Talk:Stigma (ligature)

Pronunciation

How is stigma pronounced?

st

probably... Wikisquared 10:31, 2 September 2006 (UTC)

Photograph

The first letter of the word in the photo looks like an elision of sigma and tau. It looks nothing like the "stigma" character discussed in the main article: such an elision could have been done by someone who'd never seen a stigma. Is this really a good example? Dark Formal 21:50, 17 August 2007 (UTC)

The legend for the photograph should make explicity which Byzantine period the writing sample is from. It may be confusing to those who just know that Byzantine means complex and old, and then to read that "stigma" (did the Greeks themselves call it that? does it have anything to do with the english word?) did not exist until Medieval times (which people associate with ideas of unfairness and crude thought, not oldness per se).

Title of article

Why is this article called "Stigma (letter)", when Stigma is not in fact a "letter" according to ordinary definitions of the word (but rather is a ligature)? The name used on the Unicode character list doesn't really count for much by itself... AnonMoos (talk) 15:52, 13 December 2007 (UTC)

Agreed, the article title is confusing. PeterMottola (talk) 04:40, 12 January 2008 (UTC)

The article calls it a ligature, and the manuscript example looks like a ligature, but in the article the character itself looks nothing like a ligature of sigma and tau, at least on my browser. It just looks like a word-final sigma. Is it rendered correctly? Dark Formal (talk) 15:21, 12 January 2008 (UTC)

It is a ligature in fact, but before reading this article, the casual reader might easily suppose (as I did) that it was simply an obsolete letter of the Greek alphabet, like Digamma.

As it does look very like the final-form of the letter Sigma, I wonder if perhaps that is the source whence the final-form derives. Do we have any insight on that?

Nuttyskin (talk) 18:03, 14 April 2009 (UTC)

Definition

The article omits the most famous and well known use of the letter Stigma: The famous χξς or 666 of the Book of Revelations 13:18, the Stigma being the last 6.

Also, the statement: "...did not occur in ancient times and only dates to later medieval manuscripts." is questionable. The argument is based on the omission of Stigma in one Greek source, while the presence of Stigma is found in the Koine Greek book of Revelations. Many scholars place the writing of this book in the life time of St. John, making it an early 2nd century, and not a "later medieval manuscript."Prelgovisk (talk) 15:10, 26 June 2008 (UTC) —Preceding comment added by Prelgovisk (talk • contribs) 14:43, 26 June 2008 (UTC)

No, I think this is a misunderstanding. What the Revelations example shows is that the numeral sign for "6" existed. But that numeral sign is the original digamma, which of course always had that function. What did not exist at that time was the s-t ligature, used as an s-t ligature. That arose later and was then graphically conflated with the other. Fut.Perf. ☼ 15:16, 26 June 2008 (UTC)

To the best of my knowledge, the ancient manuscripts for Revelation 13:18 spell out "six hundred sixty six" (ἑξακόσιοι ἑξήκοντα ἕξ) in words, anyway. I believe the Chi-Xi-Stigma (χξϛʹ) form in numerals only occurs in medieval copies. -- Radagast3 (talk) 13:20, 16 December 2009 (UTC)

Although it seems Papyrus 47 uses numerals in upper case: ΧΞϜ -- Radagast3 (talk) 05:28, 20 December 2009 (UTC)

Template

Why is this not listed in the template? —DIV (128.250.80.15 (talk) 06:35, 15 April 2008 (UTC))

Distinguishing it from sigma

How is lowercase stigma distinguished from an end-of-word sigma? A²Kafir (and...?) 22:54, 3 February 2009 (UTC)

I have no idea. I'm trying to understand this myself, but the most I've gotten is that stigma is mostly unused nowadays as a letter. Maybe stigma always used to end a word, which means that every word that ends in lowercase sigma in modern greek actually used to end in stigma. Thus, sigma used to only have two cases, whereas now it has three. For what reason? I'm working on that. That could be and probably is totally wrong: Someone please explain it better, the article doesn't really. -Panther (talk) 22:19, 24 June 2009 (UTC)

This is incorrect. The letter is known as “sigma,” not “stigma.” There is no “t” in the name of the letter. The character that takes a different form in final position is known as τελικό σίγμα (final sigma). Pbot1959 (talk) 01:05, 10 June 2018 (UTC)

This is incorrect. The letter is known as “sigma,” not “stigma.” There is no “t” in the name of the letter. The character that takes a different form in final position is known as τελικό σίγμα (final sigma). Pbot1959 (talk) 01:05, 10 June 2018 (UTC)

I agree. I think there is some confusion with final sigma and stigma. You can find examples of final sigma throughout ancient Greek biblical writings. — Preceding unsigned comment added by 2600:1700:DC00:11F0:A149:F4F7:C7DC:CAB3 (talk) 15:09, 30 January 2020 (UTC)

Fake origin of stigma

There's an image captioned

Incorrect transition of sigma and tau to stigma.

If it's incorrect, why is it even on the site? Could someone who knows something about this replace it with something that's actually informative? —Preceding unsigned comment added by 24.238.113.249 (talk) 16:50, 15 November 2009 (UTC)

Merging

I am considering merging this page into Digamma. I find that most good references treat both as essentially the same thing. The only reason why "stigma" is notable (as a subject of an article, as a typographical element in modern print, and as a codepoint in computing) is because of its role as a numeral, and that role is the role that marks it as a mere continuation of digamma. The function as an s-t ligature is a side show (or else we could just as well start doing articles on , and all the rest.) The historical process of how digamma morphed into "stigma" is a long and complex one, and there is no real cut-off point at the stage when the s-t ligature happens to come into play.

There are some other funny stories to be told somewhere along the way, about yet more alternative names. Still researching. Fut.Perf. ☼ 15:41, 8 October 2010 (UTC)

Makes sense. There are two things that distinguish stigma from other ligatures: that it has a name; and that it is used as a numeral (but as a substitute for digamma). Merging into one article doesn't mean that the two things are exactly the same, only that they are better treated together than separately -- which in this case they should. --Macrakis (talk) 18:42, 8 October 2010 (UTC)

Okay, thanks for the feedback. I'm collecting some stuff for the merger. Seems I'm having lots of fun with Sampi, Digamma and friends these days. Who'd have thunk our lowly numerals are in fact mystical symbols of the Holy Trinity? Fut.Perf. ☼ 19:24, 8 October 2010 (UTC)

Merger done, reviews welcome. Fut.Perf. ☼ 15:47, 9 October 2010 (UTC)

moving here from user talk:

Yes, stigma is otherwise not a "letter," but it is only in a very limited sense that it can be classified as a "name" for digamma. This is just ontologically wrong. Stigma properly refers to a ligature used in writing Greek -- as its name attests -- and even its numeral usage cannot be fully classified under the heading of digamma--that would be to confuse origin with result, and to call even the Modern Greek usage στʹ (implausibly) a funny way of writing digamma. It's my opinion that your drive to rationalize the treatment of Greek writing went a couple of steps too far in this case, and that stigma should be broken back off of digamma. The constellation Ϝϝϛ, explained "Ϝ, lowercase ϝ; as a numeral symbol: ϛ," is WP:OR or close to it. Which authoritative treatments of Greek writing agree with thus subsuming stigma in general as a variant of digamma? (I would have no objection to, "as a numeral symbol, also represented by ϛ.") I'm hard pressed to apply such an understanding of stigma to reading the stigmas here. They are neither digammas nor properly considered under the heading digamma. Wareh (talk) 17:47, 11 October 2010 (UTC)

