Talk:Shia Islam

Title grammar of Shi'a vs Shi'i

As currently written, isn't "Shia" functioning as an adjective to "Islam". Considering that "Shia" technically is a noun, should it not be written using the adjective form "Shi'i", i.e. Shi'i Islam? --HyperGaruda (talk) 08:49, 14 September 2019 (UTC)

Negative, “Shi’a” refers to who (group), and Shi’i refers to a singular Shi’a. JasonMoore (talk) 12:41, 11 August 2020 (UTC)

True if used as a noun: both Oxford and Merriam-Webster define "Shia"/"Shi‘a" as a noun exclusively, not as an adjective. In these, might I say, authoritative dictionaries, only Shi'i and Shiite/Shiite are (also) used adjectivly. --HyperGaruda (talk) 04:13, 3 December 2020 (UTC)

Explanation of term ‘Shiite’

The first parentheses end with ‘sometimes spelled Shi'ite is also used in archaic English’. I tried to change it but the change was reversed. I will change it back unless there’s a reasonable counterargument to the following:

1. This is ungrammatical.

2. ‘Shiite’ is not archaic, but very common today, and in no way a slur or inappropriate. Someone behind the edit may think that any word from Arabic must be replaced by as close a transliteration of the Arabic as possible, but that doesn’t mean usage they don’t like is ‘archaic’ if many people use it. Nor is the Arabic word for ‘English’, ‘al-injliziya’, in any way archaic just because it differs from the source language. Harsimaja (talk) 05:50, 6 January 2020 (UTC)

“Nor is the Arabic word for ‘English’, ‘al-injliziya’, in any way archaic just because it differs from the source language.”

1. Which is precisely the point. You say إنقليز when saying the word “English” in Arabic. Which is a direct Arabic transliteration of the word. Likewise it is more grammatically correct and proper to refer to a Shi’i individual as a Shi’i, and not a Shi’ite.

2. Forcing adjectives to be created in lieu of already existing (and more correct) adjectives is redundant an unnecessary. JasonMoore (talk) 12:44, 11 August 2020 (UTC)

Semi-protected edit request on 31 August 2020 there are some grammatical issues such as , and curly bracket etc

Just want to correct some grammatical issues 39.32.34.61 (talk) 06:44, 31 August 2020 (UTC)

You can suggest edits here on the form "Please change X to Y" – Thjarkur (talk) 08:51, 31 August 2020 (UTC)

Requested move 22 January 2021

Shia Islam → Shi'ism – Requesting page move based on WP:COMMONNAME. According to Google Ngram analysis Google Ngram analysis, "Shi'ism" has significantly more usage among sources than "Shia Islam". Such article naming already has precedence in regards to religious sub-groups (e.g. Protestantism, Anglicanism and Sufism).

(I'm seeing an issue with Ngram processing the request; just press enter in the search bar and it should work fine.)
Alivardi (talk) 12:37, 22 January 2021 (UTC)

Survey

• Comment But "Shia" is much more common than "Shi'ism", so if we move the article, why wouldn't we move it to Shia, which already redirects here? Rreagan007 (talk) 18:17, 22 January 2021 (UTC)

@Rreagan007: I would say because "Shia" is an all encompassing term that could refer to anything that relates to Shi'ism (Shia practices, Shia places, Shia people). "Shi'ism" and "Shia Islam" meanwhile can only refer specifically to that group's body of beliefs. I guess it's much the same reason why the articles we have here on Wikipedia are "Protestantism" and "Hinduism", as opposed to "Protestant" and "Hindu".
Alivardi (talk) 19:20, 22 January 2021 (UTC)

My impression is that when English speakers use the term "Shia" they are almost always referring to the branch of Islam. If they are referring to people they usually use the term "Shiites". Also, as far as the term "Shi'ism" is concerned, English-language dictionaries seem to prefer the term "Shiism" without an apostrophe, with "Shi'ism" being an alternate spelling.[1][2][3][4][5] Rreagan007 (talk) 19:48, 22 January 2021 (UTC)

