View source for Talk:Pronunciation of English ⟨th⟩

{{WikiProject banner shell|class=b|
{{WikiProject English Language}}
{{WPLING|phonetics=yes}}
}}

{{ArticleHistory
|action1=GAN
|action1date=22:07, 3 June 2006
|action1result=not listed
|action1oldid=55583869
|currentstatus=FGAN
}}
{{split article
|from=T
|from_oldid=81697987
|to=Pronunciation of English th
|diff=81704209
|date=16 October 2006
}}
{{annual readership|expanded=true}}

==Geographical Terminology==
This article is fairly backward in its geographical terminology.  The terms are generally outdated and/or too vague.  "Britain and America" (which has been changed to the United Kingdom and the United States) and "the South", "the North" in relation to Scotland for example.  <span style="font-size: smaller;" class="autosigned">—Preceding [[Wikipedia:Signatures|unsigned]] comment added by [[Special:Contributions/62.31.15.112|62.31.15.112]] ([[User talk:62.31.15.112|talk]]) 18:44, 28 November 2009 (UTC)</span><!-- Template:UnsignedIP --> <!--Autosigned by SineBot-->

==Old Talk==
Does anyone know if the two pronunciations of ''with'' are distributed geographically in America, as they are in Britain?  --[[User:Doric Loon|Doric Loon]] 20:49, 26 May 2005 (UTC)

:I don't have any published information on it, but I once did an informal survey of my acquaintances, and my impression was that people from the Northeast and the Inland North tended to say {{IPA|/w&#618;ð/}} while people from the South and West tended to say {{IPA|/w&#618;&#952;/}}, but those were only tendencies, and it was anything but a scientific investigation! --[[User:Angr|Angr]]/[[User_talk:Angr|<big>&#53449;</big>]] 21:21, 26 May 2005 (UTC)
::I suspect {{IPA|/w&#618;ð/}} is more common everywhere.  AFAIK this is the General American :pronunciation, and hence the pronunciation you will see in most Midwest and West speakers, e.g. [[User:Benwing|Benwing]] 8 July 2005 19:32 (UTC)
:::That is interesting, as from what I have always heard, the GAE ''with'' is /wɪθ/, ignoring the GAE being a rather vague notion to begin with.  In addition, at least here in (southern) Wisconsin, [wɪθ] is by far the most common realization of ''with'', with [wɪːð] being ''far'' less common than it, and with [wɪt̪] being a sporadic realization of it in informal speech in unstressed positions before stops.  [[User:Travis B.|Travis B.]] 11:03, 27 February 2005 (UTC)


==Problem paragraphs==

I have removed these two paragraphs pending discussion
:Speakers of "Standard" English split into two groups in their articulatory implementation of /θ/ and /ð/: roughly half produce one or both [[interdental]]ly, whereas the rest produce them with the tongue tip behind the teeth. (/s/ and /z/ show the same variation, but their interdental variants are stigmatized as a lisp and therefore are significantly less common in the adult population than their non-strident counterparts.){{Talkfact}}

:Some [[Scots_language|Scots]] dialects, particularly [[Ulster_Scots_language|Ulster Scots]], pronounce initial /θ/ as [h], and /ð/ between vowels as [l]. For example ''mother'' is pronounced /m'ɑlɹ̩/ in Ulster Scots, and ''think'' is pronounced /hɪŋk/ throughout most of Scotland and Northern Ireland. Only a few words beginning /θ/ are affected (''think'', ''thing'', ''thanks'', ''nothing'' /n'ɑhɪn/ and ''anything'' /'ɛnɪhɪŋ/) while most words with intervocalic /ð/ are affected in Ulster Scots (''father'', ''mother'', ''brother'', ''other'', ''rather'' and ''bother'' amongst others).{{Talkfact}}

Angr rightly put the {{Tl|Fact}} tags here.  These statements cannot stand.  Apart from in certain consonant clusters like /ðz/ (''clothes''), I doubt very much that a significant proportion of speakers of standard English pronounce /ð/ with the tongue behind the teeth.  I can attest that it is definately not true that ''think'' has [h] or ''mother'' has [l] "throughout most of Scotland".  Sources, please, or these claims cannot remain on the page.  --[[User:Doric Loon|Doric Loon]] 16:01, 12 January 2006 (UTC)


