Talk:English alphabet

From Wikipedia, the free encyclopedia

You have new messages (last change).

For previous discussion relating to deletion of this page, please see Talk:French alphabet/deletion

This has nothing to do with English alphabet. What are you trying for? Rmhermen 00:33, Nov 25, 2003 (UTC)

Contents

[edit] Pronunciation

Since someone has kindly added IPA pronunciation, the SAMPA is redundant and clutters the chart. Is there any reason not to delete it? —Michael Z. 07:12, 2004 Dec 20 (UTC)

The best thing to do when there are mistakes is to fix the mistakes, rather than obliterating the entire page. --Evice 21:02, Dec 24, 2004 (UTC)
What I mean is: is the IPA alone sufficient to convey pronunciation? If so, can I delete the SAMPA to make the page less cluttered? —Michael Z. 21:54, 2004 Dec 24 (UTC)
Well, the page for SAMPA does say IPA is preferred over SAMPA. --Evice 00:00, Dec 27, 2004 (UTC)
Well, that took five hours to get reverted, and in the middle of the night. I don't know why I post questions like these on the talk pages; someone who disagrees only comes by after the change is made. Oh, well. —Michael Z. 21:05, 2004 Dec 27 (UTC)

[edit] Curious

Curious, under what circumstance is "W" used as a vowel?

That sentence points to semivowel, which explains it a bit.

W is a vowel in Welsh, pronounced "oo", and therefore in a few Welsh words that have become (moreor less) naturalised in English, such as cwm, crwth. rossb 19:31, 25 Feb 2005 (UTC)
In the word kwm (pronounced "koom").Cameron Nedland 23:58, 2 March 2006 (UTC)

[edit] Sections

I thought splitting "notes" up into 2 sections made the article much easier to read. Why was it reverted?66.99.9.30 18:54, 31 Jan 2005 (UTC)

[edit] Yogh

I am aware that yogh didn't just stand for /g/. I am talking about the grapheme. They didn't just make up the symbol, it is based on the shape of the latin letter g as written by Irish and Anglo-Saxon scribes, see e.g. [1]. dab (ᛏ) 13:34, 16 Mar 2005 (UTC)

That's one theory. However, Yogh was used for completely different sounds than G. It is just plain wrong to state that Yogh derived from G. Jordi·✆ 13:44, 16 Mar 2005 (UTC)
I'm sorry. There seems to be a misunderstanding. Do you agree we are talking about graphemes, not phonemes? Of course a Yogh is different from G, just as Eth is different from D. W is not "also" for UU, because when it was a ligature, there was no distinction between UU and VV (U was introduced in the 16th century, i.e. in Early Modern English). Now, what other theories for the origin of the grapheme Yogh are there? I thought it was very obvious it is a variant of g, but I'm willing to be educated, of course. dab (ᛏ) 14:24, 16 Mar 2005 (UTC)
The letter yogh is derived from Insular g, but it is not identical to Insular g. It is not derived from Carolingian g. I have fixed the text. Please see http://www.evertype.com/standards/wynnyogh/ezhyogh.html Evertype 14:27, 2005 Mar 16 (UTC)
that's what I meant all along. My statement was meant as 'ultimately descended from latin g', nothing else. dab (ᛏ) 14:50, 16 Mar 2005 (UTC)

sigh, I am not sure what is so contentious about all this, but somehow yogh and wynn are mixed up in the present version (I mean the glyphs. or is it just my browser??). if you agree that they are, could somebody fix that please? dab (ᛏ) 14:08, 21 Mar 2005 (UTC)

The characters were swapped in the introduction. For reference, wynn looks like a p, and yogh looks like a 3. Jordi·✆ 14:21, 21 Mar 2005 (UTC)

[edit] spellings

I added the modern names of the letters per the OED: letter + ee, or e + letter, for the consonants, and the vowels for themselves, with the exceptions of aitch, jay, kay, cue, ar, ess, double-u, wye, zed. Variants occur, such as obsolete cu, qu, que, kue, kewe for Q, and in verbal forms, such as effing. --kwami 03:08, 26 Apr 2005 (UTC)

