Wikipedia talk:Manual of Style/Dates and numbers

Do counts of of different, but related things count as comparable numbers for these purposes? In particular, I'm thinking of when death and injury counts are given for disasters. Would these be comparable or these purposes? Or would that only apply if we were comparing, for instance, death tolls from different events/locations? For instance:

• "The attack killed 3 people and injured 26."
• "The attack killed three people and injured twenty-six."

Or would it be:

• "The attack killed three people and injured 26."

I haven't really seen it done consistently either way. TornadoLGS (talk) 03:48, 17 December 2025 (UTC)

Opinions will likely differ, and I suspect this is left to editor consensus per article. My personal preference is for the first of your 3 bullets, but I consider the second acceptable if that is preferred by others. The third looks awful, and unnecessarily hard for the reader to parse. Dondervogel 2 (talk) 09:42, 17 December 2025 (UTC)

• In my view this is a very silly rule, which I ignore: single digit normally spelled out (unless with an ISO unit, etc); two or more digits numerals. Tony (talk) 10:39, 17 December 2025 (UTC)

Deaths and injuries are usually comparable things, often added together as a single metric, 'Killed or seriously injured' (|KSI). In general I'd use one of your first two examples, but if the numbers were grossly dissimilar, it would be strange to impose consistency ("Only one person was killed but 2,173 were badly injured" is better than either alternative). It would then be more a matter of writing good English, rather than of Wikipedia's style choices, and I'd hope we wouldn't need WP:NUMNOTES to dive into differentiating between comparable things and comparable magnitudes. NebY (talk) 11:16, 17 December 2025 (UTC)

Yeah, I've even seen it applied differently within the same paragraph (actually edited from one I added): ... killing four people and injuring 88... But a few lines down: ... heavily damaging 30 government and residential buildings ... 5 of which were destroyed. TornadoLGS (talk) 17:57, 17 December 2025 (UTC)

Three vs. 26 may still be a borderline case but i think with four vs. 88 it's very clear that we are NOT talking about "comparable numbers" anymore, hence writing them differently is very reasonable. You can compare anything if you really want to, but I would understand "comparable" in this context to mean fairly close to each other, which is not the case when they differ by an order of magnitude or more. Gawaon (talk) 02:42, 18 December 2025 (UTC)

Okay, I think I had a misunderstanding in that "comparable" in this sense meant the things being counted were similar in nature. TornadoLGS (talk) 02:59, 18 December 2025 (UTC)

My interpretation of "comparable" in Comparable values near one another should be all spelled out or all in figures, even if one of the numbers would normally be written differently: patients' ages were five, seven, and thirty-two or ages were 5, 7, and 32, but not ages were five, seven, and 32 is comparable in nature, not in value. Dondervogel 2 (talk) 18:47, 18 December 2025 (UTC)

In that case there is some ambiguity that should be cleared up. TornadoLGS (talk) 18:49, 18 December 2025 (UTC)

Do we need to? The examples you quoted yesterday are fine. Gawaon's right to point to one pair differing by an order of magnitude or more but what would be the significant benefit of determining and laying down a precise boundary? NebY (talk) 19:16, 18 December 2025 (UTC)

I think you are missing Tornado's point. Some of us are interpreting "comparable" as referring to the nature of the quantities, while others interpret the same word to refer to their value. Dondervogel 2 (talk) 20:43, 18 December 2025 (UTC)

TornadoGLS's initial examples are pairs that are comparable in both senses of quantity - that which is measured, and the amount of it. A later example was of strikingly dissimilar amounts, perhaps another one would be of strikingly dissimilar things. The word "comparable" serves us well in both cases. However, TornadoGLS's initial query was explicitly in terms of different, but related things. It's true that your initial responses didn't address that, but does that mean the MOS has an ambiguity that needs clearing up? NebY (talk) 20:50, 18 December 2025 (UTC)

Dondervoge is right about the point I'm making. I gave an example that was comparable in nature and quantity. But this discussion raised a second matter: the sense of "comparable" is ambiguous as to how it should be if the values are comparable in nature but not in quantity, as in the example of four deaths and 88 injuries. TornadoLGS (talk) 21:11, 18 December 2025 (UTC)

• Comparable absolutely means comparable in nature, not in magnitude. I've expanded the range of values in the "ages" example to help make that clear . (I considered changing the example to "5 officers, 32 enlisted men, and 842 civilians" to make the range even wider but I was reminded (by a hidden note I placed in the wikitext twelve years ago!) that I had chosen to use ages as the example exactly because questions about how represent ages come up over and over.) EEng 01:40, 2 January 2026 (UTC)

