Wikipedia:Use our own words



It may seem obvious that editors should choose their own words when writing articles. We have a long content guideline on plagiarism and another explanatory essay on close paraphrasing. And it is obvious and normal for editors to choose their own words, rather than lift them from our sources. It is quite normal for a copyeditor to revise article wording without even glancing at the sources. And yet when editors get into a dispute over word choice, someone may shout that we must WP:STICKTOSOURCES. They will do that because they believe that the cited sources (or most reliable potential sources) agree with their word preferences, and they assert that this policy compels us to do likewise. This misleadingly elevates that editor's opinion to one having the backing of policy or wide community consensus. It is a fallacy commonly employed to advocate for a conservative language position, sometimes expressed as the belief that "Wikipedia should follow, not lead" when it comes to language. This over-simplifies difficult editorial decisions by appearing to delegate word choices to the authors of our sources. While sources can guide us, along with style guides and publications aimed at audiences similar to ours, we alone are responsible for the words we write, and we should own that choice.

Policy
WP:STICKTOSOURCES links to the section in our No Original Research policy that deals with choosing and using sources, to avoid introducing facts, allegations, and ideas that are not present in reliable sources. The relevant linked paragraph is (with my emphasis): The best practice is to research the most reliable sources on the topic and summarize what they say in your own words, with each statement in the article being verifiable in a source that makes that statement explicitly. Source material should be carefully summarized or rephrased without changing its meaning or implication. Take care not to go beyond what the sources express or to use them in ways inconsistent with the intention of the source, such as using material out of context. In short, stick to the sources. Earlier in the lead of this policy it says: Despite the need for reliable sources, or violate their copyrights. Rewriting source material in your own words while retaining the substance is not considered original research. Policy requires us to rephrase or summarise our sources in our own words.

Deviance from sources
We write differently to our sources for many reasons, not just to avoid plagiarism or copyright violations.


 * Audience: Our sources are written for a different audience than Wikipedia. Often their audience are professionals in the field or workers in a particular sector. We cannot assume our audience has the same knowledge of technical terms and familiarity with jargon. We may choose "baby" rather than "foetus" or "womb" rather than "uterus". WP:TECHNICAL suggests that even for advanced topics, we write "one level down" from the level of knowledge and linguistic familiarity that our sources assume.
 * Point of view: Our sources may be advising their reader on matters where roles are different to our reader at home. For example, as WP:MEDMOS notes, our sources may refer to "patients" or "cases" or even a "cohort" whereas we would talk about "people" with a certain condition or disease or undergoing a treatment.
 * Neutrality: Our sources are not required to be neutral. They may indeed be highly partial politically advancing one point of view. This may be reflected in their word choices. MOS:WTW advises us on words to watch that are hard to use in a neutral manner.
 * Advocacy: Our sources may be recommending, advising, instructing and proposing something that the author believes should be done, or complaining about something that the author believes should not be done or done differently. For example, informing doctors what therapy to pick, or parents how to bring up their child.
 * Tone: We are an encyclopaedia and our tone differs from our sources. A dictionary is typically extremely terse and relies on cross referencing. A newspaper headline may be clickbait. A blog may be jovial. An academic work too dry.
 * English language variants: WP:ENGVAR notes that, somewhat unusually, we have no house style when it comes to regional variants of English. We do, however, insist on consistency within an article. The paracetamol article uses the international name but half the sources use the US name acetaminophen. Sometimes it is possible to choose a word or use a descriptive phrase that is clear in multiple English variants, such as autumn (rather than fall) or ground floor (rather than first floor).
 * Change happens: As a wiki, we can quickly adopt new words and terminology. When a virus is named, an actor announces a new name and gender, or a king accedes to the throne, we can update our articles even while the majority of our sources are stuck referring to the old name.
 * Evolution: The meaning of words can change over time, sometimes dramatically. It is a fallacy to examine the origins of a word or its historic use and consider those meanings must be applicable today. Our sources may use words in a way that our readers don't, or they would today regard as wrong or even offensive.
 * We have our own house style: Our WP:MOS describes the consensus of how we format, punctuate, capitalise, abbreviate, hyphenate, deal with currencies and measurements and a small number of word preferences that have community agreement.
 * Variation: Writers may substitute synonyms to avoid repeating a word. While this practice can sometimes make things worse, it is nevertheless common. Additionally, the subject, such as the president or an actor or a pop group or a drug, is not always referred to by name. Instead, we write they, it, them, those, he or she.
 * Jargon and specialist terms: While we may often avoid jargon and specialist terms if we can succinctly use everyday language, sometimes it is necessary to teach the reader terms that are essential for the subject. Thus, the article may at times use both specialist and everyday synonyms to help the reader.
 * Foreign language sources: Our sources are usually English but are not required to be. The article Venezuela has 50 sources in Spanish.

Exceptions
The principal exception to this is when we are not writing in our own voice, but quoting someone else's words. There are many reasons to do this but, with regard to word choice, quotations can be useful when the words would not make for good encyclopaedic text. For example, if the idea expressed is controversial, biased or offensive, uses archaic or outdated language, offensive words or inappropriate tone.

If we have indicated someone "stated", "said" or "wrote" something but put that something in our own words, we should stick to a fairly direct paraphrase.

Although public domain text (sources) can legally be copied word-for-word, this may still be plagiarism if done without appropriate attribution.

Motivation
What then is the motivation for asserting something that clearly isn't true in practice, and has no foundation in policy or guideline? The misciting of WP:STICKTOSOURCES most commonly occurs when an editor is advocating for a conservative language choice, and their opponents are proposing a progressive language choice. A variant on misciting WP:STICKTOSOURCES is to claim "Wikipedia should follow, not lead", which likewise is core policy on facts and viewpoints but not on word choice. Rather than having the solidity of established consensus policy, this belief about the language in our articles is a personal expression of conservative wishful thinking and no more.

There are many reasons why our sources can be more conservative in their language choices than contemporary modern English. Wikipedia is the encyclopaedia anyone can edit, with new editors and new ideas arriving all the time. Unlike journalists at a conservative newspaper, or researchers pitching their findings to a stuffy academic publication, or an embarrassingly out-of-date book from the 1980s, Wikipedia editors will naturally reflect how people across the English-speaking world write today. This is both good and inevitable.

Editors should debate what words we choose to use with honesty and respect for the variety of opinions that exist. We should own that decision rather than offload responsibility for suboptimal choices onto the authors of our sources. There are no policy shortcuts to finding consensus, instead we must work in good faith to find some way to cooperate, collaborate and compromise with others.