Talk:Premise

I'm thinking about implementing changes to this article with the hope of moving it in the direction of GA status. With a mere 550 words readable prose size, the current version leaves out many key topics. It only discusses deductive reasoning, but premises are equally relevant to non-deductive reasoning (e.g. inductive, abductive, and analogical reasoning). There are various types of premises that should get more attention, such as explicit vs implicit, independent vs dependent, and first vs intermediary. Another key topic is premise adequacy or under what conditions a set of premises actually provides support for a conclusion. An overlapping issue to explain concerns fallacies where the error lies with the premises, such as ad hominem, begging the question, red herring, and false dichotomy. Premises are relevant to various fields, including logic, argumentation theory, philosophy, mathematics, science, law, and critical thinking, so having a section to cover this would be good. Sources often discuss the problem of how to identify premises in natural language arguments, such as expressions that act as premise-indicators (e.g. since, because, given that, as indicated by,...), so this should be covered somewhere. It might be good to start the article with a definition section so the lead section can summarize the body of the article rather than present new information.

Sourcing is another point that needs to be addressed since various passages currently have no references. There are also problematic claims in need of fixing. For example, the paragraph starting with "An argument is meaningful..." makes the confused claim that true premises are about meaning instead of saying that they are about epistemic support. The paragraph later claims that denying a conclusion because of a false premise is the fallacy of denying the antecedent. This is not correct: denying the antecedent requires a conditional premise (if ... then), not just any denial of a conclusion. I'm not sure why the article cites Euclid when discussing the nature of syllogism; this topic is primarily associated with Aristotle. The sentence "A premise can also be an indicator word..." is problematic since a premise is a statement, not a word.

There are more issues to consider, but they can be addressed later since the points mentioned so far will already involve a lot of work to implement. I was hoping to get some feedback on these ideas and possibly other suggestions. Phlsph7 (talk) 09:19, 16 May 2026 (UTC)

@Phisph7 Thanks for your contribution: it is a big improvement. I think I can make a few further improvements, particularly to the lede, which still focuses too much on deductive arguments. I don't want to tread on your changes, so perhaps you could comment below when you are done for now, then I'll make some changes and we can go from there if anything else needs to be done. Dezaxa (talk) 10:11, 21 May 2026 (UTC)

Hi Dezaxa and thanks for your input! I agree, the lead needs various changes. Generally speaking, the lead should summarize the main points of the body of the article. This is difficult to do if there are still major parts of the body missing. I usually work on the body first and fix the lead as the last step. There are still various topics to be covered in the body of the article, such as a section on the role of premises in deductive and non-deductive reasoning. The different expansions will probably take me another week or two. I'll ping you once the major revisions are finished to get your feedback and to decide what further adjustments are needed. Phlsph7 (talk) 12:44, 21 May 2026 (UTC)

Hey Dezaxa, I've implemented most of the proposed changed and I'm planning to nominate the article for GA status. I would be interested to hear if you have more improvement suggestions. Phlsph7 (talk) 17:16, 26 May 2026 (UTC)

Phisph7 Thanks again. Here are my comments:

1. The article is quite long. I wonder whether it really needs the sections on Role in Arguments and Fallacies. Those are covered in other articles. I'd like to have someone else comment on that.

I appreciate the input and the reformulations. The article currently has a readable prose size of 3280 words, which is not particularly long relative to Wikipedia policy and GA standards. According to WP:SIZERULE, articles below 6000 words should not be trimmed just because of their length. The section "Role in Arguments" covers premise adequacy and premise relevance among other things. I'm not sure that any other article is better suited to discussing these specific topics. Blair 2012 has separate chapters explicitly dedicated to "Premissary Relevance" and "Premise Adequacy", so we are just following the weight given to these subjects by reliable sources per WP:PROPORTION. Similarly, regarding the section "Fallacies", the fact that premises can cause bad reasoning seems to be a key aspect of the topic. I'm pretty sure that GA reviewers would complain if the article did not discuss premise-related fallacies since articles should be broad in coverage and discuss all the main aspects. I think the better approach here is to look for specific passages that are not relevant rather than simply ignore topics that reliable sources deem central. Phlsph7 (talk) 12:15, 28 May 2026 (UTC)

