User:Dominic Mayers/Find and then present the points of view

An editorial process should aim to include all relevant information (describing debates, providing arguments, attributing opinions, etc.). This goes beyond an editorial process that only uses UNDUE, BALANCE, etc. and a neutral tone: it also requires actively finding all relevant informations and then present them. Finding relevant information is not a simple process, but it is part of the responsibilities of Wikipedians. Better explaining this will improve the NPOV policy.

The policy does not explain well what is a neutral point of view. Perhaps, it has never explained it. The fundamental idea of a neutral point of view is to present ideas and facts in such a fashion that both supporters and opponents can agree. A good way to understand this idea is to make an analogy with science. The goal of science is also to present knowledge that is as neutral as possible. This is why people with different interests and problems can agree on scientific knowledge. In science, it does not come easily. Patterns such the use of statistics, meta-analyses, Hilbert Spaces (in Quantum Mechanics) are used in science to achieve the goal. The NPOV policy also offers patterns such as the process of attribution that help achieve the neutral point of view: someone might not agree with "P," but yet will agree with "Joe said P". There are other patterns such as providing the arguments: someone might not agree with "P", but yet might agree with "the offered arguments for P are...". It is the responsibilities of the Wikipedians to use that kind of patterns to achieve the neutral point of view.

The neutral point of view outside Wikipedia

This section shows using examples that, outside Wikipedia, "adopting the neutral point of view" or "not taking sides" describes a process of research of in formation, not a simplistic process that is an afterthought once we face the outcome of that research of information.

Presenting views that are taking sides as a tool to develop critical thinking

McGraw-Hill says that each book in its series "Taking Sides" covers a controversial issue "in a debate-style forma designed to stimulate student interest and develop critical thinking skills". In each of these books, different clashing views are presented. In its review of one of the books, Taking Sides: Clashing views on African issues, Andrew Newsham wrote:

William Moseley, its editor, executes with aplomb the role of the helpful guide who introduces the protagonists in each controversy with respect and, by and large, with admirable impartiality. Having an editor who deliberately does not take sides nicely balances the yes/no structure, thereby also serving as a reminder to students (and academics) to understand opposing viewpoints before formulating their own opinions about them.

Alison McCartney, professor of Political Science at Towson University, says "these books are sometimes criticized for only offering either/or answers".

The process described by “not taking sides” is the one that results in an introduction of the protagonists by the editor. The presentation of the protagonists is comparable to a very detailed attribution by a Wikipedian. The Wikipedian must carry out a different editorial process, but one which also requires a complex search of relevant information, such as arguments to include. The strategy adopted during the process is important: “with respect and, overall, with admirable impartiality.” The goal is also important: to enable students “to understand opposing points of view before formulating their own opinions about them.” These strategy and goal are not very different from these of Wikipedia.

Not taking sides in court of law

Hanna Panreck of Fox News wrote "'Courts work because people trust judges. Taking sides in this way erodes that trust,' Ponsor, a senior judge on the U.S. District Court for the District of Massachusetts, continued, criticizing the Supreme Court." This extracts illustrates that not taking sides refers to the attitude of the judges. It is not about the final outcomes. This is clearer in the next extract.

The Nineteenth Judicial Circuit of Lake County Illinois says "A ruling by the judge does not mean he or she is taking sides." In this extract, it is clear that the details of the ruling is irrelevant: "not taking sides" describes the process that determines the ruling, not the ruling itself.

The lawyer Russell Alexander wrote : "Unfortunately, this is a misguided presumption: As a broad legal principle all judges are duty-bound to remain impartial as between the litigants that appear before them. So a judge cannot give the appearance of taking sides or actively advocating on one party’s behalf." The judge must even avoid the appearance of taking sides in his decision-making process. Not taking sides is used to describe proper ruling by a judge in many other sources.

Due weight in social and environmental policies

Geoffrey Hammond, Director of the interdisciplinary International Centre for the Environment at the University of Bath, wrote a correspondence piece in Nature about the fact that "Carbon footprint" is not the correct expression, because "footprint" means it is measured in area units whereas what is meant by "Carbon footprint" is measured in weight unit. The title of his correspondence piece was "Time to give due weight to the ‘carbon footprint’ issue". He was, of course, playing with words: the expected result of giving due weight, in that case, is the use of "carbon weight" instead of "carbon footprint". But, putting asides this play with words, "give due weight" in the title means "give more attention". This describes the process that leads to the decision (whether or not to use the term “carbon weight”), not the decision itself.