== th rune ==

if i am not mistaken, runic didn't distinguish between, or did't include both, voiced and unvoiced dental fricatives. there is a single rune which looks like an angular thorn, the third letter in the [[futhark]]. this ought to be mentioned by someone who knows the subject well enough to add it precisely. thanks. [[User:Aaronbrick|Aaronbrick]] 21:43, 11 March 2006 (UTC)

:Actually it did.  The voiceless was represented by the [[Thorn (letter)]], and the voiced was represented by the [[Eth (letter)]].[[User:Cameron Nedland|Cameron Nedland]] 20:08, 5 June 2006 (UTC)
::Runic didn't have the [[eth (letter)|eth]], only the [[thorn (letter)|thorn]]. The Roman alphabet as used for Old English was expanded to include eth and thorn, but they were used interchangeably: both could have either the voiced or the voiceless pronunciation. It was only in Icelandic (Old and Modern) that thorn is used for the voiceless and eth for the voiced sound. [[User:Angr|Angr]] ([[User talk:Angr|talk]]) 20:28, 5 June 2006 (UTC)

:::Sorry. My bad. Here I thought I knew something.[[User:Cameron Nedland|Cameron Nedland]] 19:05, 9 June 2006 (UTC)

So then using thorn (þ) instead of θ might be better? Then the first sentence would be like:
In English, the digraph ⟨th⟩ represents in most cases one of two different phonemes: the voiced dental fricative /ð/ (as in this) and the voiceless dental fricative /þ/ (thing).

Theta comes from IPA chart, but speaking about English, and if you refer to OE text: <ref>http://www.heorot.dk/beowulf-rede-text.html Beowuld</ref>, my humble opinion is that thorn is more appropriate here. Besides, isn't þ the only surviving truely English letter? [[User:Ahangama|JC]] ([[User talk:Ahangama|talk]]) 14:33, 21 October 2015 (UTC)

:θ is the phonetic symbol. Between two slashes, which indicate that we are talking about phonemes, you would use /θ/. So we can never write /þ/. But if you are talking about Old English seplling, then by all means use <þ> - between pointy brackets to indicate a grapheme. --[[User:Doric Loon|Doric Loon]] ([[User talk:Doric Loon|talk]]) 20:02, 5 January 2016 (UTC)

{{reflist-talk}}

==Illustrations==
The section on phonetic realisation would benefit from a diagram showing the two tongue positions.  Most phonetics books have such a diagram, but these presumably are copyright.  Would anyone here feel competent to draw one?  --[[User:Doric Loon|Doric Loon]] 21:41, 23 May 2006 (UTC)

==GA Nomination==
I am failing the GA nomination.  While there is a lot of good material in the article I felt that the lack of in-line citations and only two references for what is a well known topic in phonology were simply not sufficient.  Also the flow of the article could be improved.  The sections don't follow from one another logically and simply seem to be strong along with out any overriding order (historical, phonological, orthographuc).  Almost but not quite where it needs to be. [[User:Eluchil404|Eluchil404]] 22:07, 3 June 2006 (UTC)

==Th...==
Am I only person who knows that the "th" in "the" is shorter, different sound to that in "other". It is infact, at least in my accent quite a bit different, almost like a "th" mixed with an "l" sort of position.

:So what IS your accent?  It is usual for almost any phoneme to show allophonic variations.  The articleshould only mention them if they can are described in scholarly literature. --[[User:Doric Loon|Doric Loon]] 19:14, 23 July 2006 (UTC)

== ''Bovver boy'' Standard English? ==

''...and in at least one case has been transferred into standard English as a neologism: a bovver boy is a thug, a "boy" who likes "bother" (fights).'' Surely not Standard English just a phonetic spelling representing the /ð/ to [v] realisation. Probably as an indication of the perceived social status of the ''bother boy''. [[User:84.135.240.77|84.135.240.77]] 15:08, 8 November 2006 (UTC)

:Yes, I would say it is now standard; at least as standard as "lager lout" and other phrases representing passing social phenomena.  At any rate, it's in several dictionaries that I have lying around here.  --[[User:Doric Loon|Doric Loon]] 16:49, 8 November 2006 (UTC)