@Rreagan007: Actually, 4/5 of the dictionaries you had cited have their primary definition of "Shia" as being a follower of Shi'ism.[6][7][8][9]

In regards to your second point, Ngram seems to pretty heavily favour "Shi'ism".
Alivardi (talk) 21:17, 22 January 2021 (UTC)

• Comment – (edit conflict) the original Ngrams search above is invalid. It might still be the case that Shiism is more common, but this query does not demonstrate it. See #Ngram searches in the discussion section. Mathglot (talk) 21:25, 22 January 2021 (UTC)

I've now included alternative spellings into the ngram, though only a couple of the ones I've found had meaningful results. And from what I can see, including wording variations will only make a negligible difference; see my reply in the section below.
Alivardi (talk) 14:30, 30 January 2021 (UTC)

• Comment: I would like to seize this opportunity to draw attention to the remark I made a while ago at #Title grammar of Shi'a vs Shi'i. --HyperGaruda (talk) 21:31, 22 January 2021 (UTC)
• Oppose. After giving this a lot of thought, I think the current title is the better option. I think it's more recognizable and is a consistent title format with our article on the other major branch of Islam, Sunni Islam. Rreagan007 (talk) 03:07, 29 January 2021 (UTC)

Tbf, the article titles for the two biggest dominations of Christianity, Catholic Church and Protestantism, aren't consistent with each other.
Alivardi (talk) 14:35, 30 January 2021 (UTC)

Plus, as I've stated previously, ngram suggests that it is the more commonly used term.
Alivardi (talk) 19:59, 30 January 2021 (UTC)

• Oppose. Shi'ism is unrecognisable. But perhaps just Shia is a possibility, as suggested above. Or perhaps Shi'ite? Andrewa (talk) 17:53, 30 January 2021 (UTC)

​ Like it's mentioned above, "Shia" and "Shi'ite" generally refer to followers of the faith in the way that Hindu and Christian do for their respective religions. Could you clarify what mean you by "Shi'ism" being unrecognisable? Ngram seems to suggest it's the most common term.
Alivardi (talk) 19:50, 30 January 2021 (UTC)

See #Reply to above discussion below. Andrewa (talk) 00:21, 31 January 2021 (UTC)

• Oppose. If we look at uniformity, we can see that "Sunni Islam" follows the same premise as this article, so moving would not be necessary. I believe the current title fits and should not be changed. Wretchskull (talk) 10:12, 3 February 2021 (UTC)

Discussion

Ngram searches

One has to be very careful attempting to demonstrate frequency when using Ngram searches. There are many pitfalls in constructing queries, and then in interpreting results. Certain queries are almost always an invalid comparison, such as comparing a unigram with a bigram (e.g., "Shiism :: Shia Islam". In particular, this Ngram search (listed as "Google Ngram analysis) is invalid. The assumption here is, that "Shia Islam" is the alternative form of "Shiism" and so they may be directly compared. This is false. In fact, a regular web search for Shia Islam (unquoted) demonstrates that there are all sorts of variations, often (but not exclusively) with interpolated words, such as "branch of", "version of" and so on. So as a first approximation, you would have to find all of the most common synonyms, and sum their frequencies, and then compare that sum to the unigram Shiism. Ngrams does permit summation, but the number of items is limited by the size of the input field. Here is one such comparison of Shiism to a summation of Shia Islam plus as many of the top aliases matching the template Shia * of Islam as I could fit in the input field (but not enough of them):

• compare 'Shiism' to 'Shia * of Islam'; a first approximation of a valid Ngram comparison (but not good enough)

We can see that the line graphs are now much closer, with Shiism still in the lead but not as much as before. However, this comparison is *still* not valid; you would have to substitute in other keywords into the '*' field, if they contributed any significant amount of information to the query, but the input field is already maxed out in that one, and I wasn't able to add more. (You could do it with two or more queries, manually summing the results at the end, and creating your own curve.) And even this only deals with the 4-gram template, "Shia * of Islam", and not any other 4-gram models or trigrams.

