Wikipedia talk:Scientific standards/Archive 2

Archive 1

Archive 2

The subject of "Scientific standards"

When writing articles or sections involving science, we rely on the most reliable sources to describe the scientific evidence accurately while weighting material toward prominent, majority opinion in the relevant scientific discipline.

The word "involving" in this quote has been contentious. It was switched from "about" to "concerning" back to "about" and now on to "involving" and then back and forth again. I believe the concern is that the intent of using a more general word than "about" is an effort to give science a dominion over other fields. The interest in a more general word is not some sinister plot to expand the realm of science. It acknowledges that science is used in a great many articles lots of which don't directly concern science. The type of articles that first come to mind are technology pages which are best explained through science. But even something like the moon isn't necessarily about science, should we split page into two moon (various perspectives) and moon (scientific perspective). When science is used it should live up to a set standard. I think its up to authors to decide if the science perspective is being invoked or if it should be. Once it is invoked this would be the standard to which the citation and writing must stand.--OMCV (talk) 13:21, 28 August 2008 (UTC)

One way to address this would be to say "when presenting scientific perspectives in articles..." Treating the viewpoint rather than the subject as the scope avoids the thorny issue of determining which subjects are "science" subjects. It also avoids the issue of whether scientific viewpoints should or should not dominate when multiple viewpoints are presented. It would simply clarify how to present scientific viewpoints in those contexts and to that degree to which scientific viewpoints are relevant and have weight. In my view, determining when a scientific viewpoint is relevant involves matters of philosophy which are (in my view) completely inappropriate for a policy page and something our policies shouldn't be getting into. I suggest a much more pragmatic approach which minimizes any chance of us attempting to call the shots for how the rest of the world should work or think. Scientific views are relevant when scientists have published such views through sources the scientific community considers reliable, whether other people think they should or not. No fuss, no muss. Less theory, more empiricism. Best, --Shirahadasha (talk) 18:36, 28 August 2008 (UTC)

I would prefer to address this concern in some other way. My fear is that writing something like "when presenting scientific perspectives in articles...." attempts to claim the existence of a scientific point-of-view when we don't have a consensus of Wikipedia editors that a scientific point-of-view necessarily exists. I guess I see the issue as one that is bigger than simply "perspective". I see it as one that is specifically subject-based. Yes, there are thorny borderline issues (e.g. the Virgin birth of Jesus) but those are going to be thorny regardless of whether we fall back on "perspective" idealizations or otherwise. Rather, I think that where you and I and most editors agree is that whenever we have scientific evidence that is directly relevant to an article, we must straightforwardly present that scientific evidence. Furthermore, when there is a dispute about what constitutes scientific evidence, what scientific evidence is relevant, or how scientific evidence should be presented, we must rely on reliable sources that verifiably present scientific evidence. I feel that this approach is one step more pragmatic than declaring something called a "scientific perspective" which is likely to offend the better sensibilities of those Wikipedians who disagree with this characterization of sources related to science. What do you think? ScienceApologist (talk) 23:42, 28 August 2008 (UTC)

Ok I still agree we should side-step the issue of whether science is POV. I realize my first entry addresses science almost entirely in the POV way but I don't think thats the important part. The question this section really needs to address is when do "science standards" get applied to an article. I think your idea of when its directly relevant and the text that follows covers this well.--OMCV (talk) 00:29, 29 August 2008 (UTC)

The way to sidestep discussions of SPOV is to simply say "scientific material", and then discuss when such scientific material is relevant to the topic and when it isn't, rather than saying that the material is the point of view of scientists. It's relevancy-in-context. --Nealparr ^{(talk to me)} 02:54, 29 August 2008 (UTC)

I can probably draft something to the effect of what I'm talking about this weekend (if I don't get too busy). --Nealparr ^{(talk to me)} 01:50, 30 August 2008 (UTC)

There seems to be some overlap of issues here with Wikipedia:Manual of Style (medicine-related articles) and Wikipedia:Reliable sources (medicine-related articles). Including the similar ambiguity of "medicine-related", and the NPOV question (though hardly discussed there) regarding the fact that many articles will also be "SomethingElse-related" and different fields/approaches may differ in their fundamental conceptualization and structuring and evidential methods regarding the same issues. EverSince (talk) 15:59, 5 September 2008 (UTC)

Is there a scientific opinion?

References to "scientific opinion" were removed recently. What is scientific consensus if not the collective opinion of science? At Wikipedia, where everything is about viewpoints and opinions, we should directly address scientific opinion rather than pretending it doesn't exist. I've read some editors say there is no SPOV. I agree that many times SPOV = NPOV, but I don't think we can honestly say there is no SPOV within the parameters of the Wikipedia project.

I should add that a perceived adoption of an SPOV is where past efforts at adopting a scientific standards policy failed. I feel addressing SPOV is vitally important to getting this accepted in the community. I don't see how scientific consensus doesn't equal scientific point of view, but more to the point: Whether or not there is factually a SPOV, most editors are scared of it. It needs to be addressed forthrightly. --Nealparr ^{(talk to me)} 21:23, 20 August 2008 (UTC)

Scientific consensus is what is left after everybody publishes their results. It isn't an "opinion" any more than the position of the buildings in your town is an opinion or the orientation of stars in the sky is an opinion. We can interpret scientific consensus and that interpretation is an opinion, but the existence of consensus does not require an opinion. It either exists or it doesn't.

