Knife and fork edit

I’ve more often heard of ‘fork and knife,’ but anyways, who said these can’t be interchanged because they definitely can and have. Also it appears that a Siamese twin has to have no rhyme or reason (*rimshot*) why the two items can’t be switched, but you have on the list “slip and fall” and that’s just a sequence of events (you slip then fall, you don’t fall then slip) so it shouldn’t be counted as a Siamese twin — Preceding unsigned comment added by 24.18.38.178 (talk) 04:12, 18 November 2018 (UTC)Reply

That's not true. There are clear patterns. All this was covered in the 1975 article on the semantic and phonological patterning by Cooper & Háj Ross. However, other editors apparently don't know of this paper. – ishwar  (speak) 23:41, 12 March 2022 (UTC)Reply

Non-English languages edit

I'd assume these show up to a certain degree in some other languages too.

--Special:Contributions/TheSands-12 06:28, 10 March 2021 (UTC) — Preceding unsigned comment added by TheSands-12 (talkcontribs) Reply

Doubtful examples edit

Many of the examples listed in the article are doubtful, either

  • because they often are reversed (e.g. "port and startboard" may be somewhat more common than "starboard and port" but not exclusively - see also all the examples above), or
  • because they are not commonly used expressions (e.g. "Bootleggers and Baptists" appears to be a phrase coined by one particular author and is by no means "a standard part of the vocabulary of native English speakers"), or
  • because the reason for the order of the words is semantic, not a matter of collocation or idiom, as pointed out above in the case of "slip and fall" - other examples include "catch and release" and "dusk till dawn" ("dawn till dusk" is equally possible but has a different meaning, all day as opposed to all night)

Somebody seems to have made it their mission to make these lists as long as possible at the expense of being accurate or meaningful, which is a pity

--HairyDan (talk) 14:27, 12 November 2021 (UTC)Reply

I generally agree. But "port and starboard" is about 16x more common than the reverse in UK English, and 8x more common in US English.[1] Of course, other examples are more common, like pen and ink, which is >35x more common than the reverse. Do we want to have some objective threshold? Do we believe that Google nGrams is a reliable way to measure relative frequency?
As for the semantic and syntactic cases, we should just remove them... though there are some that are unclear:
  • is "victory or death" a semantic sequence, i.e., "if we fail to win, we will die" or "if we cannot win, we will accept death"; or are they interchangeable: "the result could be either death or victory"?
  • "above and beyond" -- does this mean "both above and beyond" or "above and then even further than that"?
  • "alive and kicking" -- "both alive and kicking", or "not just alive, but even kicking"?
Thoughts? --Macrakis (talk) 17:02, 12 November 2021 (UTC)Reply
On second thought, I see that Malkiel includes logical sequences, where the second element is the "consequence, inevitable or possible" of the first. I bet that different authors analyze the situation in different ways, and it would be good if our article discussed that rather than just consisting of long lists.
Following Malkiel's section VIII, then, we might want to categorize the different kinds of irreversible binomial along different axes. His discussion of "mutual relation of the two members" includes the following categories (this is just a summary):
  • Formal relation
    • Same word: "wall to wall", "years and years"
    • B a variation of A: "bag and baggage"
  • Semantic relation
    • Near-synonyms: "checks and balances", "death and destruction"
    • Complementarity, things that go together: "assault and battery", "hat and coat", "meat and potatoes"
    • Opposites: "assets and liabilities", "dead or alive", "true or false"
    • One a subdivision of the other: "genus and species", "dollars and cents", "months and years"
    • Consequence: "divide and rule", "shoot and kill", "spit and polish", "rise and fall", "if and when"
Since we're supposed to follow our reliable sources, I guess I was wrong above to say that we should exclude logical sequences; but perhaps we should annotate or classify them somehow. --Macrakis (talk) 17:56, 12 November 2021 (UTC)Reply
Replying to HairyDan:
Your assumption is that one order has 100% (1.0) probability of occurrence and the other order has 0% probability. But, that need not be so. If there is a strong tendency/preference for one order over the other order, then that is something must be included in a description of the English language. For instance, it could that when asking an individual person, they don't have strong preferences for one order for some binomials but strong preferences for others. But, when you ask thousands of speakers, you find clear preferences over the speaker population on average. – ishwar  (speak) 23:52, 12 March 2022 (UTC)Reply
To user:HairyDan I think you're right, but being right doesn't greatly diminish the article's accuracy or meaningfulness. I agree; yet I am one of the main ones who developed the article to make it the admittedly bloated listing of odd expressions that it is. I think the reader is going to marvel at the pervasiveness of the irreversible binomial to the extent that he gives us some leeway in defining the term.
I came here to specifically call out fast and loose. I am, almost certainly, the one who added that to the article. And if I'm not, I permitted it and continued to include it through dozens of edits. I now have my qualms about it. I am revisiting the article to say that I indeed could and probably would say loose and fast given the opportunity.
But after reflection, I am deciding to keep it in the list. Why? Because the fast and loose version is well-known and well-used, and its reversal would not be frequently uttered by those who habitually read and listen to standard English. I think the reversals are, large and by (sorry, did I mean by and large?), mistakes.
And so my take on the matter of impreciseness — and occasional irreversibility — is that we as English speakers all know that there are exceptions to the rules. Siamese twins are no exception. We read the lead of the article and observe no equivocation about irreversibility. But, alas, it doesn't take too long to find reversible words (like eggs and bacon)! I think we give these imprecise examples a pass because we are learning something about English usage, and we recognize that English usage is flexible and has many variations. It's not the same as using a fraction in front of a whole number (132) or seeing °45 — it's just English. I like to saw logs! (talk)