This is a stupid idea. The IPA pronunciations are enough, without adding imprecise pronunciations. Could someone remove these inaccurate pronunciations of the letters, leaving just the correct IPA pronunciations. -- Mark 11:52, 10 May 2005 (UTC)
They aren't "pronunciations", they're the standard spellings of the names of the letters! You wouldn't remove the names alpha, beta etc. from the Greek alphabet because "the IPA pronunciations" [alfa, vita] "are enough", would you? And where else are you going to go to find out how to spell the name of the letter Y, or H? These are all from the OED, but it took me a while to find Y listed under the letter W! kwami 21:53, 2005 May 10 (UTC)

[edit] Pronunciation

Phonemic pronunciation /.../ is pretty well accepted for most purposes in Wikipedia, unless there's a specific reason to be more specific. The IPA representations here can be shown with slashes as they are, but in a few cases I might even simplify them. E.g., /keɪ/ would be acceptable for [kʰeɪ], and /el/ for [ɛɫ].

If you're going to use brackets [...] to indicate precise phonetic pronunciation of the letter names, then you should indicate what accent these represent. You may as well also start adding versions for different accents, like for Alabama, New York, Vancouver, London, Glasgow, Dublin, Delhi, etc. —Michael Z. 2005-05-15 20:33 Z

I changed them to brackets because a phonetic pronunciation was being indicated. Phonemic would be fine, but slashes with phonetic pronunciation is not. That is, if we change to slashes, we need to phonemicize the transcription: if other pages are sloppy in this regard, then they need to be fixed, rather than us being sloppy too. We should simplify more than you suggest, however. For instance, K would be /keː/ (if L were /el/), or /ke/ (if L were /ɛl/). [That's a theoretical stance, whether the difference between these vowels is phonemically one of length, or if they're different vowels like /i/ and /a/.] We shouldn't show a diphthong, because the vowel isn't phonemically a diphthong. Likewise, O would be /oː/ or /o/. H would be /eːc/ or /ec/: Since the ch sound is a single phoneme, it should be transcribed with a single letter. Anything in the IPA that's reasonably close, like /c/, would be appropriate.
However, the point of this table is to tell people how to pronounce the letters, not to make a theoretical point as to which sounds are phonemically contrastive in English. Since pronunciation guides are by their nature phonetic, a phonetic transcription would seem appropriate. Of course, we can change the level of phonetic detail: lose the aspiration marks, lose the diphthongs, lose the velarized el, etc. That's not a theoretical issue, just a judgement call of how much detail is appropriate. But this approach would require brackets. kwami 23:44, 2005 May 15 (UTC)
I'm a little confused. Bear with me, because I learned IPA on-line and am not very clear on the finer differences between some of the symbols. But I don't see how /ec/ sounds at all like the name of the letter H; wouldn't that sound more like the "ets" in "nets" (or perhaps like a Slavic "еть" /etʲ/)? /eɪʧ/, or maybe /ejʧ/ look right to me (I replaced the two-letter tsh sound with a ligature). And /ke:/ wouldn't get the name across to me either; it's "the letter kay". I suppose a Scot might say the letter name that way, but then if he read "/kej/", it may sound like that to me too.
Personally, I think it's good to keep it very simple because
  1. It's easier to read if one is not very familiar with IPA. Our audience isn't linguists, it's the general reader.
  2. We're not trying to duplicate the precise phonetic pronunciation in a particular dialect; we're just showing the names of the letters.
/el/ seems just as good to me as /ɛɫ/ at getting the name of the letter L across. When I pronounce /kju:/, I will probably automatically add the (superscript-h) aspiration, but for all I know, there are English accents that wouldn't—and I don't think it matters. —Michael Z. 2005-05-16 02:11 Z

On p. 791 of World's Writing Systems, the first four letters of the alphabet are referred to as [eːj], [biːj], [siːj], [diːj]. Unfortunately, I can't find the rest of the letters in that book, but it's as authoritative a reference as there is, I think. —Michael Z. 2005-05-16 03:10 Z

