Talk:antimuslim
Latest comment: 1 year ago by Dan Polansky in topic RFD discussion: October–December 2022
The following information passed a request for deletion (permalink).
This discussion is no longer live and is left here as an archive. Please do not modify this conversation, but feel free to discuss its conclusions.
Vanishingly rare misspelling: these are excluded per CFI. [anti - Muslim / antimuslim] at the Google Books Ngram Viewer. shows absolutely damning frequency ratio, over 200 000. This is astronomical. Could be speedied, IMHO. --Dan Polansky (talk) 15:44, 12 October 2022 (UTC)
- Oops, they are showing percentages. The ratio reaches 2000. That is not such a clear result for "rare misspelling"; it is clear for "misspelling". Do what you want; the spelling is non-standard or a misspelling either way. "(rare) Alternative letter-case form of anti-Muslim" is very descriptivist, but perhaps a tad too much. --Dan Polansky (talk) 16:00, 12 October 2022 (UTC)
- Keep. There are other entries that don't capitalize Muslim (demuslimize, muslimonazi, muslimtard, nonmuslim, promuslim, pseudomuslim), and it's not uncommon to see people use muslim on its own (to the point I'm surprised we don't already have an entry for it as an alt form). Although religious terms are usually capitalized, this isn't always the case in compounds or less formal contexts, and I see no harm in our acknowledging that fact. Binarystep (talk) 09:21, 13 October 2022 (UTC)
- These are mostly Usenet-attested forms from people who cannot spell. Some of them deserve the same fate as "antimuslim": they are non-standard vanishingly rare variant spellings of the "correct" spellings. If we should have an entry for "muslim" (wrong capitalization), I guess we may also Usenet-attest japanese, italian and perhaps london. That would be entirely pointless since a search for "japanese" takes the user to "Japanese" anyway, and muslim already has Muslim at the top. Collecting evidence from non-copyedited corpora that people sometimes cannot spell is lexicographically very uninteresting. It is one thing to collect rare words such as muslimonazi; it is another thing to collect vanishingly rare odd spellings of words that we do cover. In terms of collecting evidence of human behavior, Citations:antimuslim is still a thing and not subject to deletion. --Dan Polansky (talk) 08:30, 14 October 2022 (UTC)
- The creator was (one of many "sockpuppet" accounts of) a troublemaker who obsessively added Islam and "Zionist" stuff. This doesn't automatically disqualify the word, but it's context. Equinox ◑ 20:58, 13 October 2022 (UTC)
Keep - this nomination is just prescriptivist snobbery. Theknightwho (talk) 06:14, 20 October 2022 (UTC)
- Keep—some authors, in recent history anyway, have adopted a convention of spelling adjectives denoting religions with a lower-case initial, e.g. Gregory Dix's Shape of the Liturgy (1945) has "christian" throughout and I've seen "protestant" and "catholic" in more recent works. So on the face of it I don't think forms like antimuslim have to be misspellings, though they might be hard to track down—I think this convention was most popular 50 years or so ago, unfortunately a period for which copyright laws and lack of scanning have made it annoying to search for material. A bit of trawling on Google Books turned up an example of antimuslim in a published book here, unsurprisingly from that period and alongside "hindu" and "muslim". —Al-Muqanna المقنع (talk) 23:37, 20 October 2022 (UTC)
- My long-term position is that extremely rare variant spellings are to be treated as misspellings. If we accept the reasoning above, we need hindu and muslim as rare variant spellings in English. The linked book seems to be by a non-native speaker, who is more likely to adopt capitalization conventions of their native tongue by mistake. The question remains: what is wrong with concieve to make it a misspelling? It is very well attested in edited corpora. Etymology cannot be the problem since terms are often broken from etymological standpoint. The only problem I see is that the spelling is overwhelmingly rejected by native writers and their copyeditors as evidenced by relative frequency. Otherwise, one may claim concieve is just a proscribed rare variant spelling, not a misspelling. --Dan Polansky (talk) 14:10, 31 October 2022 (UTC)