::Which dictionary? It's not on Merriam-Websters, American Heritage, Reference.com, a few other online dictionaries and encyclopedias (including Wikipedia) just redirect to Skinhead and make no mention of the term, and even Wiktionary doesn't have an entry. For what it's worth, I had never heard of the term until I viewed this page, and believe it to be chiefly (well, exclusively, really) British. I could probably count on one hand the number of times I have heard an American use the word "bother" in that way, anyways. And even if that pronunciation is "fairly widespread," I would like to see some evidence it has any popularity in America, since I also feel that is somewhat relegated to Britain. [[User:Eebster the Great|Eebster the Great]] ([[User talk:Eebster the Great|talk]]) 05:30, 4 May 2009 (UTC)

== This article ==

If one actually takes time to read this article, it turns out to be highly repetitive. This is a common problem around here, as people contribute without reading everything that has been written before. Anyway, it ought to be tightened up. I'm not doing it now because it would take more time and effort than I feel like spending these days. --[[User:194.145.161.227|194.145.161.227]] 22:53, 13 November 2006 (UTC)

:Yeh, there's  a reason for that.  A few weeks ago someone merged a whole lot of material from another article, and did it so clumsily that although I spent over an hour tidying it up I still haven't ironed out the hicups.  In particular, the historical section needs completely rearranged.  Feel free: I don't have time, and to be honest I can't be bothered.  I spent so much time writing this article that I find it disspiriting to have to redo it all after someone messed everything around.  So, you have a go.  --[[User:Doric Loon|Doric Loon]] 23:27, 13 November 2006 (UTC)

::OK, thanks - I guess I'm going to do something about it at one point or another. --[[User:194.145.161.227|194.145.161.227]] 14:38, 14 November 2006 (UTC)

:I'm the "someone", thanks. It looked "clumsy" because I don't actually know anything about the subject myself, and so I figured it was better to err on the side of inclusion; if you felt the addition was a net negative you could have asked me to do more. But ironically it's all probably going to have to be reverted anyway, I now find out. That material was originally added to the article [[T]] by the user [[User:Macaw 54]], who it now turns out was a sock puppet of a serial plagarist and so it's all likely in violation of copyright (see [[Talk:T#Macaw 54]]). So this is a dispiriting time all around, it seems. [[User:Bryan Derksen|Bryan]] 00:32, 15 November 2006 (UTC)

Sorry, Bryan, that wasn't meant to be personal, more an expression of my own November weariness.  No, the addition wasn't a net negative, because some of the new material was interesting.  But I have had to delete a lot of duplication, and there will be more, because as I say, I haven't tried to rationalise the history section yet.  Plagiarism is only a problem if whole chunks of text are stolen, so if we are going to cut it back and keep the best bits, it is OK if some of those bits came from books.  But perhaps we can prevail on Macaw 54 to reveal his sources?  --[[User:Doric Loon|Doric Loon]] 09:57, 15 November 2006 (UTC)

:I guess it can't hurt to try. [[Wikipedia:Long term abuse/Primetime]] indicates that his source of choice is Encyclopedia Britannica, so even if he doesn't respond that'd be a good place to start looking. Since plagarism in and of itself isn't illegal, just the copyright violation part, perhaps if we can cite and rewrite the "good bits" sufficiently that they contain the same information but expressed in a different way we'll be okay. I suspect the sheer volume of material he added to [[T]] may be too big a task for that, but the stuff I split out to this article might be salvageable. Feeling a bit better now. :) [[User:Bryan Derksen|Bryan]] 17:24, 15 November 2006 (UTC)

==Removing material==

OK, I have started to redo the history section.  There was a lot of duplication, and a lot of stuff which was unclear.  I have written a new explanation of the Germanic prehistory, which is short and allows a lot of the longwinded stuff to be deleted.  I have also deleted most of the material on the history of the letters thorn and eth, since that does not belong here.  We have links to the [[thorn]] and [[eth]] articles, but just in case I am deleting anything important which is missing from those other articles, I will deposit the deleted passage here.  Then some enterprising person can merge important points into those typographical articles if it seems worth doing.  --[[User:Doric Loon|Doric Loon]] 15:34, 15 November 2006 (UTC)