The proposal says up front that Wikipedia does not assert whether a scientific opinion exists or not. We should probably try to stick to that throughout the proposal. I am here to say (and I know others agree with me) that SPOV is simply a misnomer when it comes to writing an encyclopedia.

Does a scientific opinion exist? In some venues, perhaps. For example, in a court-of-law in the United States, there is a treatment of science that lends it the veneer of an opinion. That's fine, but Wikipedia is not a US Court of Law. It is an encyclopedia. As a reference work, it relies on sources. Scientific evidence is one kind of source. It is the closest approximation we can get to a source that is simply fact assertion rather than opinion. That's why I am of the opinion that SPOV is essentially a misnomer.

Can you show me a place where an editor evinced a fear of SPOV? I have read through all kinds of arbcom nominations, and answers to questions posed by User:Ragesoss never once indicated fear of SPOV.

ScienceApologist (talk) 21:38, 20 August 2008 (UTC)

I think you're wrong about that. Interpreting evidence is what leads to scientific consensus, a mutual agreement like any consensus. It's a shared opinion. If someone interprets scientific consensus, they're interpreting an interpretation of evidence. That's a double-interpret : )

Evidence of the fear of SPOV can be found on this talk page, and the fact that WP:SCI and WP:SPOV are discontinued. --Nealparr ^{(talk to me)} 22:11, 20 August 2008 (UTC)

There is an activity I have my students do which lists a number of statements and asks them to determine what is an "interpretation" and what is a "fact". This is an example of a coding exercise and it is a valuable one because it asks some really foundational questions about the metaphysical implications of studying reality through science. Is it a "fact" or an "interpretation" that Polaris is the North Star? In some sense it is an interpretation because the idea "Polaris" and "North Star" are constrained by perception, language, definition, and representation. Ultimately, however, there is a fundamental premise that reliable sources agree upon, with the possible exception of certain purists in Berkeley's camp, that phenomena do in fact exist independent of the mind and are probed through the scientific method. The fact of Polaris being the North Star stands for an external phenomenon of a certain pulsating red giant star that is in a particular direction which aligns with the only non-rotating directional vector associated with the Earth's surface. Blah, blah, blah: I can't get across the physicality of the situation without using interpretative language: I can only get more and more detailed until I bore you to death with explanation. At the end of the day, though, inasmuch as science ever has "consensus", it is about the physical reality that is independent of our interpretations of it. Therefore it is, by design, about as far away from "interpretation" as you can possibly get. When we talk about consensus, we're literally talking about what is left when all the data are in. Consistency is the name of the game. Induction requires it.

WP:SCI was discontinued because people thought it didn't really make any notability claims not already dealt with at WP:N. As far as I can tell, WP:SPOV was rejected with little-to-no-discussion at all.

ScienceApologist (talk) 22:29, 20 August 2008 (UTC)

I appreciate that in science classes scientific consensus is taught as fact. However, outside of science (and inside of Wikipedia), it is only one approach to describing consensus reality. There's other approaches outside of science. Further, I'm sure we can both agree, science is a process rather than a product. It's not a complete, finished, "we're done" product that describes the world around us. Even in science "the view" changes, adapts, reconfigures, and what not. There are few immutable facts in science. Even the facts that are solid are added to all the time, becoming new facts (plus) that supercede the old ones. There wouldn't be the term paradigm shifts if what science calls "fact" was immutable. It's the fact-of-the-moment, which, of course, is simply the current interpretation of reality. Simply because science adopts the assumption that "phenomena do in fact exist independent of the mind", it is a view. In fact, it is a view disputed by subjective idealism. I, too, could get into long, boring explanations of why it is an interpretation, a view, but let's skip all of that and get down to what really matters -- the parameters of Wikipedia. In Wikipedia, a fact is defined as something not seriously disputed. What science describes as facts are things often disputed by other views on the matter. We thus, at least at Wikipedia, do not call these things facts. --Nealparr ^{(talk to me)} 23:47, 20 August 2008 (UTC)

So, I'm not sure I understand you. Are you saying that the statement "Polaris is the North Star" is actually an opinion that should be referenced as such? Should we write "current scientific consensus is that Polaris is the North Star"? I'm really not following you. Either we write plainly about what the evidence is or we do not. No one is arguing that things can't change and there may be things reported plainly in Wikipedia which are strictly verifiable but not true. That's not our problem, but you seem to be focusing on it. Whether the scientific evidence is "wrong" or "likely to be superseded" or "subject to a paradigm shift" is not Wikipedia's concern. Either we are given permission to write plain statements without hemming and hawing over the empiricism of the statement or we write an encyclopedia that pretends that everything is opinion and nothing is fact. Just because someone disputes something that is a plain fact doesn't mean that we must accommodate them, does it? ScienceApologist (talk) 18:50, 21 August 2008 (UTC)

Yes it is an opinion. No it doesn't have to be referenced as such in most cases, because it's most often not a seriously disputed opinion (Wikipedia's definition of fact). Here's how we speak plainly: Context. In most contexts Polaris being the North Star is an undisputed opinion. In the context of describing the nature of consensus reality, it is actually disputed that there is even an external anything. Again, the philosophical point of view of subjective idealism would say that science is a mental construct studying other mental constructs and that there is actually no objective reality. That point of view isn't really relevant in the context of an article about Polaris, but it's still a point of view that is out there. The relevant point of view on Polaris is that of astronomy, but as we move back from that micro-context into greater contexts, it gets fuzzier what is a relevant point of view. We started this conversation talking about whether science has a point of view. That's a wider context. We're out of the micro-context of a discussion on an astronomical topic and into the macro-context of approaches to describing reality. In that context, yes, science is an opinion or viewpoint.