I missed the discussion above; thanks to FPAS for bringing my comment here. "It has a name" does seem significant. There is no reason in principle not to give the most notable ligatures encyclopedic treatment (compare e caudata); in modern usage it survived when all other ligatures (except for ου) had vanished. Finally, to my mind, the fact that the history of this ligature is interestingly intertwined with digamma's via the numeral usage makes it more notable in its own right, which is why it is discussed in all kinds of sources that don't care about ligatures in general. Let me make clear, though, that I would be satisfied with a short article about the ligature here, with a reference to digamma for discussion of its numeral usage: my wish is not to create pointless and inefficient repetition, but rather to avoid the odd situation where we see the unusual alphabetic entry "Ϝϝϛ" and get the general impression that the reliable sources regard ϛ as a digamma. It would be far preferable simply to omit stigma from the alphabetic listing, and my point is that this does not entail omitting an article on stigma the ligature. (However, if stigma-the-ligature can't be found via the alphabet template, then sigma should be given a "Not to be confused with the ligature ϛ (stigma)" dabhat.) Wareh (talk) 18:21, 11 October 2010 (UTC)

(ec with the above, responding to the first post:) Well, I see your points, and I had been thinking about these things for a while myself. Sorry for not waiting longer for feedback before going ahead, but on the whole, I still think I prefer the merger. First of all: sure, the "stigmas" in the 19th-cent print you point to aren't "digammas". But then, why would we have an article about those stigmas anyway? As I said above, they are no more inherently notable than , and all the rest. Nobody has been using them for the last 150 years or so. The presence of the "ϛ" character in modern digital encoding standards is explicitly not designed represent those either. They are good for a sentence in Greek ligatures, or a footnote in digamma, but why would we have more coverage about them than that? – That leaves us with "stigma-as-a-numeral", and your "ontological" problem. I think your suggestion of changing the wording to "as a numeral symbol, also represented by ϛ" makes some sense. As so often with ontology, though, it all "depends on what the meaning of is is". The problem I think the merger solves is that I just could not find a reasonable historical cutoff point where our treatment of digamma should end and our treatment of stigma should begin. When an Attic inscription in the first century uses the numeral symbol, its epigraphical shape looks more or less like a digamma, and "ontologically", it sure "is" a digamma. But in modern editions of that inscription, it will invariably be rendered as "ϛ". So, "is" that little black wiggle "ϛ" on the printed page a digamma or a stigma at that point? – If a 4th-century papyrus uses the numeral symbol , it already looks like a stigma, walks like a stigma and quacks like a stigma, and of course it will again be rendered as "ϛ" in modern print. But "ontologically", it still "isn't" a stigma, because a stigma-as-a-ligature with which to conflate it didn't yet exist at that point. So, all in all, looking at the present digamma article, which parts of it would you want to separate out again and move into stigma? When we discuss 16th–19th century references to this thing that talk neither of "digamma" nor of "stigma" but of "the Episemon ϛ", does that go into the stigma or the digamma article? If the symbol on the ancient stone and the symbol representing it on the modern page are two different entities, how do we factor out those two perspectives into two separate articles? Fut.Perf. ☼ 18:36, 11 October 2010 (UTC)

BTW, as you asked about sources: sorry the ones I can find quickly are mostly quite old, but there certainly are plenty that treat the "ϛ" symbol as primarily an instantiation of digamma which only secondarily also happens to look identical to the st ligature, rather than primarily an instantiation of the st ligature that just secondarily happens to have taken over the function of digamma.

• T. L. Heath, A history of Greek Mathematics, p.32 has a table of the numeral system. Two lines in the table look like this:

[ϛ] = 6

[ϡ] = 900

And then it says: "The sixth sign [...] () is a form of the digamma Ϝϝ. It came, in the seventh and eighth centuries A.D., to be written in the form and then, from its similarity to the cursive ϛ (=στ), was called Stigma"

• A nineteenth-century grammar (Sophocles, A Greek grammar for the use of learners, p.2) speaks of: "F or ϛ, Βαῦ Vau, or Δίγαμμα Digamma"

• Another 19th-century grammar (Buttmann's larger Greek grammar, p.22: "after ε the ϛ, here called Βαῦ, Vau, and not στ; [...] The resemblance of the ϛ to the later abbreviation for στ is only accidental; as a numeral it is called Βαῦ, and is merely another form of the digamma, F or [...]"

• Thompson, Handbook of Greek and Latin Palaeography, p.104, has: "F, digamma, for 6, which in its early form appears as or , and afterwards, in the middle ages, becomes ", like the combined σ and τ or stigma"

Fut.Perf. ☼ 19:21, 11 October 2010 (UTC)

Adding a bit more:

• Felix S. Feldbausch, Griechische Grammatik zum Schulgebrauch: "Dieses ϛ heißt als Schriftzeichen Sti oder Stigma, aber als Zahlzeichen hat es noch den ursprünglichen Namen des Digamma: Βαῦ, d.i. Ϝαῦ, Vau." [1]
• Marc W Küster, Geordnetes Weltbild: die Tradition des alphabetischen Sortierens': "Der 29. Buchstabe war schließlich das Digamma, allerdings in einer ungewöhlichen Inkarnation als Stigma {ϛ}, das in byzantinischer Zeit wohl wegen seiner Form als Ligatur von Sigma und Tau uminterpretiert wurde" [2]

And two more modern sources:

• K. Barry, The Greek Qabalah: alphabetic mysticism and numerology in the ancient world (1999), p.17: "The digamma letter was gradually discarded in Classical times [...] and was retained as a numerical sign. It became gradually simplified in form until it was no more than a large comma, the episemon or stigma (ϛ), almost identical in appearance to a semi-compound letter known as stau [sic]" – later, p.204, in a table of the alphabet: uppercase: "F", lowercase: "ϛ", name: "digamma; episemon"

• Chambers Dictionary, s.v. digamma: "the obsolete sixth letter (F, , later ϛ) of the Greek alphabet with the sound of our w, as a numeral ϛ' = 6." – also s.v. episemon: "one of three obsolete Greek letters used as numerals: ϛ or , vau or digamma (6); ϙ, koppa (90) [...]"

Fut.Perf. ☼ 20:22, 11 October 2010 (UTC)

If we wanted to be ontologically clean, we might have to treat the ligature s-t, the letter digamma, and the-symbols-denoting-6-in-Greek-numerals in three articles, not two. Fortunately, treating topics together in one article makes no ontological commitments (unfortunately for DBPedia) -- it is simply a pragmatic judgement of usefulness to the user. Thus we do not have separate articles for the departement of Paris and the commune of Paris and conversely we do have separate articles for Smyrna and Izmir. What an entity is is not at all clearcut: Heracles/Hercules, Moscopole/Voskopoje, New Amsterdam/New York. --Macrakis (talk) 23:32, 11 October 2010 (UTC)

I am hopeful I can cut through the source of disagreement here, because I agree that it is logical to treat the numeral use of stigma together with the numeral use of digamma, from which it is graphically and functionally derived. (It's not perfect, but I am no stickler, and the idea of a third article on the Greek alphabetic numeral strikes me as absurd and very different from the spirit of my constructive suggestion.)

In terms of article division, then, 100% of my disagreement is contained in the fact that I do not consider, "We should not have an article on stigma-qua-ligature," to be obvious or convincing. The ligature is significant, notable (I can't believe anyone here really means to suggest it does not pass WP:N), and covered verifiably in plenty of reliable sources. That it is not the most obscure topic in the world of Greek writing is handily represented in the fact that it is one of the twenty-seven lowercase glyphs with which the text of Bekker's Aristotle is printed. It is hardly obscure or non-notable, and, yes, it does have a name, which does make a difference. If it weren't for the interwoven numeral/ligature story, then I'd say, fine, redirect it to digamma or to Greek ligatures, but the very fact that it is connected to both makes it most convenient to keep an article on the ligature, with links to both of those topics. (Likewise, if stigma the ligature really were something else, e.g. digamma or sigma, of course I'd want it treated there.)

I am not going to look into all the sources cited but I am hopeful it is unnecessary. It seems obvious that many of them combine the treatment of stigma with digamma because they are principally interested in stigma-the-numeral. That is to be expected. I am claiming that we should retain (not invent, retain!) an article on "those stigmas," the ones that have nothing to do, in their history or in their usage, with digamma. Accounting honestly, I think we'd be hard pressed to approve of other examples of topics treated under headings with which they have as little to do as stigma-the-ligature does with digamma (in all the fullness of the latter's reference).