I know this is kind of an old discussion, and I freely admit that I come to it knowing much less about the topic than others who have participated in it seem to know. So I totally understand that the list is descriptive, but I'm a bit confused about the idea that it's based on mere prevalence of a particular usage rather than actual irreversibility. Don't get me wrong; I fully understand and support the idea of noting prevalence of a given order within a phrase. But the title of this article is Irreversible Binomial and that's not the same thing. On the contrary, many, many of the examples in it are quite readily reversible. (Yes, many reversals create a less commonly used order, but my point is that, common or not, they usually do still make sense.) This is very different from phrases that either don't make sense at all when they're reversed, such as "a swing and a miss", or they mean something different, such as "smash and grab". Calling something "irreversible" means that it's impossible to reverse without losing or changing the meaning. The remainder of the items on this page are very interesting and completely valid to note, but they absolutely do not fit the definition of "irreversible". Could the two sets of phrases be separated in some way, even as separate articles, to reduce confusion among readers? I can especially see how it could be confusing to English language learners. -- edi(talk) 00:33, 26 April 2023 (UTC)Reply

Triple example edit

Rats, Lice and History

2001:171B:2274:7C21:88B3:909D:89DC:6D5 (talk) 21:24, 2 June 2022 (UTC)Reply

"gloom and doom" as irreversible (with a caveat) versus others that are reversible edit

KD: Did the phrase "gloom and doom" predate "doom and gloom"?

ChatGPT: The phrase "gloom and doom" actually predates "doom and gloom." "Gloom and doom" has been in use since the mid-20th century, while "doom and gloom" gained popularity later. The exact origins of idioms can be challenging to trace, but "gloom and doom" appears to have been in use as early as the 1940s. The phrase suggests a sense of negativity or a pessimistic outlook on a situation. Later, "doom and gloom" emerged, likely as a variation of the original expression. It gained prominence in the latter half of the 20th century and has become the more prevalent form of the idiom. Both expressions convey a similar meaning, describing a sense of impending disaster or a negative outlook, but "doom and gloom" has become the more widely used and recognized variant in contemporary language.

More from KD: doom and gloom isn't defined in Merriam-Webster. but gloom and doom is defined. Both the Oxford and American Hritage dictionaries define gloom and doom (see gloom and doom @ Oxford and gloom and doom @ American Heritage while their respective entries for doom and gloom merely link to gloom and doom, suggesting that the former is a solecism while the latter is a recognized idiomatic phrase.

KD: Does "ears and eyes" have the same meaning as "eyes and ears"?

ChatGPT: The phrases "ears and eyes" and "eyes and ears" convey the same general meaning, but the order of the words can subtly emphasize one sense over the other. Both expressions are idiomatic and typically refer to paying attention or being vigilant.

  • Eyes and Ears: This is the more common and traditional order. It suggests a broader sense of awareness or vigilance, as both the eyes and ears are important for gathering information. The phrase is often used in contexts where someone is expected to be watchful and alert.
Example: "We need you to be our eyes and ears on this project."
  • Ears and Eyes: This order, while less common, might be used for stylistic variation. It still conveys a similar meaning, emphasizing the importance of both hearing and seeing in a situation that requires attention.
Example: "Keep your ears and eyes open for any unusual activity."