Due weight in court of law

Article 12 of the Convention on the Rights of the Child of the Office of the United Nations High Commissioner for Human Rights (OHCHR) says: "States Parties shall assure to the child who is capable of forming his or her own views the right to express those views freely in all matters affecting the child, the views of the child being given due weight in accordance with the age and maturity of the child." Again, we see that it is not a question of the decision taken, but of giving due weight to the child's opinion in the decision-making process.

Taking sides as do activists or by taking into account personal principles

This subsection, unlike the previous ones, is about the opposite of the neutral point of view, but as a good thing, at the least as a human thing. It's about being an activist or taking into account our own personal principles, even our emotions. Because it is often seen positively, it leads to question whether Wikipedia should take sides, but neutral knowledge must disregard individual emotions and principles. The examples are presented without any additional commentary.

Sherine Tadros, an Egyptian Coptic Christian, grew up comfortably in Britain after her parents prospered. She became an award-winning war correspondent and found fulfilment as an activist. The site GoodReads, in its review of her 2023 book "Taking Sides" wrote: "As a broadcast journalist for Sky News and Al Jazeera, Sherine Tadros was trained to tell only the facts, as dispassionately as possible. But how can you remain neutral when reporting from war zones, or witnessing brutal state repression?"

The 2003 movie "Taking sides" directed by István Szabó is about americans taking sides against an alleged nasist in a court case. The synopsis of the movie says: "Maj. Steve Arnold (Harvey Keitel) is charged with making an example of Furtwangler due to his status as a high-profile cultural figure, and pulls no punches in questioning him about possible ties to the Hitler regime. However, as Arnold presses forward, his assumptions about Furtwangler don't necessarily hold up."

"Taking sides" by Gary Soto is a 1991 juvenile fiction telling the story of a Hispanic boy, Lincoln Mendoza, who moved to a white suburban neighborhood. His basketball team at his new school, in which rich and mostly white students are enrolled, faces his old team. The description of the publisher, Blackstone Audio, for the 1999 audio version says "How can Linc play his best when he's shooting against his former teammates? To find an answer, Linc will need to sort through a maze of emotions and some tricky moves on the court."

Not taking sides and due weight in Wikipedia

Not taking sides has been a part of Wikipedia since its early times. Wikipedians are not activists and leave their personal principles and emotions aside. The important point about not taking sides is that it is about the editorial process, not about the space used by viewpoints in the article only. Presenting the facts, no matter how much space is attributed to these facts, does not mean we are taking sides, just like a judge that applies the laws is not taking sides. Conversely, not presenting some facts could mean that we are taking sides. In a correct editorial process, we include the relevant information.

Rejecting information does not mean that we are taking sides

In 2003, because the "include-info" aspect of not taking sides or neutral point of view policy could be misinterpreted and misused to include theories supported only by an extremely small minority, perhaps even nonsense theories, such as the flat earth theory, the no original research (NOR) policy was added to complement it. In 2005, the section "Undue and due weight ...", which is based on the 2003 statement of the NOR policy, was added to the Neutral Point of View (NPOV) policy. Just like a judge is not taking sides when he applies the laws, Wikipedia is not taking sides when it applies the NOR policy.

Including information as a way of not taking sides

Not all information is a theory to be rejected using NOR. In an article on Earth, one must give the information that "the Earth has the shape of an ellipsoid with a circumference of approximately 40,000 km." Not doing so would be to take sides or a refusal to accept one's responsibilities as a Wikipedian: not taking sides applies to a process and, if the process did not take place, it is because we refused our responsibilities as a Wikipedian. Including this information means accepting one's responsibilities as a Wikipedian while not taking sides.

Due weight outside Wikipedia and not taking sides in Wikipedia

The due weight notion used outside Wikipedia, say in court of law, goes well along the not taking sides principle. It means that the editors must give due weight, i.e., the required attention, to all pertinent view points in sources. As in the case of not taking sides, it is as much about including information as it is about excluding information. A lesson to be learned from the analogy with a legal context is that editors must simply read the sources carefully, giving them all the necessary attention in order to determine the proper place of each point of view.