- I do feel they should be listed as variant spellings on this basis, yes, but gathering evidence for attestation is difficult owing to Google Books being case-insensitive. To me, in any case, it would strain credibility to describe (for christian) an obviously intentional usage within a standard reference work on the liturgy by an Oxford lecturer—and one that nobody, to my knowledge, has complained about—as an "error". And if one religion can be decapitalised without editorial controversy, it's hard to see why others can't be: variation of capitalisation is often used to make a point, as we know recently from black/Black. On the other hand, I find it hard to believe that there are even a handful of published authors out there intentionally using "concieve" rather than "conceive": intentionality may be a distinguishing criterion here. —Al-Muqanna المقنع (talk) 13:22, 1 November 2022 (UTC)
- I don't think intention is decisive for something to be a misspelling or not. I'll grant that christian,Christian at the Google Books Ngram Viewer. reveals that christian used to be common in 19th century; not so much muslim, Muslim at the Google Books Ngram Viewer.. I think what matters is the overwhelming rejection of christian now, not by pontificators inventing things but rather by actual language users, publishers and copyeditors. muslim is overwhelmingly rejected as well. That is, mis- not as in "proscribed by authority" but rather as in "overwhelmingly rejected by language users" to the point of "driving the frequency very close to zero". --Dan Polansky (talk) 14:14, 1 November 2022 (UTC)
- If not decisive, it's certainly relevant—I've myself been involved in publishing and it's not unusual for a publisher to ask whether some variant form is an error, with the understanding that if it was intentional it's not. If, for example, a philosopher is intentionally using "concieve" for whatever reason (cf. Derrida's infamous différance in French), then it might be a nonce word but it's not an error. In general, anyway, my sense as a native speaker is that variant capitalisation is not nearly so strictly proscribed in English as you seem to think it is—infrequent is one thing, proscribed is another, and one can't be imputed from the other. —Al-Muqanna المقنع (talk) 14:44, 1 November 2022 (UTC)
- We use the word "proscribed" to refer to what authorities and pundits do, often contrary to empirical evidence. I intentionally used the phrase "overwhelmingly rejected". My notion of "error" is "considered to be an error by all but a vanishingly small portion of the native speakers and writers". As a point of contrast, I'll pick publically: it is rare, but not so extremely rare. publicly is more than 25 times more common than publically. publically is still going to be often considered an error, but it is not so extremely rare as christian and muslim. The only plausible explanation for the extreme rarity I have is that it is overwhelmingly rejected; this can be corroborated by such observation as that when it is found at all, it is often by non-native speakers. Maybe we could at least agree that if you bring text with "christian" to a copyeditor, you will get it marked as an error, and that almost no native speaker, or even non-native in fact, is going to respond to the copyeditor that it should remain in lowercase. --Dan Polansky (talk) 15:01, 1 November 2022 (UTC)
- That is a faulty inference. Hypothetically, 95% of people might prefer spelling A over spelling B, but 80% might consider either spelling to be a legitimate variant. Theknightwho (talk) 14:08, 4 November 2022 (UTC)
- Not at these huge frequency ratios. I mean, 1:100 is damning enough, but we are talking 1:1000. Even if there is any doubt about "prefer" vs. "consider legitimate", the 1:1000 ratio provides a great buffer. Does anyone really maintain in earnest that lowercase "muslim" is an accepted spelling in English and that its extreme rarity is merely a signal of "preference" rather than "disapproval"? I for one don't believe it for a second. I maintain that frequency ratios are objective evidence not based on whims and unsubstantiated prescriptions or subjective estimates and are the golden standard. Our CFI agrees in part by mentioning "statistics" in its misspelling section. --Dan Polansky (talk) 16:38, 4 November 2022 (UTC)
- You still cannot infer that, and when multiple native English-speakers are saying you're wrong, it suggests you're wrong. Variants which are rare-but-accepted do exist. You just don't want to admit it, as usual. Theknightwho (talk) 16:45, 4 November 2022 (UTC)
- Editor opinions are not objective facts. Right, I don't accept the extremely-rare-but-accepted notion. In fact, even publically is probably not universally "accepted" and that is not "extremely rare".
- Native-speaker editor opinions that "color" is not accepted as British has no value either: it is accepted both per corpus evidence and per OED. These are wiki-external sources.
