak of to exist and procreate. And, some of that will get to less than expert in science funders who will then waste and abuse.

I don't want the job of "expert", but neither can I claim ignorance and unfamiliarity. FYI, and not to be bragging, or one upsmanshipping in any way, I've got 30+ years in RDT&E, Academia, Industry, Gov't Scientific Advisor positions, and BA, BS, MS, PH.D., post doc, and secondary ed theough grad school teaching on all this. I really fear for it being "gotten wrong".

Please advise me how I can get "reviewed", "critiqued", etc.---LekLiberty 1/4/07

There are at least 4 steps, at the base: Observe, Hypothesize, Deduce (i.e. Predict) and Test. You have to have an unknown that you are hypothesing about. No fair predicting a 'known', unless you are building a case for your 'unknown', and are trying to gain confidence in your 'hypothesis'/'prediction'. 'Observe' may not apply; in that case 'Imagine' or 'Question' or 'Inquiry' can be the placeholders. But you must genuinely not know what will happen, and be willing to take the risk to your reputation should your prediction fail. Good luck. --Ancheta Wis 00:03, 6 January 2007 (UTC)

I believe your four steps are actually three, as is practiced. Your last two, are ususally combined under the term of Test. The idea is that one is testing a Hypothesis, and to do so one must make a (new as you state) prediction (that has not already been covered/tested) and then test that hypothesis with either an experiment or an observation. It can be also that the observation is an "old one", catalogued and known as an observation, but not one "explained" by a hypothesis/theory" that is "new".198.175.175.107 18:33, 11 January 2007 (UTC)LekLiberty

It's a mental thing. Call the steps C H P T. When something is unknown, no one knows, including you. But if you understand H well enough to predict something logically, then you can marshal physical resources for the P step. i.e. lift a finger, whereas you can just imagine H in your head. Now once you have the resources to perform P infront of others, including your physical statement in front of others, then you have just put your reputation on the line. Remember, T has not been performed yet. Then you have an expectation (P) and a test T all lined up. You perform the steps, open up the box, pull out the result, in front of others. Either you have predicted correctly or your H is wrong. And as the article states, even if your T is as expected, H has just not been disproven yet. Good luck. --Ancheta Wis 22:56, 11 January 2007 (UTC)

This article seems very theoretical: is there data showing to what extent the scientific method is actually used by actual, practicing scientists or by non professional scientists ?

Do you really mean theoretical or are you meaning abstract and still not understandable? It's an important distinction for those of us who are trying to get it all right. The street English expressions along the lines of "the thoretical" verses "the practical" have no use in this effort and may be part of where you are coming from. If so, don't worry, you have much company, and if ever we converge to an agreed definition of the "Scintific Method" and "Science", hopefully it will become more clear.

I mean theoretical in the sense that it is not clear if the SM, as described, is actually used by practicing scientists (I assume that we could find people whom most of us would agree are scientists).

If there is no DATA to show that scientists use the SM described in the article, then the SM is in and of itself, non scientific - not quite sure how to deal with the sort of recursive questions that this raises. —The preceding unsigned comment was added by 24.60.137.141 (talk) 02:05, 6 January 2007 (UTC).

See Galileo(1638),Two New Sciences for the example upon which scientific method is based. --Ancheta Wis 05:46, 6 January 2007 (UTC)

You are not seriously proposing that a single scientists work from the 1600s is "data" which validates how the majority of all scientists who ever lived (those in the latter part of the 20th century) actually work ?? let me clarify what "data" is" something that shows how most scientists, most of the time, work. I'm sure all scientists use the "sm" as protrayed in the article, at least once. —The preceding unsigned comment was added by 24.60.137.141 (talk) 15:12, 6 January 2007 (UTC).

Someone had to start the huge stream of knowledge. A trickle has broadened into a torrent; Galileo is that start, unless you care to dispute that. The initial data aside, the encyclopedia shows that the steps are parsimonious. If you throw out one of the pieces, such as 'destroy a scientific community', then the method fails, as shown clearly in the history of science. It should be clear that the method is fundamental to progress. See, for example Needham's Grand Question. Awaiting your contribution, --Ancheta Wis 16:30, 6 January 2007 (UTC)

You may enjoy historiography of science which collects more critiques of science etc. --Ancheta Wis 16:35, 6 January 2007 (UTC)