Now, what context are we actually dealing with here and now? Well, we're writing a proposal to adopt a policy (or guideline, etc.) on science as a whole, within the parameters of the Wikipedia project. It's not science in a micro-context. It's science in a macro-context. In that macro-context where science is a point of view that's disputed, we need something addressing in the policy that it's one of many viewpoints. It's already sketched out in the proposal, but it needs to be made clear because someone's going to call us on it. We need to say clearly that in the context of comparing approaches to describing reality, science is just one viewpoint, before we get into describing what to do in micro-contexts where the consensus in science is the primarily relevant view.

Note that I am talking about the proposal writing and not the article writing. Polaris being the North Star is not seriously disputed in that article. The Earth being round is not seriously disputed in the Earth article. The sky is blue on a bright sunny day. Water is wet. No disputes there. These are micro-contexts where the "we live in the mind of God" argument isn't particilarly relevant. We can speak plainly.

In fact, it is actually my position that most of the headaches plaguing science-related articles can be resolved by clearly addressing context-based writing, addressing macro- and micro- contexts. It is only from contexts that you get things like weight. If there was a concerted effort to describe writing in context, we'd have a very useful policy meeting a broad consensus I am sure. Wikipedia, by nature of the NPOV, is necessarily contextualist. --Nealparr ^{(talk to me)} 20:26, 21 August 2008 (UTC)

So if I understand you correctly, the standard you are proposing is that only "seriously disputed opinions" should be described in the text as opinions while opinions that are not seriously disputed should not be described in the text as opinions. I think this begs the question of what constitutes a "serious dispute". My issue is that there are instances where people argue incessantly on whether "serious disputes" exist or not. In many cases, there are only contextual answers to this question, but this begs the question of who determines the context? It's easy to see what the context is for an article on a round Earth or the Virgin Mary, but what the hell is the context for parapsychology (to reopen an old wound)? If we take your approach we end up encouraging fights that have no chance of ending. Rather, I think if we abandon the idea that scientific evidence is opinion-based we can start to work towards the ideal espoused by WP:ASF and stop writing articles about things that have no scientific evidence that read like he-said she-said fests. Instead, we just plainly state the evidence (or lack thereof) and move on. No wrestling with opinion necessary. That would be ideal. Do you see a problem with that? ScienceApologist (talk) 20:36, 21 August 2008 (UTC)

Those disputes are going to happen anyway, because Wikipedia is source-based. Someone's going to come along and say "parapsychology is science, here's my sources". Someone else is going to come along and say "parapsychology is pseudoscience, here's my sources". Then they're going to duke it out over which sources are more reliable and the prevailing party gets to tag the article, or maybe there'll be consensus to tag it as both. The point is, it's still a process of determining the context of the topic. Once the context is determined, a treatment is decided (which views have more weight) and the article forms around that contextual treatment. It's not my standard, it's the standard (defined as wiki-tradition) already in place. It's what's already happening. A proper policy develops tools for making that process more effective. Take another topic: intelligent design. The consensus-building process looked at the sources, determined which are more reliable, prominent, and so on, and that process determined the context that intelligent design is properly treated as pseudoscience. The policies already in place for dealing with pseudoscience (WP:PSCI and WP:FRINGE) kicked in and the article formed as a result of that context-determination. It's just simply what happens in consensus-built wikis that have policies like neutrality and undue weight.

A proper scientific standards policy, in my opinion, would clearly address context-based writing, clearly point out when the context is science, and then clearly separate non-science from science in those contexts under the rationale that non-science is out-of-context. Knowingly or not, the current proposal is reflecting that, even the parts I didn't write. I feel that's because it's the natural approach of Wikipedia, or what's eventual. --Nealparr ^{(talk to me)} 20:53, 21 August 2008 (UTC)

I see things in a very different way. I think that the conflicts arise because people object to facts. E.g. people object to the fact that parapsychology is a marginalized field of study. People object to the fact that parapsychology is maligned by scientists. People object to the fact that no conclusive evidence for the existence of psi has ever been demonstrated to the satisfaction of independent evaluators. That's the source of most of the conflict. What happens is people begin "tagging" the article instead of engaging in good editing... but tagging an article is really not informative nor is it necessary to writing the best article. What we end up with is a place where people end up thinking that there is a continuum of opinions on a subject when this, in fact, may be only incidentally related to the topic itself. The opinions of Dr. Joe Shmoe or John Q. Public are really more applicable to journalism than they are to reference works because journalists don't have the time or resources to do source-based research and are generally charged with reporting on opinions as well as facts. We have the luxury of being able to decide that opinions on a subject may not be relevant to a reference work on a subject. Are the various opinions people have about the Bohr atom relevant to an article on the subject? I'm going to say probably not. If you take the tack instead that everything is an opinion then you are simply asking for more trouble than what is required or healthy. ScienceApologist (talk) 22:16, 21 August 2008 (UTC)