I don't question that the subjects in the sections on Role in arguments and Fallacies are important, it's just that those topics have their own articles in WP, and it is WP policy not to duplicate material. Dezaxa (talk) 19:56, 28 May 2026 (UTC)

I'm not sure that there is such a policy. Maybe you are thinking of WP:CFORK, which says that two different articles should not have exactly the same scope. However, it does not apply here, since we have articles with different scopes where there is merely some overlap. Generally speaking, fallacies are relevant to many articles, such as Argument, Deductive reasoning, and the like. A policy that does not allow any article other than the article Fallacy to discuss fallacies would be rather odd. There is even a template for including text from one article in another. Allowing the use of such a template would be strange if overlaps were prohibited. Phlsph7 (talk) 08:26, 29 May 2026 (UTC)

I'm thinking of the Wikipedia:Summary_style policy. This states: "A fuller treatment of any major subtopic should go in a separate article of its own." Dezaxa (talk) 10:53, 29 May 2026 (UTC)

The passage you quoted talks of a fuller treatment, not of a short summary. Wikipedia:Summary_style also says The original article should contain a section with a summary of the subtopic's article. That's what we are doing. Phlsph7 (talk) 08:22, 30 May 2026 (UTC)

2. Strictly speaking, premises do not have to be relevant to the conclusion. In what we now call classical logic, premises do not have to be used. See for example Monotonicity_of_entailment. Also, what counts as relevant? Relevance is highly context sensitive. To speak of relevance suggests we are talking about Relevance_logic which is a specific kind of logic.

The purpose of premises is to be relevant. Otherwise, they are just dead weight. Blair 2012 says that In arguments so made, the grounds ... called “premises” ... are supposed to lend support to the viewpoints or conclusion ... and continues to define premissary relevance as this form of lending support. Talking about premises as relevant or irrelevant is standard in the academic discourse on premises. For example, a major category of fallacies is called "fallacies of relevance", defined as arguments with irrelevant premises. In this sense, you are right that there are arguments with irrelevant premises: they are fallacies. In any case, relevance is not some fringe issue limited to relevance logic. Phlsph7 (talk) 12:15, 28 May 2026 (UTC)

I know some books talk about relevance, but it isn't the best term. Redundant premises should be avoided, because they are a waste of space and could be misleading. But adding a redundant premise to a valid argument does not make the argument invalid, in classical logic at least. Dezaxa (talk) 19:56, 28 May 2026 (UTC)

This is probably an issue regarding personal terminological preferences. I tried to follow your suggestion by rewording some uses of the term. However, it is commonly used in sources, so it makes sense for our article to introduce it as well. Phlsph7 (talk) 08:26, 29 May 2026 (UTC)

3. Premises don't have to be accepted for an argument to be successful: they could just be suppositions that we allow "for the sake of argument".

I think it's a good idea to hint toward arguendo. I adjusted some of your reformulations to preserve the idea that justification is transferred from premises to conclusion without implying that the premises are necessarily accepted. I hope that this addresses your concern. From Blair 2012: When we invoke a warrant, one way to conceptualize what we are doing is to see it as claiming that the epistemic or dialectical status of the premises (their believability or acceptability) will transfer to the conclusion. From van Eemeren et al. 2014: An argument scheme is an abstract characterization of the way in which in a particular type of argumentation a premise used in support of a standpoint is related to that standpoint in order to bring about a transfer of acceptance from that premise to the standpoint. Phlsph7 (talk) 12:15, 28 May 2026 (UTC)

I don't like the expression "transferring truth". In classical logic a valid argument is often said to "preserve truth" in the sense that true premises guarantee a true conclusion. In Bayesian probability, a hypothesis is supported by evidence when conditioning on it increases the probability of the hypothesis. This terminology is better than 'transferring'. I prefer the terms 'grounds', 'reasons', 'evidence' or 'support' rather than 'transfer'. Dezaxa (talk) 19:57, 28 May 2026 (UTC)