Answers for some of what you ask otherwise, follows: Theory is another name for a hypothesis or the collection of hypotheses in some domain of discussion, in the scientific method. Theories are tested by experiments, observations, and predictions of the theory that are tested in the same manner. Data is that which is observed. It is the result of the second step of the scientific method of Observe, Hypothesize, Test which I am advocating in some of the above. For an article along the lines you request I suggest: Lightman, Alan, "Moments of Truth", page 36, New Scientist, 19 November 2005.129.24.141.152 23:26, 5 January 2007 (UTC)LekLiberty

I've made some comments on Scientific Method and Peer Review in these discusssion sections and have read it all. In my opinion, and I am a practicing "scientist", and I don't think what we have here is "correct" or shows any sign of "converging to anything" near to correct. FYI on my "concerns" are that I see the base SM missed and over defined, and added to it a bunch of subsidiary issues such as propagation of "results", management of SM activities, validation "types" for SM, old historical views from the early days when it was part of "natural philosophy", and such that is basicly confused and confusing to "students and teachers" (cuz they took me to task and told me so!!)

I suggest that if you guys need to get your "verifiable" needs met and get "practitioners of science and peer review" definitions, that you/we ask for them from: scientists, scientific organizations, science funding agencies, and the same types for history of science sources, philosophy of science sources, and any others you feel you need for your verifialbility requirements. I do feel it's presently wrong enuf to worry about. I suggest you start as Wipepedians "formally requesting inputs" and would suggest you start with the Society of Sigma Xi, which is "the" non-denominational honorary science society. Their publication "American Scientist" would I'm sure receive a letter along these lines and probably publish it in one of their publications. I am a memeber, and would do it for you; but, I think the "formal Wikepedia", whoever/whatever that is, should do it!! LekLiberty

Here's an idea for a way forward. One problem with scientific method is that there really is no accepted definition. There are common notions of what scientific method is, which although similar enough to find some commonality are, nevertheless, different enough that any attempt to address all of them in one go is doomed; Abstracting away to preserve only that which is common will produce something completely banal. Doing the opposite and trying to retain every detail, I have difficulty even to imagine.

This is the main reason I started the history of scientific method article; taken individually the different ideas about method are intelligible and interesting.

What I’ve noticed while researching for that article (which it woefully incomplete I should point out) is that there are some themes that are really worth expanding on. Take for example Bacon’s eliminative induction which is almost, but not quite, identical with Mill's Methods and remarkably similar to Popper’s falsificationism. All three of these developed methods that value negative instances over positive: basically deductive reasoning, but very different to syllogistic reasoning which we might also mention.

Something I still have to look into is the development of inductive method. Some great minds have tried to supply induction with the force of deductive logic, including Jacob Bernoulli, Thomas Bayes, Laplace and Rudolf Carnap. If they don’t have something interesting for us, I’d be very surprised. --ChrisSteinbach 22:21, 24 January 2007 (UTC)

Regarding the history article, it might be worth mentioning that I've been trying to find the original source of the common scientific method. I mean the one with a small number of steps: 1) Observe, 2) Develop hypothesis, 3) Bake for 15 minutes, 4) Serve with a selection of vegetables. Or whatever it was now. Any leads? --ChrisSteinbach 22:32, 24 January 2007 (UTC)

Chris, I googled 'crucium experimentis' and found Giovanni di Guevara (1627) In Aristotelis mechanicas commentarii. The concept is the crucial experiment, which ought to set the extant issues about some hypothesis into bold relief, and allow some simplified thinking about the hypothesis. Right or wrong, Prove or disprove, True or false etc. --Ancheta Wis 23:03, 24 January 2007 (UTC)

More factors in the rise of scientific method as an explicit concept:

• Galileo's experiments 1638
• Newton's explication of induction from experiment 1666, 1687
• The general popularity of Do It Yourself experiments during the Age of Enlightenment
• Geo Boole Laws of Thought 1854
• William Stanley Jevons Principles of science 1874
• The professionalization of science in the 19th c. especially in the German-speaking countries (Ernst Mach was professor of the History and Theory of the Inductive Sciences)

--Ancheta Wis 11:00, 25 January 2007 (UTC)

Thanks for the pointers Ancheta. Guevara seems to have played a role in getting Galileo off the hook concerning his atomist ideas in The Assayer. Beyond that I'm not able to find anything (or anything I'm able translate I should say) about his own scientific ideas. --ChrisSteinbach 22:13, 25 January 2007 (UTC)

Chris, I found some citations which I put in the History of scientific method --Ancheta Wis 23:46, 4 February 2007 (UTC)