What you described as people objecting to facts (marginalized, maligned) is really people objecting to placing a subject in the context of outside opinion, something Wikipedia does through WP:PSTS. Editors like that would rather the subject be described in-universe, a narrow context, because in-universe it is not maligned or marginalized. That's a fact too. Only when you put it in the greater context (as Wikipedia tells us to do) is there objection to facts. There's not much dispute on the timeline of intelligent design. It's a historical context, not prone to arguments because it's the same within the community as it is outside the community. Where the topic becomes prone to argument is when it gets placed in a wider context, like on the question of whether or not ID is science. Unfortunately for IDers, that's a context they placed themselves in by claiming to be science. In that context, scientific opinion is introduced, and editors start complaining. What they're complaining about, though, is the outside opinion, not the facts of the matter. They know ID is not accepted by the scientific community. They're complaining that the greater scientific community has greater weight over what they say. They want their opinion described as they see it. That's the locus of the dispute. In my experience, that's always been the locus of the dispute... and it's entirely about context-placement. Even "relevance" is context-sensitive. (By the way, "tagging" was metaphorical. I meant setting out a "treatment" of the topic). --Nealparr ^{(talk to me)} 23:49, 21 August 2008 (UTC)

It doesn't have to say "context". Some other wording may work. I'll sit back and let others discuss it. --Nealparr ^{(talk to me)} 01:39, 22 August 2008 (UTC)

I agree that this is often how things play out on Wikipeida. WP:WEIGHT gets (ab)used far more often than it should. There are simple facts that are argued about all the time when there are people who disagree with those facts. I don't think just because someone disagrees with a fact that this makes the fact suddenly change into an opinion. Unfortunately, for Wikipedia, because people are so obsessed with the idea of everything being an opinion when consensus isn't obvious, people make appeals to opinions when they really are talking about facts. The issue is that if we could just take scientific evidence seriously, we would never have this problem on articles related to science. ScienceApologist (talk) 15:41, 22 August 2008 (UTC)

SPOV does not equal NPOV. While the SPOV is usually the main POV, there is usually one or two other lesser POVs that aren't science or aren't part of the main scientific consensus. As long as they abide by WP:FRINGE, then those views belong in too, otherwise the article wouldn't be NPOV. Deamon138 (talk) 23:53, 20 August 2008 (UTC)

We can all agree that there is a "scientific consensus", though, right? Using that seems to be uncontested. --Nealparr ^{(talk to me)} 00:23, 21 August 2008 (UTC)

Assuming there is one on a case-by-case basis, yes of course, but if there are other minority views that are notable, inside or outside science, then generally, they belong in the article, but with less weight to the consensus view. Of course, the longer the article, the more likely it is that some of the non-science alternatives would have to be split off into a separate article. Deamon138 (talk) 00:43, 21 August 2008 (UTC)

In order for a Wikipedia article to say what the scientific consensus is about a subject, wouldn't there need to be a citation to a reliable source saying what the scientific consensus is about the subject? Unless we are talking about obvious, uncontroversial and unmistakable conclusions, it seems that for editors to put in their own opinions about what constitutes the scientific consensus would be original research. Dlabtot (talk) 00:51, 21 August 2008 (UTC)

There would have to be a demonstration of consensus, yes. We don't have to get all weird about it and require a quote that says exactly "There is a scientific consensus that such and such..." but there does have to be reliable sources backing it up. That was put into the policy proposal yesterday (I think). --Nealparr ^{(talk to me)} 04:47, 21 August 2008 (UTC)

Hemming and hawing about consensus seems to be a purely Wikipedia-based phenomenon. I have read hundreds of thousands of pages of science text and very rarely do they say "the scientific consensus is..." More often than not they simply state the scientific evidence/observation/theory plainly without trying to frame it as an opinion. That's also the way other encyclopedia approach it. ScienceApologist (talk) 20:18, 21 August 2008 (UTC)

There are opinions of individual scientists, and there are numerous unscientific opinions touted as scientific authority or common knowledge. I have seen enough of the latter here at WP to surely raise the spirits of any modern day Lysenko or Torquemada. The latter often involve severe cases of WP:TROLL & WP:IDONTLIKEIT, often associated with severe deficiencies of WP:NPOV, WP:AGF, and perhaps Li, from sources that are both dead and wrong (contradicted by more recent, better sources or actual facts) although sometimes recycled into nominally WP:RS, uninformed sources that still fail fact checking.--TheNautilus (talk) 20:57, 21 August 2008 (UTC)

I have two issues with the verbiage of "scientific consensus." First, like the word "theory", consensus in science has a somewhat different meaning than in the non-science world. The "debate" is not in a big group, and then there's a vote, but it's through research. Second, I contend that scientific consensus refers to issues where there is debate. Otherwise, it's just science or scientific (or medical) research. When we use terminology like consensus we make it sound like there is an SPOV (I contend there is no SPOV, because science lacks a POV) that counters the consensus. I would never write "scientific consensus supports Evolution." There is no consensus about Evolution, because there's no minority report available (aside from the non-scientific ID and creationist arguments). On the other hand, consensus could be used with theories or ideas that have a significant minority view (non-fringe, of course). What comes to mind is the Cretaceous-Tertiary extinction event. The consensus is that dinosaurs went extinct because of a bolide impact 65.5 million years ago. But there are minority theories that a) dinosaurs were dying out before the event, b) the bolide impact had nothing to do with it, but triggered some other event, c) the environment was changing, d) excessive vulcanism. But these minority theories have significant sourcing and study. OrangeMarlin ^{Talk• Contributions} 22:53, 22 August 2008 (UTC)