These paragraphs were deleted:
:In the earliest known Old English writings in the Roman alphabet thorn was represented by ''th,'' the voiced spirant being often represented by ''d'' /д/ (sometimes by ''th''). Before 700 probably, the character ''<span lang="und-Latn" class="latinx">ð</span>,'' formed by a bar across the stem of ''д,'' was introduced; it appears in a charter of Wihtræd, king of [[Kent]], 700-715.<ref>Henry Sweet, ''The oldest English texts'' [Early English Text Society: 1885], p. 428</ref> Apparently it was first used to denote the voiced spirant: see the proper names in the Moore Manuscript of Bæda, ca. 737, and the ''Liber Vitæ,'' Cotton Manuscript, ca. 800, and charters before 800 generally. But in the ninth century it was used for both spirants, as in the Vespasian Psalter, ca. 825 (e.g., iv. 5: "''<span lang="und-Latn" class="latinx">ð</span>''a ''<span lang="und-Latn" class="latinx">ð</span>''e cweo''<span lang="und-Latn" class="latinx">ð</span>a<span lang="und-Latn" class="latinx">ð</span>''"), and in a West Saxon charter of 847.<ref>Sweet, op. cit. 433</ref> In the 8th century apparently, the thorn,'' <span lang="und-Latn" class="latinx">Þ</span>,'' was adopted from the Runic ''futhorc,'' the earliest charter showing it being one of Coenwulf, king of [[Mercia]], of 811 (Sweet, op. cit. 456); but it was not much used until late in the 9th century. A Surrey charter ''a'' 889<ref>ibid. 451</ref> has 34 examples of ''<span lang="und-Latn" class="latinx">ð</span>'' initial, and 25 medial or final, with 49 of ''<span lang="und-Latn" class="latinx">Þ</span>'' initial, and 1 medial. From the later years of the 9th century ''<span lang="und-Latn" class="latinx">ð</span>'' and ''<span lang="und-Latn" class="latinx">Þ</span>'' were used promiscuously in West Saxon works, with some preponderance of ''<span lang="und-Latn" class="latinx">Þ</span>'' initially and ''<span lang="und-Latn" class="latinx">ð</span>'' finally. This continued in Middle English until the 13th century. On the other hand, the Durham ''Rituale'' and the Lindisfarne Gospel Gloss, ca. 950, have uniformly ''<span lang="und-Latn" class="latinx">ð</span>'' in all positions (except in the compendium ''<span lang="und-Latn" class="latinx">Þ</span>'' for ''<span lang="und-Latn" class="latinx">ðæ</span>t''), as has also the East Anglian ''Genesis & Exodus,'' ca. 1250; while the Mercian portion of the Rushworth Gospel Gloss, ca. 975, and [[Ormin]], ca. 1200, have only ''<span lang="und-Latn" class="latinx">Þ</span>''. After 1250 the ''<span lang="und-Latn" class="latinx">ð</span>'' speedily became obsolete; ''<span lang="und-Latn" class="latinx">Þ</span>'' remained in use, but was gradually restricted more or less to the pronominal and demonstrative words. In later times its manuscript form approached, and at times became identical with, that of ''y'' (the latter being sometimes distinguished by having a dot placed over it). As the continental type used by William Caxton had no ''<span lang="und-Latn" class="latinx">Þ</span>,'' its place in print was usually supplied by ''th'' for both sounds and in all positions. But in Scotland, the early printers, especially in the demonstrative and pronominal words, continued the ''<span lang="und-Latn" class="latinx">Þ</span>'' as ''y,'' as in ''ye, yis, yat, you''--"thou," a practice also common in England in manuscripts, and hardly yet extinct. Confusion with the modern ''y'' consonant, Middle English ''<span lang="und-Latn" class="latinx">ȝ</span>,'' was avoided in Scotland, sometimes by writing the latter ''yh,'' but usually by continuing Middle English ''<span lang="und-Latn" class="latinx">ȝ</span>'' in the form [[Image:Insular_G.GIF]] or z, so that ''ye zeir'' stood for ''<span lang="und-Latn" class="latinx">Þ</span>e <span lang="und-Latn" class="latinx">ȝ</span>eir,'' i.e. ''the year''. It is remarkable that, when Old English ''<span lang="und-Latn" class="latinx">Þ</span>'' and ''<span lang="und-Latn" class="latinx">ð</span>'' were both in use, no attempt was made to differentiate them as breath and voice spirants, and apparently no serious attempt even to distinguish them as initial and medio-final, as was done in Norwegian when the Roman alphabet was adopted , ca. 1200, and in [[Icelandic language|Icelandic]] before 1300. At an earlier date (probably ca. 800) the character ''<span lang="und-Latn" class="latinx">ð</span>'' was partially adopted from Old English in Old Saxon, and was used generally in the middle and end of words, while ''th'' was usual as the breath spirant initially.