In everyday usage, the choice between "eyes and ears" and "ears and eyes" is largely a matter of preference or stylistic variation, and the meaning remains fundamentally the same. Kent Dominic·(talk) 06:03, 18 January 2024 (UTC)Reply

I asked for a concise post, and you enlist chatbot to generate this waffle? I see the OED's entry for "color" as basically a single line pointing to "colour". I don't stretch that to mean that "color" is a solecism, and neither should you. Can you show any fundamental difference between the meanings of "gloom and doom" and "doom and gloom"? In simple fact, the latter has been significantly more prevalent for about half a century, while the former remains well-represented. Just plain Bill (talk) 07:27, 18 January 2024 (UTC)Reply
I asked for an erudite post, not red herrings re British versus American spellings. The fundamental difference between the "gloom and doom" and "doom and gloom" is that the former meaning is defined in accredited dictionaries; the latter – although colloquially pervasive – isn't similarly defined but merely redirected to the former sense as defined in two of the abovementioned dictionaries and wholly absent in the remaining abovementioned dictionary. Kent Dominic·(talk) 08:15, 18 January 2024 (UTC)Reply
Borderline red herring: few and far between is irreversible solely because you can't find far and few between in any accredited dictionary. If the latter somehow became more prevalent with the troglodytes of the world, and was listed in a dictionary with a redirect to the former, I'd similarly scoff at the argument that a troglodytic neologism equates to a reversible binomial. Kent Dominic·(talk) 08:29, 18 January 2024 (UTC)Reply
Red herring, you say? Patent nonsense like "far and few between" has no place in a dictionary because it is, well, patent nonsense. "Far between and few", although a more accurate reversal, is no better, being awkward and tone-deaf. Instead of letting erudite ad hominem slurs like "troglodyte" slip out, how about you show different meanings of "gloom and doom" and "doom and gloom", or admit that there is no meaningful difference between them. Just plain Bill (talk) 12:20, 18 January 2024 (UTC)Reply
Asked and answered. Now ask yourself, which came first: the egg or the chicken? Oops, Wikipedia is not a textbook on causality or on a naïve theory that the set of [A and B] = [B and A] as some troglodytic disciples of Georg Cantor still maintain despite succeeding (pun intended) proof otherwise. Oopsie again, another ad hominem slur slipped out as applied to anyone for whom the shoe fits. Kent Dominic·(talk) 13:53, 18 January 2024 (UTC)Reply
The egg came first, before there were jungle fowl or domesticated chickens, as any fule kno. Thinking that language is governed only by logic is a serious error. If I am anyone's disciple, George Lakoff is on the short list.
"Ain't in the dictionary" does not answer my question. Is lexical "redirecting" your own coinage? It may indicate synonyms, not solecisms; can you show otherwise?
What meaningful difference is there? Just plain Bill (talk) 14:11, 18 January 2024 (UTC)Reply
"Ain't in the dictionary" ain't part of my reply. To reiterate, doom and gloom IS in the dictionary albeit as an entry that's defined solely via a redirect to gloom and doom. An argumentum ad populum that troglodytes overwhelming use doom and gloom is ridiculous. Asserting a synonymous relation between gloom and doom and doom and gloom smacks of an argument from invincible ignorance that would similarly hold thin and thick is reversibly synonymous with thick and thin. Have a go at deleting the majority of such phrases in this article.
Bye for now. I'm off to add no pain, no gain to the With rhymes and similar-sounding words section to see how long it lasts until failing some editor's rm reversible false equivalence reversion and curmudgeonly mutterings since no gain, no pain is obviously synonymous under an argument from ignorance. Kent Dominic·(talk) 17:12, 18 January 2024 (UTC)Reply

OK, I will take that as an admission that you are unable to show a difference in denotation between the phrases, specious deflections to bologna like "thick/thin" or "no pain/no gain" notwithstanding. Just plain Bill (talk) 17:57, 18 January 2024 (UTC)Reply

I'm loath to condescend, but I'll stoop to spelling it out: gloom and doom suggests a progression from a state of gloominess or darkness in anticipation of impending doom or fateful catastrophe. It implies pessimism that is fatalistically intensifying. A syntactic reversal yields a semantic distinction since doom and gloom more directly connotes the anticipation or recognition of impending doom or disaster followed by a concomitant sense of gloom or sadness. Absent any minor premise, any assertion of similarity between the two phrases is a textbook case of false equivalence. Kent Dominic·(talk) 22:27, 18 January 2024 (UTC)Reply
Wow. That was... imaginative. Just plain Bill (talk) 14:10, 19 January 2024 (UTC)wReply

That fanciful explanation begs the question that this binomial describes or evokes a sequence rather than a situation. For example "stick and rudder" is irreversible in the context of aviation, but a pilot does not execute a turn by first applying aileron pressure (with the stick), then rudder. but uses them both simultaneously, in a coordinated fashion.