Finally, I hope I can get some consideration, quite apart from this ligature-notability question, for my other position, which is that printing Ϝϝϛ in the template is reifying at too general a level the idea that stigma "is" simply a digamma. Yes, there is a strong case to be made for considering that to be the case in certain ways--but surely that case should be left for the digamma article, where it can be handled with more nuance than this insinuation that Ϝϝϛ is something just like Σσς. Wareh (talk) 00:30, 12 October 2010 (UTC)

Well, taking your point that stigma-as-a-ligature is "notable" – but, quite practically speaking, what is there to be said about it, to fill an article with? (a) it existed, (b) it looked like "ϛ", (c) it was used longer than the others. That's all that can be said about it, without duplicating large swaths of material from "digamma". And even all of this is of course also said in the Greek ligatures article, and must be said again somewhere along the way in the digamma article. My question is: for a reader who is reading "Greek ligatures", is it really worth clicking another link, expecting more info about "stigma-as-a-ligature", when the info they will find there is just the same as what they have already found in the ligatures article anyway? Same for a reader in the digamma article. And the marginal benefit of having a separate article will come at a very considerable cost, of having to introduce a whole nightmare of additional disambiguation links in all sorts of places ("This article is about the use of "ϛ" as a ligature. For its use as a numeral, see...", etc.) I really can't see the practical benefit of such an arrangement.

About your other point, the legitimacy of the "Ϝϝϛ" string in the infobox: if, as you concede, we are going to treat stigma-as-a-numeral together in the digamma article, I think it is quite crucial for the navbox to connect the glyph shape to that article. The box is not about making any "insinuations" about what "is" what; it's just for people to quickly find their way to the relevant article. 90% of all readers who have come across this particular black squiggly thing and wonder what it is will have seen it in the context of the numerals and therefore need to be directed to the digamma article. The next 5% may have seen it as one among innumerable other weird unknown squiggles in an old print, and should be led to Greek ligatures. The remaining 5% who have specifically come across just stigma-as-a-ligature alone, reading Bekker's Aristotle, still won't be disadvantaged if we first direct them to the digamma article, where they find the info relevant to them in the first paragraph, and some more that is relevant to both the numeral and the ligature below in the "typography" section. Fut.Perf. ☼ 06:15, 12 October 2010 (UTC)

P.S.: if you worry about names in the navbox, how about the following (just brainstorming):

Greek alphabet
Μμ	Mu	Ωω	Omega
Archaic letters
·  ·  ·  ·
Numerals
ϛ (6) · (90) · (900)
Ligatures
ϛ (στ) · ȣ (ου) · ϗ (και)

(i.e. let's just imply we're calling the whole bunch "cypher, anonymon semeion, charaktera, skope, and wiggle". We wouldn't be the first. ;-) Fut.Perf. ☼ 11:56, 12 October 2010 (UTC)

I do think your brainstormed template suggestion would be far preferable. I hadn't even known we had articles on ȣ and ϗ; I think it strengthens the case to treat ϛ in parallel. I think the present state of stigma (letter) shows pretty well what there is to be said about the subject in its own right. Overlap with the related subject digamma is to be expected, but it doesn't seem too much to me, and there is not much need to repeat what is here at digamma except for the last paragraph, which is only meant as a short statement on what the dabhat promises will be treated fully there. Wareh (talk) 13:35, 12 October 2010 (UTC)

Yes, a third article on the Greek alphabetic numeral would be absurd -- but that might be required if we were to insist on ontological cleanness. I'd think the most reasonable solution is to redirect Stigma (letter) to Greek ligatures#Stigma since, as FP says above, there's not much to say about it as a ligature. Compare the treatment of the Latin alphabet ligature ff, which is surely many orders of magnitude more common by just about any measure. Compare on the other hand the treatment of the esszet (ß), which does get its own article because there's a lot to say about it (orthographic rules etc.). --Macrakis (talk) 13:58, 12 October 2010 (UTC)

I'm a little mystified by the assertion of how little there is to say: my starting position is that we already have approximately 4,641 bytes' worth to say about it (the current state of the article). This is enough for an article on a topic. Wareh (talk) 14:05, 12 October 2010 (UTC)

Well, yes, we can of course say a number of things about it. But there is still hardly anything here that is actually unique to this article. An article whose factual coverage (if not the "ontological" status of its nominal subject) is almost exhaustively a proper subset of a second, closely related article, is still a somewhat questionable thing. Anyway, if we were to go with your model, I think we'd still need a few more tweaks. For instance, the images and the discussion of typographical variety are only really relevant for the numerals (but we can of course get a few new images where it's used in text). The coverage of the computer encoding needs to point out that those Unicode codepoints are in fact not for stigma-as-a-ligature-in-text, so they are strictly speaking off-topic here (their only legitimate use with regard to non-numeral stigma is when citing the symbol, not in actually using it for encoding text; see N.Nicholas ref). -- Speaking of typography, do you happen to have any information about uppercase stigma-as-a-ligature? I remember having seen "Ϛίχ." for "Στίχος" somewhere the other day, but I'm not sure how rare it is. Fut.Perf. ☼ 14:26, 12 October 2010 (UTC)

Shameless OR, for later use: non-use of uppercase ligature stigma: [3]; use of uppercase ligature stigma: [4]. Fut.Perf. ☼ 14:50, 12 October 2010 (UTC)

I just added an article on the Grecs du roi with an EL pointing at a full (I think) set of punches -- it might be possible to see what ligatures are included there. --Macrakis (talk) 19:34, 12 October 2010 (UTC)

Nice. The photo you mention does seem to have some stigmas (with combinations of third and fourth letters), middle of second row from the bottom. No upper case forms. Not surprising – the very idea of having ligatures would have been solidly connected to the concept of the minuscule at that stage. Do you know of any good scans of old book pages using that font? Fut.Perf. ☼ 20:06, 12 October 2010 (UTC)

Google Image Search seems to find some images -- but I don't have time right now to look at the results. --Macrakis (talk) 21:49, 12 October 2010 (UTC)

So, what next?

Another brainstormed model
Μμ	Mu	Ωω	Omega
History
Archaic letters:  ·  ·  ·  ·
Numerals: ϛ (6) · (90) · (900)
Ligatures (ϛ, ȣ, ϗ) · Diacritics

So, what are we going to do now?

1. We've got consensus the numeral function of stigma will be treated inside digamma.
2. This means there will have to be some way of linking from the "ϛ" glyph to the digamma article in the navbox.
3. The disambiguation between the two functions of "stigma" can partly be handled through two entries on the stigma page, which is a disambiguation page anyway.
4. Remaining question: where to handle the ligature function of stigma.
   1. treat as a side issue inside digamma (my initial solution)
   2. merge into Greek ligatures (Macrakis' suggestion. That article is currently unsatisfactory and would need some expansion.)
   3. keep as separate article (Wareh's proposal)
5. Remaining question: how much coverage of the ligature signs in the navbox? Note that "kai" and "ou" in my first draft above are a bit of a cheat: neither of the two target articles is in fact centrally about that ligature (the one about ou is primarily about the Latin and Cyrillic reincarnations, the one about kai is primarily about the word, not the symbol.) On the right: another draft, reflecting a merger solution.
6. Since we're tinkering with the navbox: I'd like to renew my suggestion that there may in fact be more important entries than the individual marginal signs that are currently missing. I just added History, and sneakily added a link to what is currently a sandbox page. What about Scientific symbols? Local variants (currently also in my sandbox)? Typography?

Again

Sorry, but after a year's standstill, I have re-merged the articles. The sketch Stigma (letter) article created by Wareh's unmerger last year never really took off and was never formed into something really coherent, and I still can't see what its advantage would have been for a reader. Fut.Perf. ☼ 09:34, 23 September 2011 (UTC)

So, now we're reverting? That's a bit disappointing.