The second meaning of due weight inside Wikipedia and its significance

Of course, english editors understand the usual meaning of "give due weight" and if they read "give due weight to all view points in sources", they understand that they must give proper attention to them, but in Wikipedia there is an automatic extension of the meaning to the outcome in the article: the editors also understand that the space given to the view points in the article must correspond to the importance of the view points in the sources. This is similar to what happened when Geoffrey Hammond played with the meaning of "weight" in his title (see above). In one meaning, the usual one, it refers to the attention given to the issue. In another meaning, it refers to the weight of carbon. In Wikipedia, the second meaning is the space given to the viewpoint in the article. However, Wikipedia is not playing with words, because the second meaning is dominant in the explanation of the policy, whereas the usual meaning, the attention given to the view points, is only there in the background. Undue weight in Wikipedia does not mean that too much importance has been given to the view point. It means that two much space was given to the view point in the article.

The dominance of the second meaning is highly significant. It makes a big difference, because the usual meaning refers to the attention given to view points in the editorial process, whereas the Wikipedia meaning refers to the final outcome in the article. In particular, because undue weight means too much space and due weight does not mean too little space, these expressions are only about rejection of information. In practice, in the text of the policy, these expressions say nothing about the editorial process, which, of course, is primarily about including information: the purpose of an encyclopedia is to provide information.

Principles to reject information are useful, but not enough

The reject-info principles are well known and they are useful in a search for information, because they limit the search space. They are the no original research (NOR) policy, the verifiability (V) policy and the reliable sources (RS) guidelines. The UNDUE rule (including BALANCE, etc.) can also be used to reject information, but only when we have the proposed article, because it refers to the relative space used by the information. We can add the requirement that what is included must be pertinent in its context, though this is implicit in the fact that the reliable sources depend on that context. These have been added between 2003 and 2006. There has been no attention on patterns of neutrality since 2003. Except for "attributing opinions", no pattern of neutrality has been considered. The general principle for these patterns "describing debates instead of engaging them" has received little attention and in fact some patterns and explanations have been removed.

Editors of Wikipedia are all interested in including information, because it is the goal of an encyclopedia. In that context, how is it that all the attention has been on rejecting info and on explaining how important this is for the neutral point of view ? Not that it is incorrect. Of course, we want to reject theories that have no place in Wikipedia. But how come there is not a similar interest in patterns that help achieve a neutral point of view, which seems to be equally important if not more important? The last question is considered in the next subsection.

The search for simplicity explains the emphasis on rejection of information

Rejection of information is simple: a statement is proposed for inclusion (or already seen in the article) and the editor uses NOR, V, RS or UNDUE (including BALANCE, etc.) to determine if it must be rejected. If the statement is not part of a normal understanding of sources of quality or if it violates UNDUE, it must be rejected. The correct way to add information is more complicated. To achieve the neutral point of view, it is not obvious what information needs to be added. The text of NPOV says or used to say that we should provide the arguments, explain what is at stake, etc., but doing this well in a neutral way is not easy. For example, providing factual information about a point of view can respect the neutral point of view, even if this additional information calls the point of view into question. Even the way to attribute opinions without taking sides is not obvious. Some have argued that explanations could be misinterpreted and used to include theories that have no place in Wikipedia. Explanations have been removed. Currently, we have almost no such explanations in the NPOV text. This last consideration was one of the motivations for writing this essay.

Complexity becomes a much more important factor when combined with another factor: the editor's goal. If the editor's goal is to include minority view points in Wikipedia, he will tend to like rules for inclusion and dislike rules for rejection. It will be the opposite for an editor that wish mainly to remove theories that have no place in Wikipedia. The viewpoints of a majority are not much affected by inclusion or rejection rules, so they are ignored in the current analysis. The goal of an editor should be balanced. One should not focalize on one aspect, rejection of unacceptable theories or inclusion of view points of minorities, and, as a result, see opposition between rejection and inclusion. One should see instead that inclusion and rejection of information are working together toward a same goal. In this way, a balanced explanation of the role of adding and rejecting information will make its way in the policy. It is possible that some editors, due to their situation in their personal lives and on Wikipedia, will always maintain as their sole objective to reject unacceptable theories, which is entirely laudable. Unfortunately, too absorbed in this sole objective, they will also oppose a rational approach in which all relevant points of view have their rightful place and are presented in a factual manner without omitting the information necessary for their proper understanding.