So you're saying if the balance in science is (I'm using numbers here, but pretend the numbers are valid arguments, i.e. not a vote) 90% for one view, and 10% for another, then the first view would be called the scientific consensus, but if there is only one view, so that it has 100%, the there isn't a scientific consensus. That makes no sense. It is obvious that while there might not be much dispute within the scientific community about Evolution (forgetting that some scientists are ID proponents), the very fact that scientists come out and make statements in support of Evolution against ID shows that there is a consensus, and that each summary/conclusion to a paper is an affirmation of a view given by its writer based on the evidence. Consensus is just defined as the overall view of a group based on the best reasons. 100% support for something is still consensus. Ask all scientists if they believe in Evolution and you will get their views. There's your consensus. Deamon138 (talk) 00:40, 23 August 2008 (UTC)

There is no voting in science. OrangeMarlin ^{Talk• Contributions} 03:52, 23 August 2008 (UTC)

Err, I know that, that's why I said, "I'm using numbers here, but pretend the numbers are valid arguments, i.e. not a vote". The comment was basically to deny your idea that science doesn't have a consensus, even in uncontroversial topics, when it clearly does. Deamon138 (talk) 14:57, 23 August 2008 (UTC)

Frankly, I am really surprised to see so many people saying "there is no scientific point of view". It seems like such a strange statement because it's glaringly obvious that there is. Just saying that the scientific method is the best method for describing the world around us is a point of view. Just saying that there is an objective reality to study at all is a point of view. Just saying that some logical reasoning is valid, and others aren't, is a point of view. Doing research and forming conclusions is a point of view, because, well, a conclusion is in fact a point of view. Choosing what to study and what not to study is a value judgement, even if it's a judgement made for practical reasons (like falsifiability), and judgements are points of view. I'm so frustrated by the denial that there is actually a point of view, I'm not sure how to proceed here. Typically it wouldn't matter, but Wikipedia is all about various points of view, and for a group to say that they alone have none, it feels like we've reached an impasse. It's a deadlocke in the conversation, at least at Wikipedia. Is this a case of not being able to think outside your own circle? --Nealparr ^{(talk to me)} 01:00, 23 August 2008 (UTC)

Point of view, by definition, is rather arbitrary. Point of view is based on opinion, much of it subjective. Science is based...well, read about it. OrangeMarlin ^{Talk• Contributions} 03:52, 23 August 2008 (UTC)

Point of view, by definition, is attached to the viewer. A collective that shares a point of view isn't arbitrary at all. It's a standard. You have terms like scientific standards, scientific consensus, conclusions of science, scientific this and that but Oh No! It's not a point of view! Anything but that! Never mind that the term is all over the place[1] Science is based on what? Draw it out. Science is based on a process designed to determine fact from observations. From those facts, science as an institution draws conclusions. Those conclusions, assumed to be facts themselves, are entirely a point of view stemming from the science process. Both the existence of objective facts and whether observation leads to facts is questioned by other philosophical viewpoints once you step outside the circle of science. Outside the circle of science, it's one of many points of view.

Yeah, science is awesome and all that. But this is Wikipedia. We deal in points of view. If we can't agree on that, the Wikipedia-process becomes deadlocked. --Nealparr ^{(talk to me)} 05:05, 23 August 2008 (UTC)

The pains that practitioners of the scientific method take to remove the biasing effects of POV from their observations and experiments is what gives "science" the ability to say that it is not simply a point of view, but rather, the nearest thing to a correct description of reality we can manage. Any process that can produce that is valued well above a "viewpoint".--Father Goose (talk) 07:28, 23 August 2008 (UTC)

I get that. I really do. But there are still fundamental assumptions made for practicality that does in fact bias their description of reality. I hate to keep repeating it, but the simple assumption that there is an objective reality at all is biased. It's contrasted by some forms of idealism. No matter how much they remove bias and point of views in describing objective reality, they still work in the framework of objective reality. Way above in this section I was trying to address how we can possibly resolve that bias by addressing contextual relevancy (ruling it as a primary view on some topics and a comparative view on others), but I kept hearing "there is no scientific point of view" at all. That's simply not true, and the denial of any point of view needs to be addressed before I, personally, can work on the proposal. To me, it needs to say "Wikipedia does not have/endorse a SPOV", or something similar, and explain that in certain contexts the scientific point of view is a comparative view, not a primary view. To me, that's fundamental in a science standards policy at Wikipedia because of NPOV. --Nealparr ^{(talk to me)} 08:18, 23 August 2008 (UTC)

By comparative view I mean a view to be compared with other views. --Nealparr ^{(talk to me)} 08:33, 23 August 2008 (UTC)

I agree with Nealparr. It fits much better with the WP rulebook if we don't take a stand and claim that "Science" (or rather, the consenus among scientisists in 2008) is without bias and represtent an objctive truth. (even though that for most practical purposes is the case).MaxPont (talk) 08:51, 23 August 2008 (UTC)