Wareh: about "consensus", you were the only person who insisted on this being kept; of the three persons involved in the discussion, two favoured merging.

About your question of "how this isn't content-destroying": everything in this page was already covered in the other article, and better.

Currently this article says: "Both closed () and open () forms were subsequently used without distinction both for the ligature and for the numeral, and the name "stigma" was applied to both functions." – This sentence was simply copy-pasted here and ripped out of context; the phrase "for the numeral" isn't even understandable at that point. You then have the same content duplicated further below.

We agreed last year that the numeral function of "ϛ" should primarily be treated over at Digamma, and that this article should deal with the use as a textual ligature. Now please check: how much of this article was actually pertinent to this topic? I see the following: "The στ-ligature became common in minuscule handwriting from the 9th century onwards". That's one sentence. Then comes the bit about the name, which is also covered at the other article. Then comes the thing about typography and modern encoding. This is exclusively about the numeral, including the images. It doesn't belong here at all. There is not a single sentence about the textual ligature use in all that part of the article.

Now, please look at this under the perspective of the reader for whom we want to package information efficiently. If a reader wants to learn about Greek ligatures, they will first go to Greek ligatures. Now, once they have read that article, what more is there here that would make it worthwhile for them to click a link to this article? Nothing. This article offers nothing new to them. Same with a reader who has first gone to Digamma: this article offers nothing new to them either. Fut.Perf. ☼ 13:35, 23 September 2011 (UTC)

I'm not trying to take a radical position here. But we're seeing a few things differently, and I have hope we can understand each other better. For starters, could you please point me specifically to the two opinions favoring merging this article out of existence? (Of course this will not be the same thing about consensus about treating the numeral six at digamma.)

I think the major possibility you're not considering is that there's such a thing as a user who wants to know about stigma, not about ligatures or numerals. Stigma meets the WP:N criteria for a coherent topic, with plenty of material for an encyclopedic treatment, as we already see. I.e. Wikipedia's answer to the question "What is stigma?" should not be, "It seems you wish to know something about the subject properly called digamma." (If you believe it's not treated well or coherently, that's an article-improvement issue.) I have no objection to a section of this article on its use as a numeral that points elsewhere for more info, and likewise on ligatures. I have no objection to this article being shorter than it was. But redirecting to digamma still seems perverse to me, because "stigma" corresponds to a meaningful bundle of information on its own, and that bundle is not a subset of the subject "digamma." Wareh (talk) 14:14, 23 September 2011 (UTC)

I'm not seeing that "meaningful bundle". The meaningful bundle of information that is specific to this article is: "it stood for s+t. It existed. It looked like this: ϛ." And nothing more. And no, stigma-as-a-ligature does not meet any notability criteria. Only stigma-as-a-numeral does, and that is treated elsewhere. We have not a single source that takes any interest in stigma-as-a-ligature, except in order to explain the development of the numeral (or as just another entry in a list of ligatures in general, among dozens of others). – The consensus for merging was between Macrakis and me; please read his opinions again, above. Fut.Perf. ☼ 14:23, 23 September 2011 (UTC)

If I understand you, by denying a "meaningful bundle," you are not actually saying that the question, "What is stigma?" is unanswerable by an encyclopedia, but that you feel strongly it should be answered in 2-3 different places. Well, perhaps an acceptable compromise would be a disambiguation page: "The ligature stigma (ϛ) may be used as..." I'd just want to be assured that all the content here is completely covered (hopefully better, but my sticking point is "completely") by a page on the list below that introduction.

My bottom line is that under your proposed redirect, the following unfortunate chain of events happens to a reader: (1) She is interested in learning the nature and use of the symbol ϛ. (2) She is redirected to digamma. (3) She gets a small fraction of the information on this page and has to scratch her head and keep clicking to find where the rest of it is and how it fits together in a complete picture.

I'm not making a deep analytical claim when I say "meaningful bundle" and "complete picture." But I believe you should acknowledge a bit better that the content collected here--however disparate--is unified by its common reference to the symbol ϛ. Wareh (talk) 15:09, 23 September 2011 (UTC)

P.S. However, I've realized that because the symbol ϛ is not the primary usage of the word "stigma," a disambiguation page would not serve the purpose. Therefore I suggest a short article on the symbol ϛ guiding the reader to where every shred of information the encyclopedia should and does contain about it can be found. Wareh (talk) 15:13, 23 September 2011 (UTC)

Well, and I still don't agree the reader in your example would get "a small fraction" of the information of this page. The way I had merged the pages, digamma gave that reader 100% of what this page is now giving her, and more. No, I don't want the topic to be treated in two or three different places. The topic is treated in a single place right now, at digamma, and that has been the case for over a year. You were denying above that the information in this article was a proper subset of what digamma contained, but that is simply not true. It is a proper subset. If the reader now searches for stigma, gets to the dab page, and the dab page directs her here, then in fact she has to scratch her head and is left struggling to piece things together from different places. See the highly cumbersome dab hatnote at the top of this article. "For the use of stigma as a numeral, see digamma". Oh yes. Its use as a numeral makes up about 99% of what users would ever want to know about it. Fut.Perf. ☼ 15:31, 23 September 2011 (UTC)

I'm sorry, you lose me with exaggeration like "99%"; stigma's use as a ligature for στ is not 1% of its interest. I can concede a couple of things. First, I may have missed some of the relevant content at digamma (though that just shows that digamma is indeed incorrectly trying to be an article on ϛ in addition to being an article on ϝ). Second, I can readily admit that e.g. the second paragraph of the article here is mostly worthless (i.e. not about ϛ). But you have not begun to persuade me that we do not need to offer an encyclopedic answer to the question "What is ϛ?"

Let me try another angle, again in hope you'll see why I think the merge-to-digamma approach is not only rigid but mistaken in its ontology. By this approach, not only does the article digamma have to receive readers interested in the ligature (which can be read in 19th century printed texts, so I really don't buy the 1% claim), but the article digamma also has to admit, when it comes to the numeral/ligature, that what it's talking about is commonly known a "stigma" and not "digamma" (I have in mind the sentence, "the name stigma is today applied to it both in its textual and in the numeral function"). Now, this second point seems of greater concern. If the generally used name of the numeral and the ligature is "stigma," then the article discussing these two things should be at stigma, not digamma.

So let me modify my proposal and suggest that an article stigma (plus whatever disambiguator: perhaps "Greek ligature and numeral") should describe the history, paleography, uses, etc., of ϛ. I hope you can see I'm trying to be flexible and find some reasonable accommodation. If you can't accept this, then I'd like to discuss on different lines: I claim that the WP:COMMONNAME for the numeral ϛʹ is stigma, and that, given that the variant ϝʹ is secondary, the article digamma should refer to stigma for the main treatment, with whatever notes there that may be appropriate for the digamma variant.

In other words, the issue is not whether the subject "ϛ" is a subset of what happens to have been included at digamma, but whether the subject "ϛ" is a proper subset of the subject "ϝ." We still have no evidence that this is the case, and it defies common sense in my opinion. Wareh (talk) 18:26, 23 September 2011 (UTC)

Won't work, for several reasons:

1. The history of "Ϝ" and the history of "ϛ" cannot be separated from each other. There is no cut-off point. It makes no sense to treat the symbol "Ϝ" as it appears in this Athenean inscription separately from the symbol "ϛ" that appears in its stead in its modern edition. They are as much the same as "Ω" and "ω" are the same. It also makes no sense to make some arbitrary decision about which of two articles the symbol "" should be treated in, when it occurs in a fourth century AD papyrus and is called neither digamma nor stigma but "episemon".
2. There are plenty of sources that have called "ϛ" "digamma".
3. No source has ever called "Ϝ" "stigma".

So, if we have to treat "Ϝ" and at least some part of "ϛ" in one article, we have no choice but to call it "digamma". Everything else would simply be wrong, WP:COMMONNAME or not WP:COMMONNAME. And yes, it has to be an article on both Ϝ and ϛ, just as Sampi has to be an article on both and Ϡ, and Koppa (letter) has to be an article on all of .