Generally speaking, I tend to agree. However, I think the transfer-idea should be mentioned somewhere. I reformulated the text so we don't speak of transferring truth. Phlsph7 (talk) 08:26, 29 May 2026 (UTC)

4. Arguments don't have a minimum number of premises. I have come across texts that speak of an argument with no premises, e.g. to say that a logical truth may be thought of as the conclusion of a valid argument with no premises. An argument may just use the rules and axioms of the underlying logic without needing any premises. In which case, the argument is valid if and only if its conclusion is a valid sentence or logical truth. That is why I prefer the wording that an argument consists of a set of premises and a conclusion. Sets can be empty. Dezaxa (talk) 00:38, 28 May 2026 (UTC)

From Copi 2019: An inference, from one or more propositions to an inferred proposition, is an argument ... Thus, in the technical sense of logic, an argument consists of one or more premises and a conclusion inferred from the premise(s). From Dowden 2020: Any argument must have one or more premises. I'm not sure that simply writing down an axiom is considered an argument. However, the expression "a set of premises" also works for me. Phlsph7 (talk) 12:15, 28 May 2026 (UTC)

I have seen texts say that one premise is a minimum and also texts that say two is a minimum. An argument with no premises is a degenerate case, but it is no more weird than accepting that the empty set is a set, or that zero is a number. Dezaxa (talk) 19:57, 28 May 2026 (UTC)

Nominator: Phlsph7 (talk · contribs) 08:41, 29 May 2026 (UTC)

GA review

This review is transcluded from Talk:Premise/GA1. The edit link for this section can be used to add comments to the review.

Reviewer: Z. Patterson (talk · contribs) 01:19, 14 June 2026 (UTC)

I will be reviewing the article. Z. Patterson (talk) 01:19, 14 June 2026 (UTC)

Subject is clearly notable, based on what I find in searches. Article has no copyright violations. No recent edit warring occurred. Images are either in the public domain or are licensed under Creative Commons, thus having appropriate licenses. Images are also appropriately used. Therefore, so far, the following criteria are met.

• Neutral: it represents viewpoints fairly and without editorial bias, giving due weight to each
• Stable: it does not change significantly from day to day because of an ongoing edit war or content dispute
• Illustrated, if possible, by media such as images, video, or audio
  • media are tagged with their copyright statuses, and valid non-free use rationales are provided for non-free content
  • media are relevant to the topic, and have suitable captions

A more detailed review will be forthcoming. Z. Patterson (talk) 01:31, 14 June 2026 (UTC)

Hello Z. Patterson and thanks for reviewing this nomination. Phlsph7 (talk) 08:22, 14 June 2026 (UTC)

@Phlsph7: I have some points to address.

• The Dictionary of Philosophical Logic lists premisse as an alternative spelling.^[1] Would you mind including that in the lead's footnote?

  Added. Phlsph7 (talk) 10:24, 15 June 2026 (UTC)

• When I attempt to go to the PDF version of Logical Reasoning by Bradley Dowden, I get a 404 error. Can you provide a different link or remove it?

  Fixed. Phlsph7 (talk) 10:24, 15 June 2026 (UTC)

• Can you quote premises that appear in sources that the article cites?

  Most of the examples are from sources, some verbatim and others modified. For example, the Luis example is verbatim from Copi, Cohen & Flage 2016 and the example "she is not at home since she has gone to the shop" is a variation of Dowden 2020 ("She’s not here, since she’s gone to the supermarket"). Phlsph7 (talk) 10:24, 15 June 2026 (UTC)

• If an argument is in quotations, I would think you would not end the quotation with a period if the sentence continues. For example, the following sentence: For instance, the argument "This song will either be an enormous success or an utter failure. It will not be an enormous success. Therefore, it will be an utter failure." has two premises: one states that there are only two options, and the other rules out one option, leaving the other option as the only candidate. Another example is For instance, the argument "This song will either be an enormous success or an utter failure. It will not be an enormous success. Therefore, it will be an utter failure." has two premises: one states that there are only two options, and the other rules out one option, leaving the other option as the only candidate.