  That idea sounds neat in theory, but doesn't always work in practice. I recently came across this sentence in a book: "Only one in all these three groups was born before 1800, thirty-one were born in the first half of the nineteenth century, and 162 in the second half." Per our usual rules, we would more likely write the middle number as 31, but should we seriously suggest writing either 1 or one hundred sixty-two in this case? I for one certainly wouldn't, and I consider it best to accept that while nature matters, magnitude can't be totally neglected either. Gawaon (talk) 09:25, 14 January 2026 (UTC)

  I can see why they wrote out 'thirty-one', as it avoids confusion with the date immediately preceding it. To me that sample passage reads OK. Skeptic2 (talk) 10:33, 14 January 2026 (UTC)

I am working on James Thurber, and have found myself using a lot of season names (as the sources do). I've been through and converted some to "mid-year" and so on, but there are some usages that I would like to keep, but which are not called out as exceptions by MOS:SEASON.

• spring semester and fall semester. These are standard terms in discussing US colleges and it would be odd to refer to them as the end-of-year semester or first semester of the year.
• That year they summered near Geneva, They spent the summer on Martha's Vineyard, and to escape the Parisian winter. Here it's not the dates that are relevant, it's the season -- the point is that they did this during the hot or cold weather.
• They spent the summer and autumn of 1955 in Europe, arriving in Paris in May: here I hope the month serves to let the reader know that this is a Northern summer, and prevents there from being any ambiguity.

More generally, is it possible to establish within an article that one is talking solely about one hemisphere and not the other, and so make the use of season names acceptable because they are unambiguous? I worked on Gerald Durrell, who traveled all over the world, and it wouldn't have been possible there, but the action in James Thurber never leaves the US and Europe, so the season terms are not ambiguous. It would be nice if a conventional way of establishing that could be agreed. 14:50, 20 January 2026 (UTC) Mike Christie (talk - contribs - library) 14:50, 20 January 2026 (UTC)

The examples you've given above look fine to me. The context makes the meanings unambiguous and any changes would be awkward. Clearly you've thought this through. Problems come about when people assume everyone shares the same assumptions, but you've avoided that problem. I'd not worry about it unless someone tries to change it, then, usually a courteous discussion on the talk page will resolve the issue. SchreiberBike | ⌨  16:26, 20 January 2026 (UTC)

I was a bit concerned that this part of the MoS might be enforced without regard to this kind of exception, but it sounds like that's not intended to be the case. If a conversation on the relevant talk page can't reach a consensus I might come back here for more discussion. Thanks. Mike Christie (talk - contribs - library) 22:25, 20 January 2026 (UTC)

A counterpoint from an Australian (ie Southern hemisphere). Every time you say "spring" I naturally think of Sep-Nov (which is correct for the southern hemisphere), then have to stop my chain of thought, figure out if I need to add or subtract 6 months (will it be in the prev/same/next calendar year) to make it "autumn" and then continue. Same deal for the other seasons. For us in the southern hemisphere it makes reading northern hemisphere information almost like a foreign language that has only been learnt at a beginner level. Very tedious.

So, unless the point is that they spent time in the cold, please avoid things like Parisian winter (your example dealing with the temperature, so this example is fine). Yes, it is unambiguous, but it is very tedious for us to read and we are spending as much brain power figuring out which time of year you mean as understanding the rest of the sentence. Worse, it causes us to do a context switch - which almost always causes information to be forgotten and requires re-reading.

Or to put it another way, unambiguous doesn't mean that it is simple to read when the reader's worldview is the opposite of the author's.  Stepho  talk  23:39, 20 January 2026 (UTC)

Thanks for that perspective! Gawaon (talk) 01:51, 21 January 2026 (UTC)

Stepho-wrs: I agree; thanks. There are probably twenty season references in James Thurber; it sounds like at least a couple are OK, but I'm sure some are not based on what you say. How would you feel about letting me know which ones need to be rewritten, and providing some ideas about what rephrasing could work? I sometimes get stuck trying to find phrases that both sound natural and give the reader the information intended. Mike Christie (talk - contribs - library) 02:28, 21 January 2026 (UTC)

I looked over the article (searching for spring, summer, autumn, and winter). A lot of the usages were of the sort where the climate would be more central than the date. And most of those were explicitly connected to Northern Hemisphere locations (Europe, New York, etc). I found the following usages that probably do need changing:

1) He was eventually accepted that winter.