Hi, Michael. As for the /ec/ notation, I was trying to make a point about phonemic spellings being a bit too abstract for a pronunciation guide, in my opinion. (IPA [c] (in brackets) is a palatal stop, and in most languages that have it, it's pretty close to English ch.)
World's Writing Systems is a great book for all the stuff in it, but the quality is uneven. (I mean, the chapter on Japanese is terrible.) On p. 791, O'Conner simply got it wrong. I'd check out Ladefoged's intro college text A Course in Phonetics for English and a general intro to phonetics. It's outrageously priced, but it's ubiquitous and goes through so many new editions (with little point other than making more money) that it should be easy to pick up a used copy somewhere. (Another book of his, The Sounds of the World's Languages, covers a lot more territory and at a lower price, and is a great book, but doesn't have much room for English.)
There are three common approaches to representing English "long" and "short" vowels. (I'll use [i], as in sit and seat, for an example, cuz there's a lot of dialectical variation with the low and back vowels.)
  1. The difference, essentially, is phonemically length, and the rest is phonetic detail, so they can be best transcribed as [i] and [iː].
  2. The difference is essentially quality (vowel height for some, "tense/lax" for others, etc.), and the length is detail, so they can be best transcribed [ɪ] and [i]. Or these two could be combined, for [ɪ]/[iː].
  3. The "long" vowels are really diphthongs: [ɪ] and [ij], or maybe [ɪ] and [ɪj].
You wouldn't have a vowel that's both long and a diphthong, like O'Conner. Everyone agrees there's a single feature here, even if they can't agree on what it is.
Regardless of which system you choose, a native speaker will get it right. But we're not just aiming for the native speaker, which is why I think the little details like aspiration on Q might be worth keeping. If you actually said [kju:], a native speaker would hear gew. This actually happened to me: I lost my aspiration while living abroad, and when I got back, my own mother couldn't understand me. When I said take, she heard day!
Now, there are underlying theoretical assumptions behind the three systems, but much of the disagreement may be due to the dialect of the person making the argument. For me, the "long" vowels are clearly diphthongs (I can most easily hear it in pairs like pill vs. peel, with peel almost being a syllable and a half, but I can feel my tongue move in a word like sea too). But my phonetics prof, who was from the opposite side of the country, insisted that the difference was length. However, according to Ladefoged (as good an authority as you're gonna get), for most of the US, the long vowels are phonetically diphthongs. This is very likely not true for the UK, or at least not for all of it.
BTW, for the mid vowels (set, sate), these approaches would give you [e]/[eː]; [ɛ]/[e]; and either [e]/[ej] or [ɛ]/[ɛɪ̯] or [ɛ]/[eɪ̯], etc. So the person who wrote the article decided that length was the critical feature for the high vowels, and for mid vowels it was the diphthong. That's pretty common, as the diphthong's easier to hear in the mid vowels.
So, IMHO, I think we should stick to a phonetic representation. Personally, I'd include a lot of phonetic detail like aspiration and the velarized el, for the non-native speaker, even if it's redundant for you and me. I'd also make all the vowels diphthongs ([ej, bij, sij, dij, ij, ɛf], etc.), because that's what I hear, but I'd expect someone who doesn't have that feature (or maybe who can't hear it) to object. There is no one correct answer, and it doesn't really matter a whole lot, except that someone kept putting the phonetic transcriptions between /slashes/, which was making a theoretical claim that was just wrong. kwami 07:06, 2005 May 16 (UTC)
Thanks for the patient explanation. Since all those experts can't agree on an authoritative representation, I'll feel a little more confident about my own attempts at formulating pronunciation in IPA. I think it would be useful to put some of this into the article on IPA, to show that the system isn't absolute but subject to interpretation.
I'm Western Canadian, so I think there may be a touch less diphthong in some of my vowels, but I see what you mean. When I say the name of the letter "C", there's a very subtle /j/ on the end, or maybe it's a sort of stop or release. But in some phrases the word "see" sounds like /si:/ or even /si/. I definitely perceive a difference in quality, not just length, between short and long vowels.
I can see some of the benefits of using phonetic transcription here, but I'm still not sure that that's the right way to go. The best approach seems to depend on the audience. Your pronunciation of the word "take" changed over time, but neither version was wrong. Couldn't the way you said it in both periods be phonemically transcribed as /teɪk/? In this context, shouldn't we try to show how the basic word is said, without relating minor features of regional accent?
I'll look for those books. There doesn't seem to be a copy of WWS in any library in my city, including two universities. I've been relying on Amazon.com's "search inside the book" to do a little reference. Looks like there are several Ladefoged books in the local University library. Cheers. —Michael Z. 2005-05-16 16:11 Z