About the 1% of actual ligature-related content: if you think there's more than that, prove it. Here's the deal: if you want a separate article on "stigma" (as-a-ligature), go ahead and write one. Currently, there is not a single sentence in this article that isn't copy-pasted and cobbled together (not very carefully) from stuff I wrote for digamma, and there is not a single sentence that isn't also still covered in the other article, and must remain there. If you can write up at least one paragraph of, let's say, more than a hundred words of new, original, interesting content on stigma (letter) dealing with something the digamma article doesn't deal with, you can keep your article.

Sorry if I'm now sounding a bit cranky, but, you know, I wrote this whole damned thing; for the last three years or so I've been doing virtually the whole grunt work of trying to get this group of articles up to a decent level virtually single-handedly, and, forgive me, your intervention now leaves me with a little bit of the feeling of fighting over the color of the bike shed. Fut.Perf. ☼ 19:08, 23 September 2011 (UTC)

For any outside person who'd like to comment and is now confused, here's an approximate visualization of what we're discussing. We have three interwoven "stories": that of the alphabetic character for "w" (green), that of the numeral for "6" (blue) and that of the ligature for σ-τ (red). Circles indicate the approximate scope of each of the overlapping terms and names that have been used over time.

• The "w" character existed in preclassical Greek and again as the modern typographic symbol Ϝϝ.
• There's an unbroken continuity from the numeral use of the "w" character in antiquity to the numeral "ϛ" today, both in glyph development and in function.
• The ligature sign developed separately and only accidentally has the same shape.
• However, the presence of the modern "στ'" spelling as part of the "numeral" story is taken over from the "ligature" story, and the modern name "stigma" is motivated by it.
• The modern names are overlapping: "digamma" is always used for the "w" character and (more rarely) for the numeral, while "stigma" is used for the ligature and (more often) for the numeral.
• In modern typesetting/transcription of ancient inscriptions, the modern typographic "ϛ" is used for any instantiations of any of the early forms on the "blue" bar. The modern typographic "ϝ" is used for any instantiations on the "green" bar.
• Retroactive application of the name "stigma" to any of the early instantiations of symbols on the "blue" bar is found very occasionally in modern practice, but is anachronistic.
• In terms of encyclopedic coverage, we have plenty of good and well-sourced material about the "green" and "blue" stories, but not more than two sentences so far covering the "red" story in its own right.

So, the question is: at what point in this story can we make a cut and divide our coverage into two separate articles? My position remains: at no point. The only practical, reader-friendly solution is to treat all of this together in one article, as is in fact currently done at digamma. Fut.Perf. ☼ 07:07, 24 September 2011 (UTC)

We are discussing, I take it, this independent version of the article on stigma?

I do not find the present situation (merging that article to nothing) reader-friendly; I would find it much less friendly if I were new to the subject. There are two chief problems:

1. If I did not know anything about the subject, but were following a reference to the letter stigma, which I did not understand at all (and most readers will be in that position), I would be puzzled in being dumped on digamma at all.
2. Even though I do, I find the frequent references to "the shape of digamma" confusing. Digamma has at least two, topologically distinct, shapes: ϝ (the one referred to by the name), and ϛ ("alphabetic" and "numeric" digamma, maybe? although the F-shape is quite often used as the numeral too.) It would help, incidentally, if the digamma rendered to its normal printed form, with two equal and upward-sloping prongs. At least the suggestion that C is normally stigma (rather than sigma) has been edited out.

I would restore the article on stigma, and let digamma cross-reference to it; many of its problems would be solved by not having to discuss the /w/ sign, the /st/ sign, and the sign for 6 at the same time (and sometimes the /s/ sign). Septentrionalis PMAnderson 20:30, 26 September 2011 (UTC)

I'm not sure I understand your position. If you want the articles split in two again, then which parts of the content would you want to see treated in which of them? Where would we make the cut? And I honestly don't understand your remark about "the shape of digamma". I checked every occurrence of the word "shape" in both articles, and there doesn't seem a single occurrence in which it isn't made immediately clear which shapes are being talked about. Fut.Perf. ☼ 20:48, 26 September 2011 (UTC)

• I would treat /st/ and "6" here; I would treat all three (with /w/) in digamma, but let the mention of /st/ be summary, with a cross link back to here.
• As for the other: the lead to digamma says, with no immediate explanation, Today the numeral sign is usually called stigma, after the value of a Byzantine Greek ligature σ-τ (ϛ), which shares the same shape and was used as a textual ligature in Greek print until the 19th century. I think I see how this is meant to be read; but the first clause is false, in English, and the rest can all too naturally be read as an assertion that "digamma" has only the single shape ϛ. Septentrionalis PMAnderson 22:10, 26 September 2011 (UTC)

• Hmmm... What's false about the first clause? And the statement about the shapes comes just two sentences after the lead sentence, which explicitly enumerates this shape as one among others, so I'd reckon it really ought to be clear enough. – About the division of topics with the "summary" of the st topic in the other article: the amount of stuff we have to say about it is so minuscule (pun accidental) that we really can't shorten it any further. What's to further "summarize" about two sentences? But oh well, I guess that's what we have now, and at least with my latest rewrite it's again halfway coherent. Fut.Perf. ☼ 22:19, 26 September 2011 (UTC)

Evidence

• I cannot agree that stigma is the usual term in English for the numerical sign; the OED does not know this sense of the word, although it does mention the sixishness of digamma. Searching for it on Google Books suggests that stigma is used in this sense only by theologians, if we may so class Aleister Crowley; searching for digamma as a numeral shows it quite common, although the digamma function gives false positives.
• I don't think it's clear at all, and I believe I know what you want to say.
• But the last means all we have to do to make summary style is to link back here, and I've done that. Septentrionalis PMAnderson 00:37, 27 September 2011 (UTC)

Hmm. This is getting more and more puzzling. If "stigma" is not the most common term for the numeral, isn't that yet another argument against factoring it out into its own article under this title? And if you think the description of the relation between the two signs in the intro of digamma is confused, what good does it do to link to stigma from the section further down? Especially as this stigma article, in any of its versions, still offers the reader not a single word of additional information over and above what that section in digamma has? Fut.Perf. ☼ 06:06, 27 September 2011 (UTC)

About the name, I think Septentrionalis is right that stigma is not the usual name for the numeral. Reading the quotations (far above) literally, Heath seems to be the only one who thinks that the numeral sign is called stigma. Thompson says that the ligature is called stigma. Sophocles seems to say that the numeral sign is called vau or digamma (and Chambers agrees). Feldbausch gives a clear name for both, saying that the ligature is called sti or stigma, the numeral sign vau. Barry also gives a clear name for both, saying that the numeral sign is called episemon or sigma [sic?], the ligature stau. As for Küster, I don't know whether he is saying that the numeral, or the ligature, has the name stigma.

They appear not to be talking about classical names (though they don't make this clear); at least, neither episemon nor stigma has this sense in LSJ (though LSJ does give the names vau and digamma for the obsolete alphabetic character). The oldest of those sources quoted above is Feldbausch (1862).