The complexity factor is linked to another factor: the possibility of sanctioning pov-pushers. If the process is too complex, it is more difficult to justify sanctions: to sanction one must be able to say that the Wikipedian clearly sought to violate the rules of a correct editorial process.

The simple but vague principle of reliable sources offers little direction to guide a consensus

The requirement for verifiability in reliable sources, which is the concrete implementation of NOR, is a vague requirement in the following sense that it depends on what we mean by reliable sources. This vagueness is reinforced by the fact that what is meant by a reliable source depends on the topic. It is even further reinforced by the fact that the reliable sources for a topic are determined by consensus. The only direction offered to reject or not information is the vague notion of reliability and a consensus can interpret it in many ways. This creates an issue, because a minority has nothing to rely upon, except this vague notion of reliability, to build an argument in the process that determines the consensus. This issue is well illustrated by the recurrent debates regarding WP:NOTNEWS. Many people would like that Wikipedians avoid reporting current events, but the majority insists that newspapers can be reliable sources and the current NPOV policy provides no further directions. We see here that an overly simplistic way of rejecting information which itself is largely determined by consensus is a two-edged sword.

The challenge

The issue we face is that the general principle "describing debates instead of engaging them" is not clearly explained. The "not taking sides" and the "due weight" principles are too abstract and they are as much about rejecting info as they are about including info. The explanations must cover as many kinds of patterns as possible: attribution, arguments, historical context, what is at stake, etc. Adopting the neutral point of view may require thinking out of the box rather than simply deciding to include or reject a fixed statement.

The difficulty is not in finding examples to illustrate the role of adding different kinds of information using patterns. The difficulty is that the very goal, i.e., the problems to solve in Wikipedia, is not shared. Communication is easy, examples easily appreciated and understood, when the goal is shared. Unfortunately, Wikipedians have different domains of interest in Wikipedia and they have different viewpoints about what is valid knowledge in whatever their domains of interest are. They see entirely different problems in what is going on in Wikipedia.

A difficulty often mentioned is that anybody can edit Wikipedia and, thus, we do not have the competence required to do complex search in reliable sources. This suggests that a simple editorial process, ideally restricted to the few best secondary sources on the subject, is the only way to avoid biases, which is referred to as POV pushing. This request for a simple editorial process can take other forms. For example, for some reason, WP:SYNTH states "If one reliable source says A and another reliable source says B, do not join A and B together to imply a conclusion C not mentioned by either of the sources", which is logically problematic. For example, the logical conclusion A $\wedge$ B can always be drawn and it will not usually be found in either of the two sources. More generally, conclusions that are included in the natural understanding of A and B in the two sources might not be found in either of the two sources separately and there is certainly nothing wrong that the same conclusions are included in the understanding of A and B in the article. It is rather the opposite that would be wrong. There is a very simple solution to this difficulty: simply replace "either of the sources" with "the sources". Unfortunately, there are oppositions to this solution, because it suggests a more complex editorial process in which we must understand the two sources together. For those who like logical and simple rules, this particular way of insisting for a simple editorial process is very annoying. Another form taken by this request for a simple editorial process is the suggestion that one can determine which viewpoints to include and how much space to allocate to them in the article by a simple method to analyse the sources, which would be more systematic and less biased than the usual reading and understanding of these sources. There are wikipedians that want to believe in such a magical method, because they are not comfortable with the natural complexity of the needed editorial process to find the relevant information to include. If a content is presented and the goal is to evaluate the importance of this particular content in the article, it is different, but even in that case there is no magical method, because judgment is needed to weight the reliability of the sources and other factors. The worst form of this request for a simple editorial process is the claim that UNDUE, BALANCE, etc. together with a neutral tone is sufficient to describe the NPOV policy. This hides the complexity, because these rules apply to already proposed articles.

When an example is proposed, often the antagonistic response is simply to show, relying on this superficial simplicity, that DUE or UNDUE allows to reject or to include the information. This suggests that the text of the current rule is sufficient, but the issue with that logic is that the rule applies to the final outcome only and says little about the editorial process per se. The difficulty is not finding examples, but the fact that the objective of improving the editorial process itself, by giving examples of addition of information, is not shared.