There is another factor which makes the original query invalid. Shiism is a unigram, and my guess is that it would provide very high precision in searching for the topic, and moderate recall. I think Shia Islam would also provide high precision, but lower recall; for one thing, it would miss articles about Shia Islam which were titled differently, for example: 'Shia and Sunni Islam', 'Differences between Shia and Sunni Muslims', 'Shi'i | History & Beliefs', not to mention unigrams such as 'Shia | Definition of Shia by Merriam-Webster'.

There are other factors as well, but these two factors should be enough to point out that a great deal of care needs to be taken, when constructing Ngram (or web search) queries, and in interpreting the results. The numbers so far, shows the "Shiism" vs. "summation of a whole lotta 'Shia Islam' aliases" still in favor of "Shiism", but they are closer than before. It's possible that "Shiism" would still be the most common, in a perfectly constructed query, but I would say that we don't know that for sure yet, and more research is needed to determine that. My hunch is that they will turn out to be pretty close. Mathglot (talk) 21:25, 22 January 2021 (UTC)

HyperGaruda makes a good point above, and so to the Ngram comparison in the bullet item above, one would have to add various Shii variants as well. Based on the fact that Shii Islam seems to be about 1/15 to 1/10 as frequent as Shia Islam (graph), as a quick guess, the Shii variants might add another 5–10% to the lower curve in the bulleted search. This is exactly the kind of thing that underlay my "hunch" above; whenever you have a unigram-to-ngram comparison, people start finding other contributory variants of the n-gram query that add a few percent here, a few percent there, and after a while, the numbers are much different than they at first appeared. Mathglot (talk) 21:56, 22 January 2021 (UTC)

@Mathglot: I appreciate your imput; this is obviously something that I had not considered. I do want to highlight though that my proposed move had been to "Shi'ism". You however had been inputing "Shiism" into your searches (minus the apostrophe). Including the missing punctuation shows a significantly larger difference in the line graphs than would otherwise be apparent.
Alivardi (talk) 22:29, 22 January 2021 (UTC)

@Alivardi:, I'm not claiming that the search I added is definitive, in fact it is not. I merely wanted to point out, and then to illustrate by example, some of the pitfalls. By all means come up with better searches including the apostrophe. You may have to do multiple queries and carefully sum the results if they don't fit in the input field, which they likely won't. I didn't want to go the extra mile and do that here, because I've done this a number of times before, and this is not a major interest area of mine; just wanted to help steer it in the right direction, so regulars will be armed with the right data to make a proper decision. If you decide to go further along those lines, I know that ngrams interpets "foo-bar" as two words, equivalent to "foo bar"; not sure what it does with apostrophes, and you might want to look into that. (And let me know, when you find out .) Mathglot (talk) 23:03, 22 January 2021 (UTC)

@Mathglot: I understand that. I just wanted to point out that since a mistaken spelling had been used in the examples, the numbers may not be as close as you had suggested in your final hunch. Nevertheless, I really appreciate your suggestions and I will try to follow them. And of course, if I do figure out the effect of apostrophes, you'll be the first to know.
Alivardi (talk) 02:07, 23 January 2021 (UTC)

Ngram searche adjustments

From what I can see in the "Shia * of Islam" queries that Mathglot had used (now with corrected spelling), the added wording variations actually contribute very little to the "Shia Islam" curve, something which becomes obvious when they are separated. Trying to substitute in more keywords doesn't make much of a difference; see this wildcard search for the top ten keywords, the sum of which (as seen by right-clicking on the line) is pretty tiny. Similar results are shown when using the alternative "Shia" spellings.[10][11] Also note that the majority of the results don't actually refer specifically to the ideology so the actual result would be even smaller.