2) The Dispatch did not rehire Thurber, but he was paid his work for the Scarlet Mask that winter,

{Note that there should be a "for" between "paid" and "his work".}

3) Over the winter of 1941–1942 he had an affair with a secretary,

The following mention more specific dates that help to clarify the intent, but still may or may not be considered ideal phrasings

1) In the winter of 1938–1939 he wrote "The Secret Life of Walter Mitty"; the story was published in the March 18, 1939

2) He and Helen had hoped to stay in Cornwall over the winter of 1942–1943, but the fuel oil they needed to heat the house was unavailable because of wartime rationing, so they returned to New York in January.

3) The Thurbers spent the summer and autumn of 1955 in Europe, arriving in Paris, in May, before going on to London,

--User:Khajidha (talk) (contributions) 18:13, 21 January 2026 (UTC)

Thanks! This is really helpful. I'll make some changes in the next day or so and will ping you to the article talk page rather than continue here, since this is article specific. I don't think MOS:SEASON needs a how-to section, but some input like this is certainly helpful for me. Mike Christie (talk - contribs - library) 19:22, 21 January 2026 (UTC)

i haven't checked the sources, but I am assuming they use similar language; it would be very unnqtural and awkward use of English for biographical sources to write things like "[they] doesn't May through late December 1955 in Europe" or "he wrote the "Secret Life of Walter Mitty" between December 1938 and March 1939." That, and it would be WP:OR for us to be giving specific months when the sources only mention the seasons and years. Also, in your second example, the weather does seem central to the point being made, that they were forced to return to New York in January due to lack of heating oil. I think that sentence would be MORE confusing to a Southern Hemisphere reader without mentioning "winter", because for them January is summer and it would be very odd to be forced to leave the place you are for lack of heating oil in the middle of the dog days, no? ~2026-59608-1 (talk) 05:46, 2 February 2026 (UTC)

At MOS:DATE#Formats, the MOS says that abbreviating months is allowed. It uses the examples of 2 Sep 2001, Sep 2, 2001, 2 Sep, Sep 2, and Sep 2001 and says that such abbreviations are allowed in limited situations where brevity is helpful […] For use in tables, infoboxes, references, etc. As a tangent of this discussion at {{death date and age}}, I just want to get confirmation that the extant allowances are… correct and accepted consensus, I guess. I want to make sure I'm not operating there from a misunderstanding of this page. Thanks, all! — Fourthords | =Λ= | 20:36, 7 February 2026 (UTC)

Someone's birth and death date in the infobox on them should not be abbreviated for sure. We are not as limited of space as that. There may be situations in infoboxes where an abbreviated month name is OK (not that I could think of actual examples), but that's not one of them. Gawaon (talk) 21:02, 7 February 2026 (UTC)

Language along these lines goes back to at least 2013 : Abbreviations [such as "Feb"] are used only where space is extremely limited, such as in tables and infoboxes. I think Gawaon has it exactly right: the intent was that abbreviation months would be used in tables, infoboxes, references, and so on sometimes, but not routinely; indeed, the current text Fourtholds quoted reads, more fully: in limited situations where brevity is helpful […] For use in tables, infoboxes, references, etc. ... By default, Wikipedia does not abbreviate dates.

Personally, I think routinely abbreviating months in infoboxes would be in keeping with what infoboxes are for -- compact presentation of basic data, but neither do I feel that abbreviating months would be some great step forward either. And... I'm pretty sure that a lot of people would find it jarring, and be upset if suddenly birth and death dates in a zillion of articles are suddenly being abbreviated because of a change in a template -- don't do that without an RfC, I urge you. Beyond that I leave this in y'all's capable hands. EEng 00:08, 8 February 2026 (UTC)

Incidentally, for references, the {{Use mdy dates}} and {{Use dmy dates}} templates allow the specification that an article's references will automatically be converted to "short" dates with abbreviated month names, by using the "s" settings in the cs1-dates= parameter. These options appear to be very infrequently used. A quick search finds some 34 articles using either cs1-dates=s or cs1-dates=ss (everything short), and another 18 using cs1-dates=ls (publication dates long, access-dates short). The number of references with cs1-dates=ly (numeric access dates) is much larger. —David Eppstein (talk) 01:06, 8 February 2026 (UTC)

I'm only asking if the "limited situations" is still applicable. I'm not suggesting we abbreviate where it's unnecessary. I and others previously used {{death date and age text}} to abbreviate in just such limited situations, but with its recent redirection to {{death date and age}}, that functionality was lost, and I was told that despite this MOS, there was no consensus allowance for such occasional abbreviating. — Fourthords | =Λ= | 16:46, 8 February 2026 (UTC)

In which situations did you use the abbreviated version instead of the normal one? Gawaon (talk) 17:51, 8 February 2026 (UTC)