  I'm not sure about the right approach here since examples spanning several sentences look odd if there are periods in the middle but not the end. I implemented the suggestion, have a look if you think this is an improvement. Phlsph7 (talk) 10:24, 15 June 2026 (UTC)

• In the first paragraph in the Role in arguments section, I do not see a mention of empirical sciences in the source cited.

  From Smith 2020, p. 2: The premisses of arguments can be about all sorts of topics: their truth is usually no business of the logician. If we are arguing about historical matters, then it is the historian who is the expert about the truth of our premisses; if we are arguing about some question in physics, then the physicist is the one who might know whether our premisses are true; and so on. I slightly reformulated the passage to reflect the source text more closely. Phlsph7 (talk) 10:24, 15 June 2026 (UTC)

• The quote, "the sun will rise tomorrow because it has risen every day in the past", does not appear in the text cited. I suggest turning this into a paraphrase.

  Done. Generally speaking, the inverted commas here are used separate arguments from the regular text describing them, so they do not mean that a source is quoted. We could use apostrophes instead of inverted commas, but I don't think that this is standard. This is also how reliable sources use inverted commas. For example, Hurley 2016 uses them in For example, if the premise “One unripe apple that had been found earlier was removed” were added... without implying that any particular person is being quoted here. Phlsph7 (talk) 10:24, 15 June 2026 (UTC)

• I suggest rewording this sentence: This pattern appears in the argument "All ravens I observed are black. Therefore, all ravens are black." You may reword it as the following or something similar: This pattern appears in the argument that if every raven observed is black, "All ravens are black". The quote "All ravens are black"^[2] appears in one of the sources this sentence sites, but not the quote, "All ravens I observed are black. Therefore, all ravens are black."

  I implemented a slight variation. Phlsph7 (talk) 10:24, 15 June 2026 (UTC)

• I do not see a mention of itchy rashes leading doctors to conclude chickenpox developed in any of the three sources the third paragraph of the Non-deductive reasoning subsection. Please provide an example from one of the three sources that this cites, or provide a fourth source.

  Added. Phlsph7 (talk) 10:24, 15 June 2026 (UTC)

• Are you able to directly quote Section 2.2 of the following source: Bartha, Paul (2024). "Analogy and Analogical Reasoning". The Stanford Encyclopedia of Philosophy. Metaphysics Research Lab, Stanford University. Retrieved 23 May 2026.

  The whole section is too long to quote here. However, it is available online at https://plato.stanford.edu/entries/reasoning-analogy/ . Phlsph7 (talk) 10:24, 15 June 2026 (UTC)

• I see no mention about the premises about rats, humans, and birth control in any of the sources I could access that cite the paragraph these premises appear in.

  From Salmon 2013: For example, analogical reasoning from the observed effects of the birth-control hormone on rats (which are physiologically similar to humans) has been used to conclude that birth-control pills may affect human brain development ... In the argument that concludes that birth-control pills may affect human brain development, the properties F, G, H, and so on—although unspecified in the premisses—are the physiological properties that humans (objects of type X) and rats (objects of type Y) have been observed to share. Rats are used in medical experiments because they possess these properties. That the further property Z of brain development is affected by the birthcontrol hormone has been observed for rats and is inferred to hold for humans as well.

• Can Footnote j be paraphrased instead of quoted? I wonder if we can say, "For instance, asking whether someone stopped smoking cannabis assumes that the respondent previously smoked cannabis".