:In the demonstrative and pronominal groups of words, change of initial ''<span lang="und-Latn" class="latinx">Þ</span>'' to ''t,'' by assimilation to a preceding dental ''(t, d, s),'' appears in earlier English. Old English ''<span lang="und-Latn" class="latinx">Þæt</span> <span lang="und-Latn" class="latinx">Þ</span>e'' became ''<span lang="und-Latn" class="latinx">Þæt-te</span>, <span lang="und-Latn" class="latinx">Þ</span>ætte;'' ''<span lang="und-Latn" class="latinx">Þ</span>e læs <span lang="und-Latn" class="latinx">Þ</span>e'' appears in the 11th century as ''<span lang="und-Latn" class="latinx">Þ</span>e læste,'' thus modern ''lest''. In the last section of the Old English Chronicle, from 1132,'' <span lang="und-Latn" class="latinx">Þ</span>e'' after ''t'' or ''d'' regularly becomes ''te'' (e.g., ''<span lang="und-Latn" class="latinx">Þ</span>at'' ''te'' king, and ''te'' eorles). In the ''Ormulum'' and the [[Cotton Manuscript]] of "Cursor Mundi," this assimilation is seen in all the words of the ''the-thou'' group (''Ormulum:'' "''<span lang="und-Latn" class="latinx">Þ</span>att ''tatt te'' goddspell mene<span lang="und-Latn" class="latinx">ÞÞ</span>," "wrohht ''tiss'' boc," and "''tatt te'' follc all <span lang="und-Latn" class="latinx">Þ</span>ess ''te'' bett""; ''Cursor Mundi:'' "ne was ''tar,'' here and ''tare,'' scho serued ''taim,'' als sais ''te'' sau"). So in ''[[Ancren Riwle]]:''and ''tet'' is, et ''tesse'' uerse,'' <span lang="und-Latn" class="latinx">Þ</span>eo <span lang="und-Latn" class="latinx">Þ</span>et ''tus'' do<span lang="und-Latn" class="latinx">ð</span>, and ''tes'' ''o<span lang="und-Latn" class="latinx">ð</span>er,'' etc.). In the course of the 14th century, this assimilation was given up, and the spirant reappeared (as ''<span lang="und-Latn" class="latinx">ð</span>'').

{{reflist-talk}}

==Merge Th (digraph) into this article==
The article [[Th (digraph)]] has very little text in it, and some of what is at [[Pronunciation of English th]] is actually about other languages, and the overall history of the digraph. I think the two should be merged. [[User:FilipeS|FilipeS]] 21:10, 2 December 2006 (UTC)

:Oh no, not again!  Merges on this article seem to end up with me doing all the work to fix the seams.  But OK, that's not an argument.  I think a merge would be unhelpful.  This article is about English, and including unrelated languages would not benefit it.  Trying to reshape this article so it is about all languages equally would destroy it.  Having said that, much of the info in [[Th (digraph)]] is already here: the pronunciation in German, Welsh, and in Gaelic (which is almost Irish) are all mentioned.  The bit about Thai is nonsense: Thai doesn't use our alphabet, ergo it doesn't have the th-digraph.  The digraph is used when transcribing Thai, and has the same value it has when transcribing Sanscrit, which this article already mentions.  So there is no new insight to carry across, and just listing more and more languages brings no extra value. To be honest, I would just vote for deletion of [[Th (digraph)]] - it is so superficial, and trying to cover lots of unrelated languages under the umbrella of their arbitrary use of a spelling convension seems to me not the way linguists work.    --[[User:Doric Loon|Doric Loon]] 23:10, 2 December 2006 (UTC)

If the merger proposal is rejected, I'm going to move half of this article to [[Th (digraph)]]. The history of the digraph, including its prior use in Greek and Latin, and its use in Gaelic, do not belong in an article titled "Pronunciation of '''English''' th". :P ;-) [[User:FilipeS|FilipeS]] 00:29, 3 December 2006 (UTC)