By the way, your input would be useful at Talk:International Phonetic Alphabet for English#Pronunciation guides in Wikipedia articles. —Michael Z. 2005-05-16 16:17 Z

Oh, my pronunciation of take was definitely wrong — when your friends and family can't understand you, there's st that's not right! And yes, the diphthong in [ij] is quite subtle, which is why many people don't transcribe it. Also, I think all major English dialects have a qualitative difference in long & short vowels (at least, if they have long & short vowels); the question is not whether it exists, but whether it's important enough to transcribe — like aspiration, it's a judgement call: Is it relevant & informative, or just clutter?
And when I lost my aspiration, I also lost the diphthong. So it was more like /tek/. If you're gonna make the transcription phonemic, you'll need to decide whether the diphthong is essential or not. My impression is that it's mere phonetic detail, and shouldn't be there. If you buy that, then you need to decide if the relevant factor is quantity or quality. I have absolutely no idea (again, it's a model of a theoretical claim, not reality!), but we shouldn't have both. kwami 21:22, 2005 May 16 (UTC)
There is something that for lack of a better term we might call "dictionary standard IPA for English", being defined as generally the type of IPA transcription used in dictionaries, whether in the actual entries—as in most modern British dictionaries, like the Oxford and Cambridge dictionaries—or in the front matter which correlates the dictionary-specific phonetic symbols to the IPA—as in most modern American dictionaries, like the Merriam-Webster and the American Heritage. These dictionaries generally give the set of English phonemic monophthongs as something resembling this: [i ɪ eɪ ɛ æ ɑ ɒ ɔ ʌ oʊ ʊ u ɜ], with a few variations. The variations are:
  • including the length marker for the "long" vowels for British English
  • changing oʊ to əʊ for British English
  • merging ɑ and ɒ in American English
I didn't include the rhotics, the diphthongs or the reduced vowels, but the same principles apply. International Phonetic Alphabet for English explicates most of the symbols.
Even though the actual pronunciation of English varies (sometimes dramatically) from this standard, and even though this standard may even have some theoretical problems particularly with the representation of vowels, we should stick to it as much as possible because it's a common reference point. If we use a different set of phonemes for transcribing English on every page, we end up confusing readers who try to figure what the symbols mean and not finding the symbols they encounter on the reference pages. Some variation will of course be inevitable, as even the dictionary makers don't agree on the symbols, but generally sticking to the beaten paths is what we should do here. Wikipedia:No original research. Nohat 22:21, 18 May 2005 (UTC)

[edit] Phonemic transcription

We're getting into a bit of an edit war here. An anonymous user keeps reverting to phonemic slash notation, but mixed up with a phonetic transcription. I'll change it to a phonemic transcription to see it that's more satisfactory. Of course, this won't tell a non-native speaker how to pronounce the letters, but at least it won't be so misleading.