None of these, not even Heath, says that both instances of this letter-shape have the same name. The only evidence I can think of right now that the two instances were conceptually linked is that modern Greek printers use the combination στ' for the sixth place in the alphabetical numeral system. They would only do this (I guess) because of an opinion that the numeral sign and the ligature were "the same". Andrew Dalby 12:05, 27 September 2011 (UTC)

I quite agree the naming practices are pretty complex. Just for the record, the list of refs above makes no claim to being complete or representative; I just quoted these to counter-balance Wareh's earlier claim that only "stigma" was used for the numeral sign, and calling it an incarnation of "digamma" was wrong. Fut.Perf. ☼ 14:08, 27 September 2011 (UTC)

I believe vau (spelt βαυ) is attested in a single Byzantine lexicographer, but he is assumed to have a much older source; this sense of stigma should, like the /st/ ligature, be a Byzantine innovation. Septentrionalis PMAnderson 14:56, 27 September 2011 (UTC)

Yes, no doubt it should, but I don't find it in Du Cange (though he had a special interest in manuscript symbols &c.) Andrew Dalby 15:26, 27 September 2011 (UTC)

"Stigma" certainly can't be older than the st-ligature itself, i.e. older than minuscule writing, but I can well imagine it might be even more recent. The earliest references to it that I've seen so far seem to be from the 19th century. Can you find out what names Du Cange uses for the numerals? (I'd be quite interested in his terms for Koppa and Sampi too, actually.) Fut.Perf. ☼ 15:42, 27 September 2011 (UTC)

Yes, so would I [for clarity, the book under discussion is Glossarium Mediae et Infimae Graecitatis, 1688]. Unfortunately, as far as I can see, letters-as-numerals falls between two of Du Cange's topics and doesn't get discussed at all, either in his medieval Greek grammar, which begins with the alphabet and the accents (pp. xx-xxii), or in his long tables of manuscript signs and abbreviations (4th section, columns 1-22). I looked up stigma in his main alphabetical glossary and addenda, and I've now looked up koppa and sampi there as well. No luck. Andrew Dalby 19:38, 27 September 2011 (UTC)

(left)

• βαυ is from Eustathius, who cites his source. It's also mentioned in Priscian, who is cited as using an actual digamma.
• Stigma for /st/ is not in E.A. Sophocles [Greek lexicon of the Roman and Byzantine periods; 1914, repr. 2005] either; although a modern Greek dictionary lists it. It may be entirely modern.
• E. A. Sophocles does list stigme "punctuation mark: period, comma, colon" (with eta) from all the standard lexicographers; the articles should be adapted; the two words will tend to fall together in Medieval Latin. (Digamma should also tell the reader that Greek punctuation differs from English.) Is there a source for stigma in this sense? Septentrionalis PMAnderson 16:40, 27 September 2011 (UTC)

Re-factored out from section below; hope nobody minds. Fut.Perf. ☼ 20:34, 27 September 2011 (UTC)

It isn't easy, is it? I'm sort of setting it out for myself below:

We are talking about two different letter shapes. One is called digamma; the other is now sometimes called stigma.

Digamma was a letter in some Greek alphabets, and as such got a place in the numeral system. The name has also been used by philologists for the corresponding almost-forgotten early Greek sound, whose recovery explained much about the scansion of Homeric verse, therefore also about the traditional nature of Homeric formulae and the curious use of hiatus by some later Greek poets. All that is very notable (not really covered in the digamma article: maybe somewhere else) and has nothing to do with stigma.

Stigma has been adopted by some as the name for the Greek ligature st. There's not much to say about that, but it has nothing to do with digamma.

Un peu d'histoire. Digamma was forgotten. Its symbol in the numeral system, called to episemon, came to be identical with the ligature that we call stigma. But, the ligature now also being nearly forgotten, that place in the numeral system has been restored (by some philologists) to the rediscovered digamma, while it has been given (by Greeks) to στ.

I think, since we are talking about two different letter-shapes, whose origins are quite different and whose uses and history only partly coincide, we need two articles. I think the current digamma article is incorrect and confusing in putting "stigma" in the first sentence: in terms of letter shapes, stigma is not digamma and never was. Andrew Dalby 16:26, 27 September 2011 (UTC)

I strongly disagree with this last statement, and the very references you commented on earlier today should prove why. The symbol shaped in late antiquity very certainly is digamma. What else could it be? And it is not of "quite different" origin from the original Ϝ; in fact, it is as much its direct continuation as "α" is of "Α". And it is exactly this symbol, with exactly this shape and in exactly this function, that is today called "stigma", for better or worse. So yes, stigma "is" digamma (leaving aside any further debate about "what the meaning of 'is' is"). Fut.Perf. ☼ 17:06, 27 September 2011 (UTC)

For ease of typing I will call them F and S: OK?

I admit Chambers says that S (as numeral) is called "digamma". I wouldn't regard Chambers as a reliable source on this, but I'm sure Wikipedia would :) That aside, I don't see any crossover.

No, according to the history given in digamma, it is not a direct continuation. The C symbol that developed out of the F became similar with, and then identical to, the S symbol that was used as a ligature. They had different origins. [Your diagram, above, shows this beautifully.]

I know no evidence that the numeral symbol that looked like a C, and eventually like an S, was ever called "digamma" (until Chambers). "Digamma" was a name used by ancient antiquarians and grammarians for the archaic F symbol: it describes the way the F symbol looks, one gamma on top of another one. We on Wikipedia describe the development of the numeral symbol from F via C to S under "digamma" -- where else can we do it? -- but I don't think we know whether there was a name for that symbol. After all, we don't really have a name for our equivalent. We just call it "six". Andrew Dalby 18:02, 27 September 2011 (UTC)

You overlook that the numeral ϛ, in pretty much exactly this form, was already used as a cursive on papyrus at a time overlapping with the use of Ϝ in stone inscriptions. And its root, the angular was a variant of alphabetic Ϝ in inscriptions from the very beginning. Numeral digamma didn't just "become similar with" the st-ligature. It was there all along and had the same shape as today, long before the st-ligature even existed. I don't have a citation for any contemporary author explicitly naming the numeral sign or identifying it with the "digamma aeolicum" either, right now, but significant parts of the literature are very clear about identifying it this way, which is all that matters (besides Thompson, quoted above, there's also Gardthausen, and Tod, loc. cit. as given on the digamma page). Fut.Perf. ☼ 18:53, 27 September 2011 (UTC)

Well, I've given my view, which is that there is a place for an article at stigma (ligature) (vel sim.), and that for the digamma article to name stigma in bold in the first sentence is misleading. I'll leave it to others now, with my best wishes for success! Andrew Dalby 19:15, 27 September 2011 (UTC)

Just for the record, another brief ref: Graham Flegg, Numbers through the ages (1989), p.88, "The old form of the letter digamma or vau was Ϝ, but the later form was Ϛ, which was also called stigma, because it served as an abbreviation for st." Instructive is the terminology chosen by Tod, "The alphabetic numeral system in Attica" jstor, talking about epigraphic numerals in postclassical Athens. He first identifies the numeral symbol in question as "that which we usually call ‘digamma’, then renders it as Ϝ in a table of symbols normalized to modern lowercase letters, and then gives an illustration showing that the symbol in question comprises a range of realizations including all manner of "stigma"-like S forms, including (while "Ϝ" itself, curiously, "does not appear as a numeral in Attica and very rarely elsewhere"). He also explicitly describes the S-like variants as "cursive developments" of , which is a canonical realization of both alphabetic and numeral Ϝ elsewhere. Fut.Perf. ☼ 20:22, 27 September 2011 (UTC)

Reader service

I see that Future Perfect disagrees with me on the fundamental question: what is most useful to the reader?