I also mostly agree with Nealparr here. The scientific consensus of the day (assuming good faith here thus presuming all scientists are good people, which I would say most are anyway) is based on the scientific method. But there are flaws in the scientific method, such as it is based on induction, which logically, is fallacious. This for me is part of a distinction between logic and reason. Logic just tells you what is possible, and because science is based on induction (and also due to other philosophies like idealism and solipsism), logically, it is possible that science is wrong. But reason allows one to evaluate the likelihood of each possibility. Is it likely that science is wrong? No, it isn't. The amount of inductions that science does is evidence each time that it is working, as more and more evidence is in favour of each theory. So in my mind, it's logical to be open to the possibility that science is wrong, but not really reasonable. So faith is still required to follow science, but less than in any other philosophy, therefore, though science is a POV, it is the best POV we have, as the scientific method is the best tool we have to gather information. Basically, although I can only prove that I alone exist, it is fairly unreasonable for me to only believe that I alone exist, thus denying everything else I can empirically see (my laptop, Wikipedia, my fellow editors, everything). Hence the SPOV, while it is the best POV, it is not the same as the NPOV, and we would be kidding ourselves if we said it was. Anyway, that's my two cents.

Oh and Nealparr, one slight point I don't fully agree with you on, science doesn't completely assume an objective reality, as I would say that most interpretations of Quantum Mechanics have to deny it, though there are some interpretations that still use it. Deamon138 (talk) 14:54, 23 August 2008 (UTC)

There is this weird false dichotomy being painted here that goes something like this: "If we don't want science to be painted as objective and the absolute truth then we must acknowledge that a scientific point-of-view exists". This argument seems to be made by editors who are adamant that Wikipedia maintain neutrality. However, this statement also is, ironically, one which requires Wikipedia to take a point-of-view on a meta-issue in favor of the existence of SPOV. It is clear from this very talk page that the existence of SPOV is at least debated, if not actively disputed by a large segment of the editors. There is the point-of-view of scientists, there may even be a point-of-view of a majority of scientists, but scientific consensus is not necessarily a point-of-view (although, I admit, sometimes it is treated as one) and certainly science itself, as a process has no point-of-view any more than any other algorithm has a "point of view".

I think that we can simultaneously posit that we shouldn't paint science as objective and the absolute truth while remaining neutral (imagine that) on the issue of whether of not there is a scientific point-of-view. The clearest way to deal with this situation is to take no stand on the dispute: it is entirely irrelevant anyway.

ScienceApologist (talk) 01:15, 26 August 2008 (UTC)

There's a difference between conversation on the talk page and what goes in the proposal. I'm saying there is clearly an SPOV. The proposal doesn't have to. Wikipedia doesn't have to take a stance that says "There is an SPOV, and it is not neutral in all cases." Instead, the earlier wording was "Wikipedia does not have an SPOV" (which doesn't go so far as to acknowledge an SPOV exists, but just says Wikipedia doesn't have one). There's other ways to word it that would be less contentious in the proposal than what's here on the talk page. As I said above, it comes down to relevancy. That's not really controversial. Sometimes material is relevant, and sometimes it isn't. Side-step the whole point of view business. We can talk about relevancy in the proposal without implying that science does/doesn't have a point of view. Rather, science and anti-science material is sometimes relevant, and sometimes not. --Nealparr ^{(talk to me)} 02:09, 26 August 2008 (UTC)

Agreed there is no need to decide the "nature of science" to have a "scientific standard". Could we bullet the points we want to get across in this policy that isn't covered in other policies and then remove all the extraneous text? I think that would help a lot.--OMCV (talk) 02:18, 26 August 2008 (UTC)

The whole idea of Wikipedia editors deciding whether there is a scientific consensus bothers me. A few examples:

• A debate has been going on since 1990 about the evolutionary position of the Halwaxiids. Try reading that and deciding what the consensu is - and there are only about 5 significant players in that.
• Evolution says that belief in Darwin's theory of evolution or one of its genetics-boosted enhancements is the scientific consensus. There was quite a debate over that in the Talk page, until I got lucky and found a couple of refs. But if these refs hadn't turned up, or in other caases where there are no similar pronouncements from scientific associations, how do you prove consensus?
• But how neutral and authoritative is the consensus-proclaiming group? Lysenkoism springs to mind.
• Apparent scientific consensuses have turned out to be very mistaken, e.g. over continental drift, the origin of tetrapods (see Evolutionary history of life), dinosaur metabolism (see Physiology of dinosaurs). Science does not work by consensus, it works by evidence and logic.
• Scientist A quotes B who quotes C, where C is the only one of these who has actually done research in that field. How many "votes" should A and B get? What if D's opinion differs from C's?

On highly-politicized scientific issues, counts of papers or authors only show how research grants are being distributed. -- Philcha (talk) 17:59, 6 September 2008 (UTC)

Yeh. Read this. The statement is against policy [2]. However, the consensus section has been getting better since I had problems with it before. It's now almost up to policy, like when you shoot at a target and are only a few feet off. I suggest we merely copy the policy section over here, and be done with it. But it's just an essay. The problem is that people might point to it, and some noob would not know that it actually is just someone's opinion of how it ought to be, in spite of the essay box. I admit to not having read all of this. ——Martin^phi ☎ Ψ Φ—— 20:12, 6 September 2008 (UTC)

Science is not an effort to establish "consensus"

This proposed guideline is severely flawed on the outset by the claim that scientific findings "are published under a peer-review system in an effort to develop an established scientific consensus." I would dearly like to see some references, multiple please, that suggest that this is the purpose, goal, or even a consideration of the scientific method.