I've also tried searching for other variations of "Shia Islam", but nothing is coming up in meaningful enough numbers so as to show up on Ngram. Most of everything else that I'm finding doesn't actually refer to the ideology, but rather the people that follow it (à la the articles Mathglot had referenced: 'Differences between Shia and Sunni Muslims', 'Shi'i | History & Beliefs' and 'Shia | Definition of Shia by Merriam-Webster'). And after all that, "Shi'ism" is still returning nearly double the number of results. I therefore believe it's pretty clear that this is the more common name.
Alivardi (talk) 14:30, 30 January 2021 (UTC)

Wikipedia as influencer

What if the reason that "Shia Islam" is more common, is because people use Wikipedia's nomenclature, even if it may have been wrong all this time? You would have to look at results before this article started, i.e. 2004. --HyperGaruda (talk) 22:04, 22 January 2021 (UTC)

That is indeed a problem, and is systemic to Wikipedia, which is why it's so important to get these titles right. This is a much bigger problem than just this article as Wikipedia does not lead; it follows and should be addressed at a higher level, and link to it from here, and use this as an example. (I took the liberty of adding a subsection title above your comment; feel free to revert.) Mathglot (talk) 22:25, 22 January 2021 (UTC)

Further: if you are referring to web search, then I agree with you. If "people use Wikipedia's nomenclature" even if it's wrong, then presumably those people would not be historians or academics; their career would be over, if they used Wikipedia's term rather than the academically accepted one. That's why ngrams offers better numbers; ngrams results come exclusively from published books, which can be presumed to be reliable sources in most cases, rather than web search results, which often are not. If we believe that there are published books indexed by Google that are unreliable (such as self-publishing) and that may affect the numbers in a properly designed ngram search in unequal measure, then additional means should be sought to help bolster the conclusion.

One possibility would be to use tertiary sources as a proxy for the set or reliable secondary sources in order to evaluate WP:DUE WEIGHT, which is essentially what we are trying to do here with respect to title. For an example of how this was done in an Rfc on another topic, see Talk:French Revolution#Survey of 3ary sources. Mathglot (talk) 00:37, 23 January 2021 (UTC)

Previous moves

Evidence of some previous moves can be seen here.

Hopefully this RM will lead to stability. Andrewa (talk) 17:57, 30 January 2021 (UTC)

Reply to above discussion

Replying here to this edit as such discussion does not belong in the survey.

Shi'ite and related terms were unknown to most outside of Islam until the Iran hostage crisis of 1979, when suddenly the term Shi'ite was in the papers on a daily basis. It explicitly and unambiguously refers to a particular form of Islam and to its followers.

When I google Shi'ite I get more than two million ghits. Now that means nothing, as do raw Ngram figures (as discussed above). But it is easily interpreted by looking at the first few hits. Are they relevant? And they all seem to be. So now I Google Shi'ism. I get about 320,000 hits. There is no need to go further. Whether or not these are all relevant, Shi'ite is the winner by a significant margin. And this is as would be expected from the history above.

Another reason to avoid Shi'ism is that it's easily confused with schism, a related term. That would not be an overriding consideration but it makes it even less recognisable. Andrewa (talk) 00:19, 31 January 2021 (UTC)

​That would be an invalid comparison. Again, "Shiite" does not refer to the ideology, but it's followers.[12][13][14][15] With such logic, you could argue that Judaism should be moved to "Jew" since they have a similarly proportioned difference in relevant Google search results.[16][17]
Alivardi (talk) 01:38, 31 January 2021 (UTC)

No, because Judaism is far more recognisable than Shi'ism. We use a certain amount of commonsense. Andrewa (talk) 01:58, 31 January 2021 (UTC)

​If we're gonna start nitpicking analogies rather than addressing the main point, I'd rather stop this discussion here. I can see the page move is going nowhere anyway.
Alivardi (talk) 02:14, 31 January 2021 (UTC)