If I were a reader who met a reference to a letter or sign stigma, and looked it up on Wikipedia, I would expect to find my answer in an article headed Stigma or Stigma (letter). FP disagrees, and I'm not sure I understand his grounds for doing so. Septentrionalis PMAnderson 15:00, 27 September 2011 (UTC)

It might make sense to have a minimal article under stigma (letter) simply explaining that it is the name both of the ΣΤ ligature and of the Greek numeral six, and referring to the digamma article for everything else. I don't think it makes sense to duplicate any content (such as the development of the shapes) between the two articles. Interestingly, the 1911 Britannica only mentions the name 'stigma' (other than botanical uses) and for that matter the name 'vau' under the article on the letter 'F'. --Macrakis (talk) 15:19, 27 September 2011 (UTC)

The suggestion of a "minimal article" seems to meet one of Wareh's ideas from the other day. The problem I have with this is that it would leave the reader with more open questions than it would answer. The reader would be told that stigma has this double function – but he wouldn't understand why. In order to understand that, they'd still need to read everything that's now covered in digamma. That's why it would be more reader-friendly to send him there at once. What we have now in the stigma article after my recent tweaks [5] is, I feel, the absolute minimum of what the reader needs to even begin to understand how things are related – and that means an article that is hardly shorter than the digamma article itself. Fut.Perf. ☼ 15:33, 27 September 2011 (UTC)

(ec; response to Pma): Wikipedia treats synonyms together in the same article all the time; that is not the problem. The surprise felt by a reader on being redirected from stigma to digamma need not be greater than the surprise on being redirected from, say, Angma to velar nasal, or from Hyperesia to Aigeira. The problem is not with the questions our readers ask; it's with the answers. Because the answer to the question "what is 'stigma'?" and "what is 'digamma'?" involve precisely the same information. 80% of the meaning of these two terms overlaps, the two concepts denoting exactly the same thing; and the remaining 20% that fall exclusively under the denotation of only one of the two terms are still so important to the understanding of the overlap that they can't be omitted from the description of the other term either. That's why at the moment, the "stigma" article is an exact subset of the contents of the "digamma" article, and if we were to expand it to its fullest possibility,the two articles would become 100% identical in coverage. What I find not useful for the reader is the redundancy. It is not user-friendly to send them from one article to the other through hyperlinks, as if promising some additional useful information behind the link, only to make them read through precisely the same stuff again. Fut.Perf. ☼ 15:25, 27 September 2011 (UTC)

Well, I feel surprised by Aigeira. Why should any resource that uses Aegina do that? Septentrionalis PMAnderson 18:46, 27 September 2011 (UTC)

If you mean the difference between "ae" and "ai" spellings, that is obviously entirely irrelevant to the point I was making. Fut.Perf. ☼ 20:52, 27 September 2011 (UTC)

But the point you are making is equally dubious. All we say about Hyperesia is that Aegira was Homeric Hyperesia, and a one-sentence story from Pausanias about the name change. All of it is immediately behind the section redirect. (Now, as fact-checker, I doubt both claims are anywhere near as certain as the article claims they are; but that's for that page.)

This would not be feasible if it were not a section redirect, a new idea for this page. It would not be feasible if everything notable about Hyperesia weren't two or three sentences; we now have four or five paragraphs about stigma. Septentrionalis PMAnderson 21:04, 27 September 2011 (UTC)

No, we have no more than two sentences about stigma (the ligature). Everything else on this page, as I've been saying, is redundant stuff that really belongs into digamma but needs to be repeated here for the sole purpose of making the relation of the two articles understood. Fut.Perf. ☼ 21:36, 27 September 2011 (UTC)

That's not what I see; please state your two sentences.

I see one paragraph, out of half a dozen (the one beginning The numeral symbol, originally quite unrelated...) which could be so described. Septentrionalis PMAnderson 23:19, 27 September 2011 (UTC)

Add to that the passage about computer encoding (the Unicode codepoints, while named "stigma", are meant to be used exclusively for the numeral, never for the ligature as such, so they too belong in the digamma article). Add to that the passage on the name "stigma" (that name is of interest almost exclusively as a modern synonym for numeral digamma; we actually have not a single reference for anybody using it for the ligature as such, except in the context of explaining that the ligature is the reason for calling the numeral thus). Add to that the whole introduction, whose sole purpose is to solve our self-created problem of having to explain how the topics of the two articles are related. And the remaining meagre material about the ligature use as such is exactly that part of the article that has always been unsourced. Fut.Perf. ☼ 05:27, 28 September 2011 (UTC)

(left, again) It's easy to find references to the ligature as such if one looks. This, for example is expressly discussing /st/, not 6 - again a theologian, but they may be more likely than philologists to discuss a text symbol by symbol. Septentrionalis PMAnderson 13:53, 28 September 2011 (UTC)

Point taken, but still doesn't change the basic fact that this article is an exact subset of the other article and therefore 100% redundant. Despite all this debate, no explanation has yet been offered why this should not be a reason to merge it, which means I'm still determined to do just that, unless one of you guys goes ahead and actually expands it with some content that is not redundant. Fut.Perf. ☼ 11:14, 30 September 2011 (UTC)

We are not paper. We could afford to be redundant for clarity; there are at least three conceptually distinct, if intertwined, topics here, and we can afford to have an article on each of them. If we do not have an article on each of them, we should have self-contained sections (and I don't see them). But I am not convinced that this article (six paragraphs) can be completely redundant with the two sentences Fut. Perf. sees on stigma in digamma. Were it so, WP's normal response would be to shorten what is in digamma, and rely on the link. Septentrionalis PMAnderson 17:51, 30 September 2011 (UTC)

You are "not convinced" it can? For chrissake, then go and check. Compare them sentence against sentence, then show me the one sentence you find in Stigma that you didn't find in Digamma. And no, there is nothing in Digamma that can be shortened. Even those few bits that are more about the ligature than about the numeral are needed there also, to make the relation between the items understood. Fut.Perf. ☼ 21:39, 30 September 2011 (UTC)

Hi, I'm back. I'm not intentionally stirring up a conversation just to abandon it. You'll see my continuous active editing has suffered a rare interruption; I have been absent from Wikipedia for the last week, and I cannot return with full attention at the present moment either. So I apologize in advance for not engaging at once with everything above. I fear most of what I have to say now will not persuade Fut. Perf., but I humbly ask that it be sifted for points of agreement and possible compromise.

I'm glad you brought up the morphing of digamma into the numeral now called stigma (by the way, why is that image not used at digamma?). But it is now called stigma, right? Are you seriously claiming that a roomful of connoisseurs of texts that print ϛʹ, if they had to refer to what they saw in front of them, would not call it "stigma" but would call it "digamma"? (To put it bluntly, I claim that the recent edit at digamma of "Today the numeral sign is usually called stigma," changing "usually" to "often," is a blatant error. The circumstances in which anyone considers calling it digamma are, I say, an insignificant fraction of the total.) This is a living, functional glyph, and it has a living name among its present-day English-language friends: stigma. Historians of writing are welcome to point out that it has become conflated with something else, but we are not allowed to name topics according to standards of etymological purity. If so, we'd have to rename all kinds of articles where words have regrouped.

Look, I don't care if the article on the symbol ϛ is short, I just care if it's nonexistent (and I wouldn't care about that if the reason were that no one had treated the topic). I look and look at digamma and I don't find contained within it an overview of the existent topic (this symbol ϛ). If we renamed it digamma and stigma, and didn't make the readers interested in the stigma do so much work to know about ϛ (its different uses and pedigrees), then that would be another avenue I could accept.

Your tone about the 1% ("prove it") and elsewhere (~ "write your own damn content before you have an opinion about mine", shed-painting) does not seem helpful to me. In fact, once Oxford University Press etc. etc. chooses the symbol ϛ in an OCT to print a numeral, they are printing what takes its standardized form as a representative of a ligature. I use other, 19th c., printed editions that use ϛ for στ. It's the same symbol. The typographers and readers who use these texts have the same name, "stigma," for them. (Frankly, it would be worth treating the symbol ϛ directly just because of how widely pre- and post-digital publishers confuse it with ς).

In short, it seems Fut. Perf. happily concedes that the numeral ϛʹ "is today called stigma." Don't we name subjects after what they're called today, in English, rather than their etyma or their glyphs' etyma's names? Instead, we get an article digamma with the organization Digamma#Names#Stigma. This is an etymological approach and it takes the admirable tidiness and clarity of a single editor's mind, who thinks etymologically, and forces readers down the same theoretically determined path, against the categories of common usage, when the reader has the right to encyclopedic treatment precisely according to the categories of common usage. I really do respect your work, Fut. Perf., and I sympathize with your protective feelings towards it -- I am the same way -- but in this one case, based on all the discussion so far, I see this still as a point where the single-user-input to a topic (praiseworthy work) should not be allowed to dictate that we have to call stigmas digammas. (Falsify my "roomful of connoisseurs..." assumption above, and I will have to modify myself to say you only want to call some stigmas--the ligatures--digammas.)