The Wikipedia's concept of weight is not an issue as long as we accept that its evaluation requires and understanding of the sources as a part of complex editorial process. The challenge is to find an explanation of NPOV that unify the rational people who can possibly be unified: an explanation that considers, not only DUE, UNDUE, BALANCE, etc. on the final outcome, but also the different kinds of information that must be added in the editorial process under the guidance of patterns.

The case of general articles in philosophy

In the French Wikipedia, there is currently a debate whether Épistémologie should be a general article around the word. At the same time, in the English Wikipedia there is a suggestion that perhaps Epistemology should become a general article that covers all the diverse aspects related to the word. It is natural to ask whether some general patterns of neutrality similar to attribution can be helpful to address this issue. After all, the goal of neutrality is to present knowledge in a way that every one can agree.

This question has likely more than one answer, but to get the ball rolling, I propose one. The idea is that we should only join together in an article contents that have the proper context in the sources. This does not mean that the sources must be entirely centred on the topic of the article. For example, if we have an article about a person and we need to include the person's birth place, it can be found in a reliable database about famous persons. The database is not at all centred on that specific person, but it has for exact purpose to fulfill our need in the context. In contrast, if I consider the word "epistemology" in the post Gettier context and the use of "epistemology" by Karl Popper in his book "An epistemology without a knowing subject", then the shared context is small. Post Gettier epistemology almost completely ignores Karl Popper. On his side, Karl Popper reacted to the the notion of a belief by a knowing subject, which is central in post Gettier epistemology, but it's more that Karl Popper said he was not interested in that way of thinking. So, in accordance, with the proposed principle, we should not try to cover even briefly Popper's epistemology in an article that covers post Gettier epistemology. We can mention what Popper said about a knowing subject if it happens to be really and honestly relevant to explain post Gettier epistemology and put it in perspective, but this is different from trying to cover Popper's epistemology.

In a schematic way, the principle says that we should avoid an article that looks like: "In a context, authors A, B, C, ... have discussed this and that and have different opinions, etc. In an unrelated context, authors U, V, W, ... had a different discussion and didn't think much about what authors A, B, C said.

This principle does not require that the authors explicitly communicated with each others. The case of a mathematical theorem helps explain this last point. If two sources are about the same mathematical theorem, there is a clear non ambiguous context of communication, even if the authors do not refer to each others. In contrast, Popper was not interested in a knowing subject: there was no obvious context of communication.

The motivation for the pattern or principle proposed here is that it becomes easier to agree on the proportion to give to each view point in the article. This is the idea of a pattern of neutral point of view: it allows people to agree on a neutral article. Without this, there is a huge risk that the proportions will be subjective. For example, only a small section on the post Gettier epistemology in the french article Épistémologie, while this subject is the main content in the English article Epistemology. If the article is clearly on say the French approach to epistemology, then people can agree that, in the context of that approach, there is only a small mention of the post Gettier epistemology to helps putting the approach in perspective. The same kind of idea is at work here than in the case of attribution: when we take a neutral perspective people with different points of view are more likely to agree.

It is a big challenge to agree on the meaning of “a clear and unambiguous communication context”. But, we can hope that when it is really clear and unambiguous, consensus will be reached easily. Regarding this challenge, the principle proposed here should not be confused with the idea of restricting oneself to a few reliable secondary sources of quality that cover the subject. This other principle suggests a certain simplicity, but we explain in this essay that this attempt to simplify the process can go against the goal. So, we must accept that determining whether there is a communication context requires a good understanding of the subject. Even in the case of a mathematical theorem, the same theorem can have different formulations which are obviously equivalent for the mathematicians only. There is no doubt that the challenge is big. However, this challenge seems intrinsic to the goal and trying to avoid it with simplistic solutions amounts to burying our heads in the sand.

Some people might complain that this principle invites to create a POV fork. This is not a POV fork, because the POVs are not connected in the sources. Only Wikipedians want to connect them, because they use the word "epistemology".

It still remains that it would be useful to have a list of the articles that are on a topic that is or could be identified with the word "epistemology". However, it is very important that this list is not really an article, because that will violate the principle that is discussed here. It has to be no more than a simple list of articles in Wikipedia, a list that is useful, even though perhaps boring.