Gramley and Pätzold give more weight to phonetic attributes than semantics in what determines the order of an irreversible binomial. If you don't believe me, go read the source for yourself. Just plain Bill (talk) 16:28, 12 March 2024 (UTC)Reply

Great. Edit the article accordingly using proper cites. Kent Dominic·(talk) 18:22, 12 March 2024 (UTC)Reply
Thanks for the input. Meanwhile, try not to misrepresent sources, as you did twice in this edit. Just plain Bill (talk) 19:59, 12 March 2024 (UTC)Reply
Let's not quibble about misrepresentations. It doesn't matter that (1) my edit properly corresponded to the cited source without invoking the proper page number, and (2) your "not in the given source" edit summary was equivocal seeing that the link to the source was broken. Anyhow...
I'm not saying that your argument against gloom and doom is patently unreasonable. You do make a good point but it's not entirely persuasive given how linguists disagree about what characteristics attach to whatever an irreversible binomial is supposed to be. By your rationale, you should delete night and day since day and night is a commonly used collocation that implies essentially the same meaning a la gloom and doom/doom and gloom.
Here's the thing: gloom and doom was coined in 1829; doom and gloom was coined in 1930 but wasn't really colloquialized until the Rolling Stones recorded a song with that title in 2012. Need evidience of the phenomena? Look here. Did Mick Jagger mean the same thing by doom and gloom that ordinary folks mean by gloom and doom? I'll ask him when I see him, but I would guess he'd give an artsy equivalent of no.
So, given what Lohman says about bacon and eggs, I'm restoring gloom and doom. I don't care if you delete it again as long as you don't mind risking an editing uproar by also deleting night & day, bacon & eggs, and about 45 other such pairs. Kent Dominic·(talk) 01:12, 14 March 2024 (UTC)Reply

Enough with the "broken link" red herring. The ISBN easily led to online text of the single page (58, as cited) where Gramley and Pätzold treat irreversible binomials. The following page is about pragmatic idioms, a different topic. (I could also have asked for a paper copy through interlibrary loan with nothing more than title and authors, but I'm lazy and impatient.) The authors express no doubt that the second constituent is usually more phonetically weighty, mentioning "bacon and eggs being one of the few exceptions."

While there is a suggestion that "the nearer the binomials are to the idiomatic end of the semantic scale, the more fixed they become." I see nothing on that page supporting "cannot be reversed without an accompanying degree of change in their semantic relationship." Anyway, I see you have removed the Gramley and Pätzold cite, replacing it with a truly broken link to Lohman. Did someone say "kettle logic"? (Speaking of which, date of coinage may be interesting to some amateur lexicographers, but has little bearing on any register of quotidian speech or writing.)

Disguising a dictionary.com link as the Oxford Dictionary carries more than a whiff of deception. Trying to minimize such misrepresentation of sources as something to "quibble" about amounts to shenanigans of the rhetorical sort. Bluntly put, to avoid being called a liar, don't lie about sources. Dragging Sir Mick into it amounts to bulking up the word count with fluff; it is sophomoric and tiresome; please don't. Just plain Bill (talk) 08:56, 14 March 2024 (UTC)Reply

Broken link to Lohman? Works on my end.
Lying about sources? Don't equate mistaken assertions with wrongful intent. Seems I misremembered what happpened with Lexico.
The fluff about asking Mick? Please dust off whatever remnants you might have had of anything resembling a sense of humor. Otherwise, there's an undeniable correlation, if not a causal relation, concerning the Stone's release date of the song and the increased frequency of the colloquial use of its titular phrase.
Bulking up the word count? When do I not? Don't like it? Better dismiss my incomparably sophomoric, tiresome, wearying, and sometimes enthymemic talk page elements that smart editors are wise to ignore.
For anyone who can't help themselves otherwise, I know a clinical psychologist who specializes in OCD behaviors. (Anyone tempted to offer a serious reply to that offer? Instead, please either (a) dust off whatever remnants you might have had of anything resembling a sense of humor, if not (b) ask me for the name clinical psychologist, or (c) consider un-reading my incomparably sophomoric, tiresome, wearying, and sometimes enthymemic talk page elements that smart editors are wise to ignore.) Kent Dominic·(talk) 14:45, 14 March 2024 (UTC)Reply