Some further specific points:

1. "The reader would be told that stigma has this double function – but he wouldn't understand why." Perhaps we should continue discussing the short-page (or disambiguation, but see my doubt above) alternative, as this doesn't seem an unsolvable problem. As long as the page provides an index to everything there is to be said, it could probably satisfy me. If, to do that, we have to link to several different sections of a single article, I say, good! That begins to remedy precisely the no-unified-treatment problem.
2. Unicode codepoint. Unicode has an ου ligature, presumably meant to be used as an ου ligature. Unless there's a separate ligature codepoint for ϛ, it seems absurdly legalistic not to discuss the availability of the standardized ligature glyph because someone has prescribed not to use it as a ligature. Surely no font designer should code ϛ as ϝ?
3. When you say "several sources call ϛʹ digamma," are you talking about sources that discuss the history of the form ϛ in this context, rather than simply calling it something for the ordinary reason (i.e. because they need to refer to it by its accepted name)?
4. Given we have articles on other Greek ligatures, no less common (ου), then why not put aside the "1%" objection? Even if it were true (I don't accept it: for Classicists I'd say 75/25), the ligature considered in its own right is an encyclopedic topic. Once more, it's ok with me if this article refers to digamma for full discussion of the conflation and even the numeral usage. But I don't see how the article can fail to say that ϛʹ, in widespread common use, and called "stigma," is a numeral and a ligature both. Denying this means an inappropriately etymological definition of "is."

Wareh (talk) 15:35, 30 September 2011 (UTC)

I'm not going by "etymology"; that's a misrepresentation of what I'm saying. I'm going by unity of topic. The Ϛ-shaped numeral discussed by Tod (op cit), and which he calls digamma (and normalizes to "ϝ" in transcription) is the same thing as the numeral that the packhum.org inscription database normalizes as "ϛ" in the exact same inscriptions – no matter what the editors of packhum would call that sign. About your "roomfull of connaisseurs": is Gardthausen enough of a roomful? Gardthausen calls the numeral sign "digamma" throughout, but happily uses the "ϛ" symbol for transcribing it at the same time. Thompson does the same.

Here's another instructive passage from Gardthausen (Griechische Paleographie, p.238), discussing the minuscule treatment of "σ", my translationː "From the uncial form C derives the modern final ς […] This development in the later minuscule has its analogy in the late cursive, where the same signs had a different sense and a different history: Minuscule C looks the same as cursive ; the former develops into ς, while the latter, digamma, develops into , which incidentally still differs from the στ at that stage. […] στ is the most frequent and most important of all ligatures of σ, but it will have to be treated only later in connection with the numeral signs. The sign ϛ, which came to be identified with the digamma in the understanding of a later age, was used more and more often and slowly lost its erstwhile firmly closed form; was changed to , and by the end of the 11th century both forms often occur mixed even within the same manuscript."

Older authors up into the 19th century did that all the time (e.g. "the digamma, or episemon (ϛ)" [6]). A few more, both older and more recent: [7][8][9][10][11][12][13]. Many of the 19th-century authors are actually very explicit about saying that "ϛ as a ligature" and "ϛ as a numeral" are not the same thing (except under the purely practical perspective of the typesetter), and are insisting on using different names for each (typically "vau" or "episemon" for the numeral, and "stigma" or "stau/sti" for the ligature only).

About a codepoint and article for the ou-ligature: actually, no, there isn't, neither the one nor the other (my mistake in our original discussion last year). The Unicode codepoint is for Latin only; it exists because Algonquian uses it as a letter. It's derived from Greek, but it isn't the Greek ligature. About the reasons why Greek ligatures don't get Unicode codepoints, see this[14] very insightful essay by our resident expert User:Opoudjis. As I said above, not even stigma-as-a-ligature has a codepoint in Unicode. If you wanted to re-create Greek printed with ligatures, you'd still encode the underlying sequence of "στ" on the character level in the stored text, and leave the ligature substitution to some secondary mechanism during font rendering, just as is done with Latin "fi", "ff" and the like, in high-end typesetting systems that render those as ligatures.

You keep complaining that the digamma article doesn't treat "stigma" appropriately. If it doesn't, add to it. It currently mentions "stigma" in the very first sentence. It then mentions that stigma was also a ligature, within the very first paragraph. What more do you want? Show me a single statement in the current stigma article that is missing from digamma. There isn't any. If there was, I'd have no objection against a separate article. Write something genuine about the ligature and we're fine.

About "usually called stigma" vs. "often called stigma", please argue that with Pmanderson; it was him who called the "usually" into question.

By the way, if you want to do something useful, could you help in finding out what the story is about this apparent nonce use of "gamex" [15]? Fut.Perf. ☼ 19:18, 30 September 2011 (UTC)

Your sources do weigh with me, which is why it's frustrating to find garbage sources (Histories of Scientific Insights by Igor Ushakov? a first-year Greek textbook making a condensed point for students? these are distractions) mixed in with ones that might merit consideration (Feldbausch's footnote). Some don't even support your view, e.g. the Ante-Nicene Fathers footnote that equates the numeral six with a "short form of στ": that's just the conflation you don't want. Gardthausen is indeed instructive, but (as quoted) doesn't make your point. As to Tod, well, his normalization practice is very unevenly followed by editors of literary texts; moreover his interest is in describing ancient numerals. You won't find me anywhere suggesting that stigma was used in antiquity as a numeral.

We do have an article Ou (ligature), a ligature "which was frequently used in Byzantine MSS" (and into modern Greek texts, etc., which it should be expanded to say). I don't see why we can't have an article on the στ ligature. If you want to cite Feldbausch's footnote to say that the stigma-numeral is properly called digamma, and point the reader to digamma for discussion of the numeral, I have no objection to that. How does this not solve our disagreement?

So I'm a bit disturbed that I could use such a very superficial Google-Book-link adducing methodology to produce this (multiplied over and over): that stigma is the Attic numeral six. (Here Buttmann even says, "its coincidence in form with the Βαῦ is only accidental." I'm prepared to trust you that this is rank error, but... it hardly helps the case against helping the legions of people, 19th, 20th, and 21st century people, who see the numeral as stigma and call it that.

I still find digamma hard to digest as the right place to house a discussion of the ligature. However, I don't have an objection if you want to put in two clear lines at stigma, one for the stigma (ligature) and one for stigma (numeral), with some intelligent statement, and then I guess both redirect to digamma. (I think if you'll look above you'll find I've been open to such a simple solution all along, and I don't think I have been forcing us to spend this much hot air on the subject.) Wareh (talk) 21:09, 30 September 2011 (UTC)

You are misreading the Buttmann source. What he calls "Βαῦ" is ϛ-as-a-numeral, and the subject "it" in the sentence you quote above is ϛ-as-a-ligature; it is exactly the mistaken identity between these two that he is rejecting. He is one of the authors who are vocally insisting that ϛ-as-a-numeral is not "stigma". You must also be misreading the "Anti-Nicene fathers" ref, if you think it doesn't support my point; it's calling numeric "ϛ" "diɡamma", which is all that mattered at this point. (The fact that it's also mentioning the "st" use in the same breath additionally strenghthens rather than weakens the case for saying that they are all too closely intertwined to separate them into two articles, but that's not the point I was making). You must also be misreading Gardthausen in some way; I said he was calling numeric "ϛ" "digamma", which is very clearly what he does.

A dab page with two entries that both ultimately link to the same page through two different redirects? That's about the most absurdly convoluted way of leading a reader into a blind alley I can imagine.

You keep asking why "we can't have an article on the ligature", and I keep responding: sure, we can have one, the moment you write one. Find something to say about stigma-the-ligature that exceeds the minimum amount of information we also need for independent reasons at digamma (not because it's centrally part of the topic there, but because it's needed to make the overlap understood), and we're agreed. Fut.Perf. ☼ 21:23, 30 September 2011 (UTC)