Of course, this requires a theoretical basis, as to whether the difference of the vowels in set and sate is essentially one of height, length, tense/lax, or monophthong/diphthong. I'm going to go with height. kwami 00:11, 2005 May 18 (UTC)

I'm reverting to my last serious edit. If anyone else wants to contribute, great. You might want to make a comment here so that your contribution doesn't get lost in the reverts: I'm going to automatically revert any edits by our anonymous editor (who by the way has received multiple requests to stop vandalizing pages). Unless he wants to join in the discussion, of course. kwami 20:38, 2005 May 18 (UTC)
Wikipedia:Manual of Style (pronunciation), which is based on the official Handbook of the International Phonetic Association, says that phoneme slashes should only be used if the distinction between phones and phonemes is significant in the use at hand. Here, the distinction is not important, so, following policy, the correct solution is to enclose minimal symbols (that represent phonemes) in square brackets. Nohat 21:57, 18 May 2005 (UTC)
It's not policy. And it's counter to the overwhelming usage on Wikipedia. It's an edit someone made, and I subsequently disputed, but hasn't been resolved yet. Operating on the theory that most Wikipedia readers aren't familiar with the Handbook of the International Phonetic Association, I'm recommending consistently using slashes for phonemic transcriptions. How would one know whether "the distinction between phones and phonemes is significant in the use at hand"? —Michael Z. 2005-05-19 02:34 Z
As in, when the article is distinguishing between phones and phonemes, and it's important to indicate the distinction, like when, for example, explaning the various rules for the realization of phonemes. For example, in English, you would say /k/ -> [kʰ] before stressed vowels; /l/ -> [ɫ] in syllable codas, etc. In almost every other case, the distinction is not important, and square brackets should be used. Phonemic analyses are not always universally agreed upon, and the most neutral description would be square brackets, which merely say "this contains phonetic information, which may or may not be minimally phonemic, but we take no stand as to whether it is phonemic", whereas slashes come with the implied assumption that everything within the slash is necessarily a phoneme and the implied assumption that the phonemes indicated are in fact separate phonemes of the language, which may or may not be disputed. It's better to just stick with the punctuation that makes the fewest number of assumptions about what it's punctuating. Not only does this policy adhere to the policy for official IPA usage, but it also adheres more easily to the Wikipedia policy of the Neutral Point of View. Nohat 03:07, 19 May 2005 (UTC)
So then in practice, would you only use slashes in contrast to a bracketed transcription on the same page? —Michael Z. 2005-07-9 07:36 Z

[edit] 7

is there a unicode character for the 'and' symbol used in Old English usually transcribed with '7'? dab (ᛏ) 8 July 2005 17:01 (UTC)

Far as I can tell from the Unicode index, no. Usually, graphic variants are not encoded in Unicode, but rather in the font. So if you had an Old English font, the normal /&/ would presumably display as [7], just as the /g/ might look like a yogh. kwami 2005 July 9 02:17 (UTC)
well there you have it, there is a glyph for yogh, and for wynn, so why not for 7? dab (ᛏ) 9 July 2005 06:17 (UTC)
Because wynn was a separate letter of the alphabet, and yogh became one as it diverged from gee. What you're asking for is merely a variant form of <&>. It's rather like asking for a Unicode value for cursive R. There's actually a font that is entirely ampersands - over 30 variant forms, I believe. That said, I believe there is an encoding for the Classical Roman ampersand; I don't know what motivating that particular choice. (There's also a Greek ampersand, based on the letter kappa, but as far as I know it's not supported by Unicode either.) Unicode is inconsistant at times. kwami 2005 July 9 06:29 (UTC)
it's ok, I understand Unicode. You may as well say that the wynn rune is a variant of wynn, or that w is a variant of wynn, or that Greek Epsilon is a variant of E, since '&' is for Latin et and 7 is for Old English and. But if there isn't a glyph, well, maybe in some future version, no problem. dab (ᛏ) 9 July 2005 08:58 (UTC)
The Unicode Consortium is really annoying in this. You may have more luck with the Medieval Unicode Font Initiative, which includes the sign at F142. If you have Junicode or Leeds United installed this will show it: . Jordi·✆ 9 July 2005 13:26 (UTC)
Also note that strictly speaking, in Medieval Latin as well as the English language up to ca. 1800, is the capital form of &. Within Unicode & is a character which lacks a capital form. Jordi·✆ 9 July 2005 13:37 (UTC)

ok, I can't use these in article namespace anyway. I was looking for the best way to transcribe the Kirkdale sundial inscription, and I'm just using '7' now. There is a reason people use "incorrect" '7', and not '&' (which is readily availabe) for Old English, indicative of the fact that it is not considered the same character, so imho, there really should be a unicode assignment. dab (ᛏ) 9 July 2005 13:46 (UTC)