If you're asking generally, I've both used them (and seen others do so) when an overlong rendered date & age (possibly as long 32 characters) caused varible or parameter lines to overflow outside editors' control (which can then have additional knock-on effects on infobox presentation and rendering). If you're asking for specific and extant uses of abbreviated months in infoboxes: I'm sorry, but I don't know how to check/search for those. FWIW, I gave a quick and tiny example of why one might want to, as allowed by this MOS, in this hastily-conjured screenshot. — Fourthords | =Λ= | 18:59, 8 February 2026 (UTC)

Hmm, I don't see a convincing case here. Two-line entries in infoboxes are normal and optimizing an infobox for a specific appearance is bound to fail anyway since that depends on what system people are on, which font size they use etc. So, by all means let the poor Mr. Public have his full September death date, he deserves nothing less. Gawaon (talk) 19:51, 8 February 2026 (UTC)

Oh, I didn't mean to suggest my screenshotted example was emblematic of ways that {{death date and age text}} was and is used. It's just something I threw together to illustrate another discussion. Like I said, I don't know how to find places the template was used in the wild for this (or any other) purpose. As for convincing, do we know what discussion led to the current coified consensus? That might have the examples or situations for which you're looking. — Fourthords | =Λ= | 19:58, 8 February 2026 (UTC)

What consensus exactly are you talking about? I'd reiterate my earlier point that {{death date and age}} and similar templates should never abbreviate the month. I don't think we have a consensus that they may; that option is meant for other situations when there are lots of dates and space is limited (maybe lists of dated events or publications). Gawaon (talk) 21:16, 8 February 2026 (UTC)

What consensus exactly are you talking about? I just mean whatever consensus, either explicit or observational, that resulted in this page (Wikipedia:Manual of Style/Dates and numbers) allowing abbreviations in "in limited situations where brevity is helpful […] For use in tables, infoboxes, references, etc." I don't know how it came about originally, but it's apparently been in place since 2013. As for your more-limited interpretation thereof, if that's the consensus understanding, we should change the MOS to reflect that (though it's not in line with my experience). — Fourthords | =Λ= | 21:41, 8 February 2026 (UTC)

The "limited situations" are already there and must, of course, be taken into account whenever the use of abbreviated dates is to be considered. So I think the wording is fine, only its application must be considered on a case-by-case. It's not a blanket permission to use abbreviated months anywhere in an infobox or table. Gawaon (talk) 09:02, 9 February 2026 (UTC)

That's correct! It's an allowance to use abbreviated months where brevity is helpful, including sometimes in infoboxes. You're also correct that local consensus is the crux upon which any individual use hangs. I'm not trying to change anthing, I'm just making sure that the MOS still reflects consensus that abbreviating months in infoboxes is still allowed sometimes—merely for the purpose of continuing (or not) to reference it in another discussion. — Fourthords | =Λ= | 13:06, 9 February 2026 (UTC)

Someone's birth and death date in the infobox on them should not be abbreviated for sure. That isn't what MOS:NUM says. If that's consensus, then should we change the MOS to explicitly prohibit such abbreviations? We are not as limited of space as that. We are, sometimes: at the aforementioned template discussion, I included this hastily-conjured screenshot to show just one easy example. — Fourthords | =Λ= | 16:46, 8 February 2026 (UTC)

In this edit, an editor feels that MOS:FRAC is a sufficient excuse to make the 1/3 page less useful.

In particular, the 1/3 disambig page, like the 1/2 disambig page, should have the relevant Unicode character, ⅓, on it.

While “Some precomposed fractions may not work with screen readers, and not all fractions are available precomposed”, there are times when one wants to have the Unicode fraction to quickly cut and paste, especially in the article about the fraction in question.

Samboy (talk) 14:09, 18 February 2026 (UTC)

That one's a disambiguation page, though. In fact we have no article about the fraction 1/3. Gawaon (talk) 15:41, 18 February 2026 (UTC)

Fair enough. I put ⅓ in the bullet list item about the fraction 1/3 on the 1/3 disambig page. Samboy (talk) 17:26, 18 February 2026 (UTC)

Yeah, that makes more sense! Gawaon (talk) 17:49, 18 February 2026 (UTC)

You've also added 1.33333... to that line. I've corrected it but it's still veering into breaching WP:D2D and WP:DABLONGDESC. Broadly speaking, each bullet of a disambiguation page should be a brief link to an article, with sufficient detail to distinguish that target from the others listed; it shouldn't be a substitute for an article. NebY (talk) 18:16, 18 February 2026 (UTC)

In abbreviating something like "18th century", is it "18th C." or "18th c."? Generating example: see subsections under Tel Lachish#Select inscriptions.