- In any case, you guys are on a public record claiming that lowercase "muslim" is "accepted". Objective evidence is not opinions of editors; objective evidence would be a poll of language users. The standard of external "verifiability" that must not depend on personal witness of wiki editors is not met by your position. Whatever the weakness of my position, it rests entirely on observations about wiki-external phenomena and on reading about "descriptivism" from notable linguists. --Dan Polansky (talk) 17:40, 4 November 2022 (UTC)
- Neither are faulty inferences. Data is useless if you don't interpret it properly. For instance, the OED does not claim to only describe British English, and other British corpora show that "color" trails "colour" far more significantly than Google's. I don't have any interest in trying to convince you of anything, because you have shown in the past that you don't know what you're talking about, are only interested in evidence that supports your position (while ignoring that which does not), and have zero interest in learning anything new from others. That is not a productive attitude. Theknightwho (talk) 17:46, 4 November 2022 (UTC)
- In this discussion, you are not trying to convince me but the reader. That is the only utilitarian reason for responding at all. I submit to the reader that witness editor evidence should not count as it fails the external verifiability wiki standard; external user poll could. If there is corpus data suggesting "color" is less accepted in British English, it would be worthwhile to add this to our usage notes, and I could well be convinced; why not. By contrast, dismissing usage frequency outright in misspelling determination is contrary to our CFI and contrary to past misspelling deletion discussions.
- OED does focus on British English, as evidenced e.g. by their including mostly hyphenated non-X forms and not at the same time solid nonX forms.
- From what I have seen, the opposition to deleting "antimuslim" did not propose any objective tests for "misspelling". If they did, it would add real value to this and future discussions. --Dan Polansky (talk) 18:11, 4 November 2022 (UTC)
- Multiple native speakers are explaining to you that you, a non-native speaker, are misinterpreting Ngram raw data. You have addressed none of the concerns other than with incredulity. That is not persuasive. Your naive conclusion that “color” is frequently used in British English simply shows the weakness of taking Google’s British corpus at face value, as it obviously hasn’t been checked for false positives compared to the others we have available. Theknightwho (talk) 21:03, 4 November 2022 (UTC)
- Neither are faulty inferences. Data is useless if you don't interpret it properly. For instance, the OED does not claim to only describe British English, and other British corpora show that "color" trails "colour" far more significantly than Google's. I don't have any interest in trying to convince you of anything, because you have shown in the past that you don't know what you're talking about, are only interested in evidence that supports your position (while ignoring that which does not), and have zero interest in learning anything new from others. That is not a productive attitude. Theknightwho (talk) 17:46, 4 November 2022 (UTC)
- You still cannot infer that, and when multiple native English-speakers are saying you're wrong, it suggests you're wrong. Variants which are rare-but-accepted do exist. You just don't want to admit it, as usual. Theknightwho (talk) 16:45, 4 November 2022 (UTC)
- Not at these huge frequency ratios. I mean, 1:100 is damning enough, but we are talking 1:1000. Even if there is any doubt about "prefer" vs. "consider legitimate", the 1:1000 ratio provides a great buffer. Does anyone really maintain in earnest that lowercase "muslim" is an accepted spelling in English and that its extreme rarity is merely a signal of "preference" rather than "disapproval"? I for one don't believe it for a second. I maintain that frequency ratios are objective evidence not based on whims and unsubstantiated prescriptions or subjective estimates and are the golden standard. Our CFI agrees in part by mentioning "statistics" in its misspelling section. --Dan Polansky (talk) 16:38, 4 November 2022 (UTC)
- That is a faulty inference. Hypothetically, 95% of people might prefer spelling A over spelling B, but 80% might consider either spelling to be a legitimate variant. Theknightwho (talk) 14:08, 4 November 2022 (UTC)