As a side note, guess why the ampersand is placed where it is on Qwerty keyboards? Jordi·✆ 9 July 2005 13:50 (UTC)
Cool! kwami 2005 July 9 19:10 (UTC)
surely this is a coincidence? After all, the first typewriters were not built for or by medievalists? dab (ᛏ) 12:19, 10 July 2005 (UTC)

[edit] Alphabetical Order

I am curious about the history of Alphabetical order. I do not see this topic addressed, and when I searched the Internet using Google, I found only pages that made references to books or other pages that no longer existed.

I remember hearing once that the "Alphabet Song" was the first time that Alphabetical Order was created, but I have no documents to back this up, and I know that is likely only an urban legend.

Also, would this history belong in this page about the Enlish Alphabet, or under Collation (which discusses Alphabetical Order, but not the history of it) or in its own page?

Thanks, Tetrakatus

It's much, much older than the "alphabet song". IIRC the oldest known example of alphabetical order is on a clay tablet found in the ruins of Ugarit, dating back to 1400 to 1200 B.C.. It contains the Ugarit alphabet in order from àlfu to tawru, but it is believed to have held the full Ugarit alphabet before being broken. See here. Ugarit even had an "alphabet poem"! However alphabetical order is probably much, much older than that even. Jordi·✆ 18:25, 5 August 2005 (UTC)
The alphabet comes from Latin; it is because these letters were all in the same order in Latin except J, U, and W, which did not exist. J was put right after I because it is just a form, and W after V for a similar reason. For why U was place before V, I'm not sure. In turn, the Latin alphabet comes mostly from Greek, and had only 3 letters, G, Y, and Z, inconsistent with Greek. For Y and Z, the reason is probably because these are the letters used only in words borrowed from Greek. For G being located in its current position, I think it is because they wanted G to be put in a location that formerly belonged to a letter Latin didn't need; zeta (which also happens to be one of the letters put at the end for use in words borrowed from Greek.) In turn, I'm not sure how the order of the Greek alphabet turned out to be that way; but if you know, feel free to answer. 66.32.221.106 00:48, 30 September 2005 (UTC)
Some of this info is in the Alphabet, History of the alphabet, or Middle Bronze Age alphabets articles (I forget which). All we know for Egyptian is that one of their dictionaries started with H for Ibis (originally Hibis), because the Ibis was the totem of Thoth, the patron god of writing. We don't know if they actually had an alphabetic order. There are two ancient orders, the North Semitic and South Semitic order, both attested in Ugaritic. (The southern order survives today in Ethiopic.) We have absolutely no idea where they came from, or what motivated the order. It could be a rhyme, there could be religious significance, or it could simply be completely arbitrary (although somehow I doubt the latter). The Semitic alphabet was gradually whittled down to 22 letters in Phoenician, but the northern order was not changed except for dropping out letters that weren't needed in one dialect or another along the way, and maybe confusing some of the S sounds.
The longer orders you can find at Ugaritic alphabet. The "Reduced Levantine abjad" (used for Hebrew and Phoenician) was,
’ b g d h w z ħ ţ y k l m n s c p ş q r š t
The Greeks kept this order (turning the consonant letters ’ h y c into the vowel letters a e i o), split w into w plus u (they kept w where it was and placed u at the end of the alphabet), and then tacked on a couple new letters (one of which remains with us as X). They later dropped the letters w (digamma), s (sampi), and q.
The Etruscans got their alphabet from the Western Greeks, whose alphabet was a little different from the Eastern Greeks. For example, the Western Greeks kept the Phoenician letter ħ as an [h], rather than as the vowel eta. (This is our letter <H>.) Also, that X-shaped letter stood for [ks] rather than for [kh]. (This is our letter <X>.) Also, the Etruscans didn't have a [g] sound, so they pronounced the letter g as [k]. (This is our letter <C>.) They pronounced the letter z as [ts], and the letter w as [v]. For the sound [f], which didn't exist in Greek, they used the digraph wh.