As a trained and practicing scientist, I have never seen such a proposal in any text book, journal article, or even in the philosophical treatises on the nature, purpose, and evolution of science and the scientific method. Only rarely have I have seen scientific bodies issue statements regarding the consensus of their opinions, and if they do so it is generally to assert themselves as citizens (not scientists) in an effort to influence some policy of interest to them...and sometimes to satisfy the requests or demands of politicians for a "consensus" which the politicians can then use in their debates.

The section "Consensus" is also not about proper science and the scientific method but is just a series of unsourced arguments, otherwise known as original research (which I thought was forbidden at Wikipedia). The statement "It should be acknowledged when there are consensus statements from scientific bodies such as the International Astronomical Union or the International Union of Pure and Applied Chemistry" is meaningless. Are these consensus statements about the naming of asteroids and chemicals? Or about the best way get more research funds, to cure AIDS, or to mix a martini? In any event, and worst of all, it appears that this section is implying that because in rare occasions some formal consensus statements are made, that the consensus of scientists opinions can generally be easily found and that editors should therefore be free to assert that what they view to be a "mainstream" consensus of opinion as a means of excluding scientific reports from articles that are not part of that "consensus."

This section would be far better if it were to properly clarify that claims of scientific consensus should only be made when there is scientifically defensible evidence, such as a poll of scientists who are experts in the field, that the claimed consensus actually exists.

Given this and many other problems with this proposed guideline, I believe it will not be widely accepted and will give rise to more disputes, not fewer. --SaraNoon (talk) 02:56, 11 September 2008 (UTC)

Thank you, SaraNoon, for articulating a fairly vague feeling I've had about this project. I do hope others will chime in here, as these issues need clarifying. --Hordaland (talk) 13:24, 11 September 2008 (UTC)

I totally agree. Even a "poll of scientists who are experts in the field" would be dubious, since expertise is not an on-off trait. For example Stephen Hawking is a huge expert in theoretical physics but there are other aspects of physics in which he is not at all an authority. Less flamboyantly, in paleontology I can see scientific articles citing other articles in fields / sub-fields in which the citing authors are not active researchers.

As SaraNoon points out, statements of "consensus" from scientific associations may reflect mainly political factors.

No hypothesis that has not withstood at least 20 years' critical examination can be called "consensus", "mainstream" or whatever. That means we actually have to quite liberal in in our criteria for articles about current issues - if an idea got published in a peer-reviewed publication and was not demolished very quickly, it's eligible. Since scientific debates move quite slowy, anything less than 2 years old must be treated as provisional, no matter how eminent the authors or how well they fit with ideas expressed earlier. -- Philcha (talk) 14:40, 11 September 2008 (UTC)

Wikipedians are overly-sensitive on sourcing. This is exampled above in SaraNoon calling the opinion of some Wikipedia editors on how to approach science articles as "orginal research". It's not an article. It requires original research to write a guideline based on editor's experience in writing here.

Science doesn't strive to establish consensus? The fact is, "The Earth is round" is an example of a scientific consensus that need not be written as "The scientific consensus, as stated by the AAAS, is that the Earth is round [1]". There are millions of such examples of scientific consensus that aren't stated "it is the consensus of science" in a book. Nevertheless, they are the consensus of science. The scientific method is, in fact, a method for building consensus through experimentation, repeatability, and peer review. The clue there is peer review. Any review process among one's peers is to establish a consensus beyond one's own experience. --Nealparr ^{(talk to me)} 16:45, 11 September 2008 (UTC)

Science does not strive to establish consensus, as everything is open to question, even the most venerable and successful theories - Einstein's revision of Newtonian mechanics is a good example. Science is an ultra-Darwinian competition between ideas. Some ideas become consensus because they win the debates, not because of some agreement between participants. However such victories are always provisional, see e.g. Newton / Einstein - that is how progress is made.

What we were discussing is whether the notion of "scientific consensus" has any place in Wikipedia. I would accept that it's a useful shorthand to avoid rehashing debates that have apparently ended. However the criteria I suggested above indicate that this is purely a pragmatic approach and that science has no sacred cows.

Peer review has nothing to do with consensus. It's just a first-cut test to ensure that the would-be article is sensible in its handling of evidence and reasoning. For it go any further would be a form of censorship.

We do not need a guideline that apparently tries to enshrine particular scientific hypotheses. -- Philcha (talk) 17:24, 11 September 2008 (UTC)

What you described above is exactly a strive towards consensus. It's actually very similar to Wikipedia's take on consensus, especially where it points out that consensus can change. Just because science does change, doesn't mean that it doesn't strive to establish consensus-based frameworks. You said that 'ideas become consensus because they win the debates, not because of some agreement between participants'. For a debate to be "won", there must be a conclusion. An open-ended debate doesn't have a winner. In a "won" debate, the conclusion occurs by opponents conceeding points until a general agreement is established. The conclusion is the agreement. It is exactly consensus-building. --Nealparr ^{(talk to me)} 21:10, 11 September 2008 (UTC)