:Well, yes they do belong here, because they explain the English situation (which for example reference to Albanian wouldn't).   There is no harm in having information in two places, especially if one of them is cursory and links to the other.  The account of the history of the digraph here is very brief, almost stub-like, because that is all that is required for present purposes.  But a full account of the history of the digraph wouldn't belong here, and if you think it would be useful to have one then by all means write it up poperly at [[Th (digraph)]].  But don't MOVE it, which is the lazy option: write what that article needs.  --[[User:Doric Loon|Doric Loon]] 09:54, 3 December 2006 (UTC)
:Perhaps I should explain more precisely why I selected the details I did for the history of the digraph section.  This article is really about the history and complexity of the English sounds.  85% of it is on English phonetics and phonology.  The last 15% is really just short notes on why English uses these letters for these sounds.  There are three types of sounds associated with these letters, so there are three short sections charting the history of these three phoneme-grapheme relationships from origins down to English (including other languages only when they influence some words in English).  The last section is just a pointer to the fact that the digraph can do completely different things in other languages, hence a the reference to Gaelic.  This is kept very short because its purpose is only to put English in perspective as not being the only system.  A proper explanation of Gaelic lenition is not attempted in this three-line section, however, as that belongs elsewhere.  I hope that helps to show how this very carefully thought-out article, into which a vast amount of work has been invested, has a clear integrity as an account of a phenomenon in English.  So please don't mess with it on a whim.  Go instead to [[Th (digraph)]] and create something equally good there.  --[[User:Doric Loon|Doric Loon]] 10:07, 3 December 2006 (UTC)

It is indeed a well-written article, and the amount of historical background is tolerable. I will withdraw the merger proposal. [[User:FilipeS|FilipeS]] 18:23, 3 December 2006 (UTC)

:Thanks.  Let's see if we can't do something worthwhile with [[Th (digraph)]] instead. --[[User:Doric Loon|Doric Loon]] 07:33, 4 December 2006 (UTC)

==  /ð/  tongue placement ==

The introduction states that for the phoneme'''s''' the tongue can be either between the teeth or behind the '''upper''' teeth.  I agree for theta, but as a native speaker, I find it almost impossible to make a  /ð/ with my tongue behind the upper teeth.  Instead, if I don't stick out my tongue, I place it behind my '''lower''' teeth.  [[User:Kdammers|Kdammers]] 04:18, 10 November 2007 (UTC)

:Sure - not every native speaker would do this.  I do it in certain collocations, for example in "clothes", where it is easier to get to the s if the th is not properly interdental. I think the statement is right: the allophones exist. --[[User:Doric Loon|Doric Loon]] 16:04, 10 November 2007 (UTC)

== Another pronunciation ==

This is only one word I can think of in which 'th' is pronounced like this, but (at least the way I say it) 'posthumous' has 'th' pronounced almost like 'ch'. Like I said, it's only one word, so I'm not sure it worthy of a mention on this article, but I thought I would bring it up. [[User:DrJorin|DrJorin]] ([[User talk:DrJorin|talk]]) 13:52, 1 August 2009 (UTC)

== Th as S?==
I have heard people saying "thank you" as "sank you". What causes this?  <small><span class="autosigned">—Preceding [[Wikipedia:Signatures|unsigned]] comment added by [[User:Minihag a|Minihag a]] ([[User talk:Minihag a|talk]] • [[Special:Contributions/Minihag a|contribs]]) 09:22, 28 October 2009 (UTC)</span></small><!-- Template:Unsigned --> <!--Autosigned by SineBot-->
That would be a [[lisp]], and in that case it would be a [[dental consonant|dental]] ''s''. (from below)

== Smooth? ==
In the article it states, in layman's terms no less, that /smuːð/ and /smuːθ/ are acceptable pronunciations of the word 'smooth'.
Now I don't know what country uses the latter pronunciation, but it sure isn't Australia- and most probably other countries in the Anglosphere.
This probably needs changing.
[[Australian English|<span style="color:black;font-variant:small-caps">'''En-AU'''</span>]] [[User:Enauspeaker|<span style="color:grey;font-variant:small-caps">Speaker</span>]] [[User talk:Enauspeaker|<sub><span style="color:blue;font-family:serif">(T)</span></span></sub>]] [[Special:Contributions/Enauspeaker|<small><span style="color:green;font-family:serif">(C)</span></small>]] [[Special:Emailuser/Enauspeaker|<sup><span style="color:red;font-family:serif">(E)</span></sup>]] 04:36, 8 March 2010 (UTC)
::Agreed. [[User:Jakob37|Jakob37]] ([[User talk:Jakob37|talk]]) 03:42, 30 July 2010 (UTC)

== medial position - comments ==

1. Arthur (Welsh has /θ/ medially: /ærθɨr/) - isn't it /arθr/?

2. brothel, Ethel - if we are including proper names, how about Bothell, a suburb of Seattle? Rimes with brothel. Also, are you sure Ethel doesn't come from Hebrew, Aramaic, etc. What is it's origin?