The Romans borrowed this alphabet, and dropped most of the letters they didn't need. They didn't have a [v] sound, and since they therefore never wrote w for [v], they simplified the wh digraph for [f] to just w. (This is our letter <F>.) They didn't need z and dropped it, and they dropped everything after X. Now, the Romans did have a [g] sound, so they created a new letter for [g] out of the original letter g (which was by then Roman <C>), which they did by adding an extra line to it. That is, <G> was derived from <C> the same way the letters Ç, Ć, and Č were.
So why did the Romans put G where z used to be? Well, they used the alphabet not only as letters, but as numerals. <A> was used for 1, <B> for 2, <C> for 3, etc. <Z> was number 7. When they dropped <Z>, they lost a numeral, and they used the new letter <G> to fill the gap. (Long after the Greeks had dropped digamma and sanpi as letters, they kept using them as numbers, before later substituting new symbols, so this was a normal thing to do.)
Now the Romans had the alphabet
A B C D E F G H I K L M N O P Q R S T U X
<V>, by the way, was the vowel [u]. Remember, the Romans had no [v] sound. The difference between <U> and <V> at this stage was just one of style, like the two forms of lower-case <a> and <g> in the English alphabet.
Meanwhile, Greek had changed. The sound [u] had shifted to [y] (that is, to the sound of a German ü). So when the Romans conquered Greece, and started borrowing lots of Greek words, they had two new sounds they couldn't write, [z] and [y]. So they borrowed the Greek letters and stuck them on to the end of the Latin alphabet, our letters Y (still known as "Greek I" in Spanish) and Z (still called "zed" (from "zeta") in the UK). (Other Greek sounds they wrote as digraphs, CH, TH, PH.)
In the medieval period, the Romance languages had changed, and the Latin alphabet was also used to write Germanic and Celtic languages. Now, many of these languages had both a [u] and a [v], both an [i] and a [j], so eventually people started differentiating <I> and <J>, and <V>, <U>, <W> (<W> is still known as "German V" in some languages). They also added the letter <&> to the end of the alphabet (for 27 letters), which remained until the end of the 19th century. This is how <&> got the name "ampersand": When you recited the alphabet, and got to the end, you sang "double-u, ex, wy, zed, and per se 'and'." Now, little children didn't understand that Latin per se meant 'by itself', and that what they were really saying was something like "and the letter that by itself means 'and'," so they slurred the rhyme together and ended it as "double-u, ex, wy, zed, ampersand". (The <&> was later dropped for being confusing. By the way, it is a ligature of <Et>, the Latin word for 'and'.) kwami 06:40, 30 September 2005 (UTC)
I can't find any evidence about the Romans using Z for 7; I always thought the Romans used VII for 7. Georgia guy 18:35, 12 November 2005 (UTC)
They did. For writing, arithmatic, accounting, taxes, etc, they used the Roman numerals. However, the alphabet also had a numerical order to it. It wasn't used for much - perhaps it was just a mystical association - but the idea is that the tradition was strong enough, or perhaps Greek influence was strong enough, for a gap to be left at position #7 when the letter Z was dropped, and that this gap was filled with the new letter G. I don't know any details, but I've come similar accounts several places. kwami 19:54, 12 November 2005 (UTC)

[edit] Dotting your wyes

I've just deleted this from the notes section of the article:

(When the letter wy was adopted during Middle English, it had a dot over it to distinguish it from thorn, so in such archaisms y without a dot is thorn, not wy.)

It's problematic for a few reasons. Firstly, wy was used in Old English (when it meant IPA /y/ and /yː/), rather than being adopted during Middle English (when it came to mean /j/, /i/ etc.). (In fact, I understand—but I’m not sure from where—that the name wy comes from an attempt to maintain the labialness of /yː/ to help distinguish it from the letter /iː/, with which it would otherwise have become homophonous. Of course, I don't know where I heard that from so I could've just made it up...)