  I implemented a slight variation. I found the previous version more accessible, but this works also. Phlsph7 (talk) 10:24, 15 June 2026 (UTC)

• I do not have a copy of the following book or access to this book through the Wikipedia Library, Google Books, or Internet Archive: Hurley, Patrick J. (2016). Logic: The Essentials. Cengage Learning. ISBN 978-1-305-07092-9. If I try to access it on Internet Archive, I get a "Borrow Unavailable" button. Therefore, I cannot confirm if the quotes in the article that cite this book come from this book. Can you check if the quotes appear, or if they do not appear, change them to quotes that appear in the book, or paraphrase them?

  I don't think that we use any direct quotes from Hurley 2016. For example, the soldier example is from Hurley 2016 p. 190 with some minimal changes: "All soldiers are patriots. No traitors are patriots. Therefore, no traitors are soldiers." The howitzer example is a simplified version of "The U.S. military has fewer battleships, M1 rifles, and horse-drawn howitzers today than it did in 1940. Therefore, the U.S. military is a less effective fighting force today than it was in 1940." Phlsph7 (talk) 10:24, 15 June 2026 (UTC)

• In the last paragraph in Fallacies section, could you input a direct quote from the following source: Walton, Douglas N. (1995). A Pragmatic Theory of Fallacy. University of Alabama Press. ISBN 978-0-8173-0798-1. I see that the last sentence of this is "Therefore feathers are contrary to darkness."^[3] Be sure to put a citation right after the quote or at the end of the sentence this quote appears in.

  As above, the inverted commas are used to separate the argument from the text describing it and not to indicate a direct quotation. The text from the source is "Light is contrary to darkness. Feathers are light. Therefore feathers are contrary to darkness." Our version is a slight reformulation to make it more accessible. I could add the original quote to the source, but I don't think that this is particularly important or helpful. Phlsph7 (talk) 10:24, 15 June 2026 (UTC)

• I cannot find this quote in either source the paragraph this quote appears in cites: "If a party breaks a contract, they are liable for damages. The defendant broke a contract. Therefore, they are liable for damages." Are you able to try to input a direct quote from either source or paraphrase what you find in either source?

  Again, this is presented as an argument, not quote. The original quote is much longer: If one party has a contracted to carry out an act for another party, but has then failed to do it, she is liable to the other party for breach of contract. The defendant contracted to carry out a certain act for the plaintiff, but then failed to do it. The defendant is liable to the plaintiff for breach of contract. I think for our purposes of showcasing the basic idea, the simplified version does a better job and it is supported by the sources. Phlsph7 (talk) 10:24, 15 June 2026 (UTC)

Otherwise, I find that the article meets the following criteria.

• Well-written. The article certainly meets this criterion.
  • the prose is clear, concise, and understandable to an appropriately broad audience; spelling and grammar are correct
  • it complies with the Manual of Style guidelines for lead sections, layout, words to watch, fiction, and list incorporation
• Sources used are exemplary for a GAN.
  • it contains a list of all references (sources of information), presented in accordance with the layout style guideline
  • reliable sources are cited inline. All content that could reasonably be challenged, except for plot summaries and that which summarizes cited content elsewhere in the article, must be cited no later than the end of the paragraph (or line if the content is not in prose)
  • it contains no copyright violations or plagiarism
• Broad in its coverage. The article is certainly broad.
  • it addresses the main aspects of the topic
  • it stays focused on the topic without going into unnecessary detail (see summary style)
• Neutral: it represents viewpoints fairly and without editorial bias, giving due weight to each, as stated previously.
• Stable: it does not change significantly from day to day because of an ongoing edit war or content dispute, as stated previously.
• Illustrated, if possible, by media such as images, video, or audio, as stated previously.
  • media are tagged with their copyright statuses, and valid non-free use rationales are provided for non-free content, as stated previously.
  • media are relevant to the topic, and have suitable captions, as stated previously.

I will give you seven days to address these points.

References

Z. Patterson (talk) 23:35, 14 June 2026 (UTC)

Thanks for the comments. My responses are above, I hope I got everything. Phlsph7 (talk) 10:24, 15 June 2026 (UTC)

@Phlsph7: There also appears to be no original research. I am passing this article. Z. Patterson (talk) 01:27, 16 June 2026 (UTC)