Unless I missed it, this doesn't seem to be mentioned in this MOS page. Could it be added? If it is elsewhere, could a wikilink to that place be added?

Feline Hymnic (talk) 19:45, 22 February 2026 (UTC)

I suggest 'C.', as 'c.' represents 'circa' to me... GiantSnowman 19:48, 22 February 2026 (UTC)

I was just about to say the same. —David Eppstein (talk) 21:10, 22 February 2026 (UTC)

The only abbreviation given in the print Collins English Dictionary is "C", prefixed, e.g. C14 = 14th century, and that's the only one I use in making personal notes. I would not use it or any other abbreviation for century on Wikipedia and see no benefit to the readers in the example you've given, only a risk of obscurity and bafflement. NebY (talk) 21:19, 22 February 2026 (UTC)

Why not use the full form? Unusual abbreviations should only be used in emergencies, and that article doesn't seem to be one. Gawaon (talk) 21:47, 22 February 2026 (UTC)

I would spell out "18th century", per Gawaon. Dondervogel 2 (talk) 23:13, 22 February 2026 (UTC)

See MOS:ABBR. We don't have space restrictions like a paper publication. Use the full word unless it's necessary like in a table. If needed, "18th C" could work, but it's hardly ever needed. SchreiberBike | ⌨  00:41, 23 February 2026 (UTC)

The primary aspect of my question is about whether the MOS gives guidance on abbreviating "century": an act that has been reasonably common in printed literature. After all the MOS is there to give guidance on a wide range of stylistic and grammatical conventions: WP's considered consensus. My particular "century" instance is simply one such instance within the range.

If MOS already covers the century case, then my question is very quick and simple: where is it, please? Have I missed it?

But if we don't already have the century case, then my question becomes a two-part (a) primary: shouldn't we have one? (b) secondary: what, after considered discussion, should it say?

Feline Hymnic (talk) 10:02, 27 February 2026 (UTC)

We tend to use less abbreviations than print publications, since space considerations are much less significant online. Plus we write for a general audience, while the publications you have in mind are supposedly for a more specialized, academic audience. So "but they do it" is not a particularly relevant or convincing argument here. Gawaon (talk) 11:08, 27 February 2026 (UTC)

If you're looking for advice on abbreviations in general, SchreiberBike already linked it: MOS:ABBR. "Century" is not mentioned there, which suggests it should not be abbreviated. Gawaon (talk) 11:12, 27 February 2026 (UTC)

Also, don't expect us to explicitly list all words that should not be abbreviated. Such a list is impossible for obvious reasons. Gawaon (talk) 11:14, 27 February 2026 (UTC)

(a) No, as several editors have already responded above. There's therefore no point in addressing (b), but the extreme difficulty (to put it mildly) that we would have in finding an instantly recognisable, unambiguous and clearly conventional abbreviation is one of the reasons why we shouldn't abbreviate "century". NebY (talk) 13:22, 27 February 2026 (UTC)

Yes, I'm inclined to agree that 'century' does not need to be abbreviated. GiantSnowman 21:42, 27 February 2026 (UTC)

Hi all. I was editing Stygiomedusa; I was uncertain whether to use fathoms (in addition to feet) for depth and mils for small units (e.g. 2 mm) or not. I decided not to use them. What do you think about?-- Carnby (talk) 19:41, 6 March 2026 (UTC)

• Ahoy, matey! Methinks fathoms may not be too useful to our readers, but I'm just shooting from the hip. And a mil is a 1/1000 of an inch -- I believe that if you try to use it for mm, someone will hit you on the head. EEng 20:35, 6 March 2026 (UTC)

  Not as a primary unit; I found there something like this:

  {{convert|2|mm|mil|abbr=in}}

  I decided to change it with this:

  {{val|2|u=mm}}

  -- Carnby (talk) 21:43, 6 March 2026 (UTC)

• You made a good choice. Dondervogel 2 (talk) 20:36, 6 March 2026 (UTC)

I found the following from MOS:CENTURY to be a bit redundant, verbose, and confusing:

• The 18th century refers to the period (1701–1800), while strictly the 1700s refers either to (1700–1799) or (1700–1709). When using forms such as the 1900s, ensure there is no ambiguity as to whether the century or the decade is meant.

So I changed it to:

• The form the 1700s can refer to either (1700–1799) or (1700–1709). If there is ambiguity, consider more precise forms such as the 18th century or the first decade of the 1700s.

The statement "The 18th century refers to the period (1701–1800)," was repeating an earlier bullet point. The word "strictly" seemed confusing and unneeded. Feel free to revert if you think it is a step backwards. Noleander (talk) 14:14, 13 March 2026 (UTC)