Secondly, thorn and wy were often drawn the same. Sometimes with a wy-like glyph, other times with a thorn-like glyph. They were also often drawn differently, when wy was usually drawn with a wy-like glyph, and thorn with a thorn-like one. Although the dotless thorn/dotfull wye distinction might've been made by some scribes, it was not as common (and certainly not nearly universal) as the sentence implies.

—Felix the Cassowary (ɑe hɪː jɐ) 11:58, 25 October 2005 (UTC)

I'd imagine that a dot would only make a wye look more like a thorn. Jimp 07:20, 21 December 2005 (UTC)

[edit] Ampersand as 27th letter

What is the source for this bit of the article? I don't doubt the information but I've never seen it in literature.

Can't find a good source offhand, though I've seen it several places. Here's a citation from the OED:
1882 FREEMAN in Longm. Mag. I. 95 ‘Ampussy and,’ that is, in full ‘and per se, and,’ is the name of the sign for the conjunction and, &, which used to be printed at the end of the alphabet.
Alt spellings include amsiam.
Also found [2],
1.OEngl.1011. Byrhtferð. 1929. Byrhtferth’s Manual (AD 1011) now edited for the first time from MS. Ashmole 328 in the Bodleian library. Samuel John Crawford, ed. (Early English Text Society, OS; 177) London: Oxford University Press.
[On p. 203 of his manuscript, Byrhtferð gives what could be considered an Old English character set, namely the Latin alphabet (which includes the sign & ‘ampersand’), and the Old English additional letters, as well as the nota ⁊ ‘ond’). Giving also the numeric values, Byrhtferð shows a concern for ordering, though his 11th-century interests are not, perhaps, the same as our own:
“Heræfter we wyllað openian uplendiscum preostum þæra <stafena> gerena æfter Lydenwara gesceade. Ærest we willað hig amearkian togædere, ⁊ syððan heora todælednyssa we willað gekyðan on þa wisan þe þa boceras habbað ⁊ healdað; ⁊ eac we willað þa stafas onsundron gewriðan þe þa estfullan preostas on heora getæle habbað ⁊ þæræfter Ebreiscra abecede we willað geswutelian, ⁊ Grecisra. ⁊ þæt getæl þæra stafena we þencað to cyþanne, forþon we witon þæt hyt mæg fremian.”
(‘We will next reveal to country priests the mysteries of the letters of the alphabet in accordance with the reasoning of the Romans. First of all we will write them down together, and then we will make known their divisions in the manner which scholars have and hold, and likewise we will group the letters separately, which devout priests have in their reckoning, and afterwards we will set forth the alphabet of the Hebrews, and that of the Greeks; and it is our intention to make known the numerical value of the letters, because we know that it may be of advantage.’)”
Latin A B C D E F G H I K L M N O P Q R S T V X Y Z ⁊
English A B C D E F G H I K L M N O P Q R S T V X Y Z & ⁊ Ƿ Þ Ð Æ
kwami 02:42, 20 January 2006 (UTC)

Still, what shall we do with "⁊"? This is the glyph for "not", not the Anglo-Saxon glyph for "and", and the similarity is coincidential; we may as well use "7". dab (ᛏ) 16:19, 8 February 2006 (UTC)

"⁊" is U+204A TIRONIAN SIGN ET, the "and" sign itself. The glyph for "not" is "¬" U+00AC and should look somewhat different. —Muke Tever talk (la.wiktionary) 19:52, 10 February 2006 (UTC)
great, thank you! dab (ᛏ) 13:54, 15 September 2006 (UTC)

[edit] Moved back to English alphabet

I thought the current title was self explainatory, but ChrisW felt that moving it to Abcdefghijklmnopqrstuvwxyz would make it move easily accessable. How?
The title "Abcdefghijklmnopqrstuvwxyz" to me is long and seems a bit silly to use as a title of an article, and I don't think it makes the subject any clearer, then the title "English alphabet". Marco79 13:30, 1 May 2006 (UTC)

Retrieved from "http://en.wikipedia.org../../../e/n/g/Talk%7EEnglish_alphabet_e765.html"