I'll be honest here, whenever I find a pseudoscientific article, I rearrange the writing to remove verbiage like "consensus", "opinion", or "majority." Although I disagree with Philcha, not on the words, but on the implication, that "everything is open to debate". That is true, but many theories are at the 99.9999% level of non-debate (say Evolution). However, the word consensus always makes me think that a bunch of scientists standing around a podium, voting on a compromise wording. It doesn't work that way. I usually write something along the line of "current scientific research supports" or "indicates". It leaves the reasonable doubt (not a legal term in this case, but a scientific term), but really states scientific thinking. OrangeMarlin ^{Talk• Contributions} 17:31, 11 September 2008 (UTC)

(outdent) Rather than attempting to define "consensus" it is sufficient to simply let the peer reviewed literature speak for itself, which also solves the weight problem. Otherwise you face disputes over defining wheter a minority view is held by 49%, 9% or .9% of "experts" and how many words should be allowed to each view.

Respect the peer reviewed process and let the citations (and number of citations) themselves shape the content and weight of the article.

Take the classic argument for exluding discussion of "flat earth" theories. If the proponent of including such material can produce a citation to an peer reviewed and indexed journal of science, published within the last 50 years, wherein some researcher, say Smith, presented (a) evidence and or (b) a serious synthesis of others dats leading him to conclude that the earth was flat, I have no problem with the article stating the fact that "In 1958 Smith published a review of the literature which he proposed supported the theory that the earth was flat.¹"

If it exists, such a citation may actually be of considerable interest to people interested in knowing who and how people argue for a flat earth theory. The way to deal with such scientific nonsense is not to exclude it but to include it in exactly such a limited way (one sentence) and then follow it with as many sentences and citations as proponents of the "roundish" earth view as editors feel is necessary to demonstrate that the weight of scholarly opinion is clearly against the flat earth theory.

My view is that the more citations any article in Wikipedia has, the better. I'm an information and citation junky. I use Wikipedia as a tool to start my research, not finish it, and the bibliographies are more valuable than the article text. So I don't want to be told what the consensus view is, I want to see the citations and make my own judgment. The peer review process does not block all "junk science" but it does prevent the accumulation of a large bibliography of junk science. So let the purported "junk science" be cited, if it has at least risen to the level of being pubilshed in a peer reviewed (and indexed journal, if you want to have a quality control on the journals accepted as "reliable"). When such junk science is included it will always be limited by few if any citations and so is easily offset by a bibliography of "good science."

Finally, remember that this whole issue of "consensus" is raised by Wikipedia editors who are dealing not with non controversial issues, like the shape of the earth, but with controversial issues in which they are trying to minimize or even keep out "minority" views. I have no interest in supporting a policy that restricts the inclusion of information from peer reviewed sources, unless the publication or authors have actually retracted the findings. And I support including new findings, properly identified as new and not yet replicated, once published in a peer reviewed journal. It's not our job to predict acceptance or rejection of ideas or to balance them. Instead, we should respect the contribution of editors who have found a study they believe is of note and allow at least a minimal notation of it within relevant articles.--SaraNoon (talk) 18:35, 11 September 2008 (UTC)

As a case in point consider Plate tectonics and its predecessor Continental drift. The old theory retained widespread support even despite some troubling suggestions of unexplained phenomena like matching coastlines, but once the picture of magnetic striping on the spreading seafloor was published (yes—this took some doing to get past peer review) consensus shifted rapidly. Not that there was a vote taken, but papers built on the new model were successfully published. Soon afterwards papers on the old model ceased to be published.LeadSongDog (talk) 21:55, 11 September 2008 (UTC)

(EditConflict)

Nealparr states: "For a debate to be "won", there must be a conclusion. An open-ended debate doesn't have a winner." That seems to suggest that a judgement is pronounced or that the losing side publishes an admission of defeat -- something that can be cited. Often, in my admittedly limited experience, the losing side, or argument, just fades away. Newer papers, in the intro, state the "winning" conclusion as background stuff - and the work goes on. I suppose we could choose to call that a consensus, that being what Wikipedia is all about, but it is Wikipedia terminology.

Philcha's demand for "at least 20 years' critical examination" may be overdoing it a tad, at least in some fields or on some types of debate. (An example, one I've followed the last few years: When I went to school there were two and only two types of photosensitive cells in our retinas. Rods and cones. Full stop. Full consensus. The third type, ipRGCs, was accepted in much less than 20 years, though mechanisms and functions are still being discussed.) --Hordaland (talk) 22:21, 11 September 2008 (UTC)

Isn't there some quote about progress generally happening by old scientists dying rather than changing their minds... EverSince (talk) 22:47, 11 September 2008 (UTC)

@Hordaland. Well, yes, actually there is something that can be cited. I feel you guys are looking at it backwards. As I said above, Wikipedians are too obsessed with citations and demand direct quotes such as "There is a scientific consensus that the Earth is round [1]". The fact is that we cite scientific consensus all the time at Wikipedia in the form of "The Earth is round [1]" or "Humans have 46 chromosomes [1]". We need not say "There is a scientific consensus that humans have 46 chromosomes" but that's actually what it is. It's a consensus.

The losing side in a debate doesn't publish an admission of defeat, but the winner actually does publish an admission of victory in the form of textbooks, continued papers, and so on. As you said, the loser fades away. --Nealparr ^{(talk to me)} 23:12, 11 September 2008 (UTC)