- We use the word "proscribed" to refer to what authorities and pundits do, often contrary to empirical evidence. I intentionally used the phrase "overwhelmingly rejected". My notion of "error" is "considered to be an error by all but a vanishingly small portion of the native speakers and writers". As a point of contrast, I'll pick publically: it is rare, but not so extremely rare. publicly is more than 25 times more common than publically. publically is still going to be often considered an error, but it is not so extremely rare as christian and muslim. The only plausible explanation for the extreme rarity I have is that it is overwhelmingly rejected; this can be corroborated by such observation as that when it is found at all, it is often by non-native speakers. Maybe we could at least agree that if you bring text with "christian" to a copyeditor, you will get it marked as an error, and that almost no native speaker, or even non-native in fact, is going to respond to the copyeditor that it should remain in lowercase. --Dan Polansky (talk) 15:01, 1 November 2022 (UTC)
- If not decisive, it's certainly relevant—I've myself been involved in publishing and it's not unusual for a publisher to ask whether some variant form is an error, with the understanding that if it was intentional it's not. If, for example, a philosopher is intentionally using "concieve" for whatever reason (cf. Derrida's infamous différance in French), then it might be a nonce word but it's not an error. In general, anyway, my sense as a native speaker is that variant capitalisation is not nearly so strictly proscribed in English as you seem to think it is—infrequent is one thing, proscribed is another, and one can't be imputed from the other. —Al-Muqanna المقنع (talk) 14:44, 1 November 2022 (UTC)
- I don't think intention is decisive for something to be a misspelling or not. I'll grant that christian,Christian at the Google Books Ngram Viewer. reveals that christian used to be common in 19th century; not so much muslim, Muslim at the Google Books Ngram Viewer.. I think what matters is the overwhelming rejection of christian now, not by pontificators inventing things but rather by actual language users, publishers and copyeditors. muslim is overwhelmingly rejected as well. That is, mis- not as in "proscribed by authority" but rather as in "overwhelmingly rejected by language users" to the point of "driving the frequency very close to zero". --Dan Polansky (talk) 14:14, 1 November 2022 (UTC)
- I do feel they should be listed as variant spellings on this basis, yes, but gathering evidence for attestation is difficult owing to Google Books being case-insensitive. To me, in any case, it would strain credibility to describe (for christian) an obviously intentional usage within a standard reference work on the liturgy by an Oxford lecturer—and one that nobody, to my knowledge, has complained about—as an "error". And if one religion can be decapitalised without editorial controversy, it's hard to see why others can't be: variation of capitalisation is often used to make a point, as we know recently from black/Black. On the other hand, I find it hard to believe that there are even a handful of published authors out there intentionally using "concieve" rather than "conceive": intentionality may be a distinguishing criterion here. —Al-Muqanna المقنع (talk) 13:22, 1 November 2022 (UTC)
- My long-term position is that extremely rare variant spellings are to be treated as misspellings. If we accept the reasoning above, we need hindu and muslim as rare variant spellings in English. The linked book seems to be by a non-native speaker, who is more likely to adopt capitalization conventions of their native tongue by mistake. The question remains: what is wrong with concieve to make it a misspelling? It is very well attested in edited corpora. Etymology cannot be the problem since terms are often broken from etymological standpoint. The only problem I see is that the spelling is overwhelmingly rejected by native writers and their copyeditors as evidenced by relative frequency. Otherwise, one may claim concieve is just a proscribed rare variant spelling, not a misspelling. --Dan Polansky (talk) 14:10, 31 October 2022 (UTC)
Keep per Binarystep and Theknightwho. Whoop whoop pull up Bitching Betty ⚧️ Averted crashes 00:33, 6 November 2022 (UTC)
Keep – This is not a misspelling; it's little different than the term antisemitism (cf. anti-Semitism). Additionally, I have just added a bunch of citations which show that although the term is less common than anti-Muslim, it's an intentional usage that I wouldn't even characterize as "rare". All it takes is a quick search to show that the term usage is not "mostly Usenet-attested forms from people who cannot spell". Graham11 (talk) 06:11, 29 November 2022 (UTC)
- It is very rare per objective frequency evidence from GNV; these numbers do not lie. The attesting quotations provided in the entry are either not from print media or from non-native speakers, except one native from what I can see. I do not see 3 attesting quotations from print media from native speakers, and even that would not show it is not a misspelling. All opinions and inadmissible self-reports (not permissible on a sane wiki), no objective facts. No sanity here; do what you want. --Dan Polansky (talk) 12:48, 19 December 2022 (UTC)