3. Speaking in general, some discussion in this section about the relationship between pronunciation and morpheme boundaries would not be amiss. [[User:Jakob37|Jakob37]] ([[User talk:Jakob37|talk]]) 03:51, 30 July 2010 (UTC)

:In Welsh, ''Arthur'' is {{IPA|[ˈɑrθɨr]}}. [[Ethel]] is of Anglo-Saxon origin; it comes from the Old English word for "noble". (The þ/ð was voiced in Old English, so I don't know why it's become voiceless in the Modern English name.) +[[User:Angr|'''An''']][[User talk:Angr|''gr'']] 08:38, 31 July 2010 (UTC)

== Pronunciation queries ==

In [[British English]] (my accent is somewhere between [[Received Pronunciation|RP]] and [[Estuary English|Estuarine]]), I would pronounce the medial ‹th› in both ''earthen'' and ''earthenware'' with /θ/. Similarly, I would pronounce the final ‹th› in ''loath'' and ''loathe'' as homonymous with /ð/ and I've never heard ''froth'' (the verb) with a /ð/.

Finally, the medial ‹th› in ''bathing'' could be either /θ/ or /ð/ for a reflexive verb — ''"I was bathing the baby"'' could be /bɑːθɪŋ/ or /beiðɪŋ/; the self-reflexive ''"I was bathing in the pool"'' would only be /beiðɪŋ/ — [[:en:User:OwenBlacker|OwenBlacker]] ([[:en:User talk:OwenBlacker|Talk]]) 23:55, 5 September 2010 (UTC)


I second the comment about ''earthen'' being with /θ/.  In fact, [http://en.wiktionary.org/wiki/earthen wiktionary] lists that as the only pronunciation. 
- [[User:Rflrob|Rflrob]] ([[User talk:Rflrob|talk]]) 07:49, 25 February 2011 (UTC)

:Then be bold! But best get a reference for anything you change. I personally do distinguish loath and loathe, but otherwise I agree with both you and Owen Blacker. --[[User:Doric Loon|Doric Loon]] ([[User talk:Doric Loon|talk]]) 08:24, 25 February 2011 (UTC)

==New York==
nu'in on new yawk? I'm from queens and my Th is a light D or T (but never CH even in cluster Thr) and I know for sure im not alone  <span style="font-size: smaller;" class="autosigned">—Preceding [[Wikipedia:Signatures|unsigned]] comment added by [[Special:Contributions/71.190.208.141|71.190.208.141]] ([[User talk:71.190.208.141|talk]]) 22:18, 19 December 2010 (UTC)</span><!-- Template:UnsignedIP --> <!--Autosigned by SineBot-->

==By far the most common speech impedimant (?)==

The section "Acquisition problems" says that the lisp "is by far the most common speech impediment in English". This is unsourced, and I think it's almost certainly untrue. Inability to pronounce "r" is a common problem; it wouldn't surprise me if it's more common than the lisp. But even if it's not, saying the lisp is "by far" the most common is wrong. So I'm going to change it to "is a common speech impediment in English". [[User:Duoduoduo|Duoduoduo]] ([[User talk:Duoduoduo|talk]]) 15:39, 22 July 2011 (UTC)
:Quite right. --[[User:Doric Loon|Doric Loon]] ([[User talk:Doric Loon|talk]]) 15:52, 22 July 2011 (UTC)

==Number of examples==

@Doric Loon: You wrote "it only makes sense to have one each for initial, medial and final position. Otherwise why not have even more?"  Maybe I'm wrong, but I think these are the ''only'' examples. (Or are there some I'm not aware of?) If I'm right, I don't think it makes sense to list three out the four examples but not the fourth.  [[User:Duoduoduo|Duoduoduo]] ([[User talk:Duoduoduo|talk]]) 16:08, 3 August 2011 (UTC)

:No, there are more further down the page: teeth:teethe, mouth (noun) : mouth (verb) etc. For phonemic comparisons, just one minimal pair in each of the three positions is what you would normally give - hence further down also the then-zen-den sets are structured the same way. (Actually, one pair is enough to prove a phonemic relationship, but since some phonemes are only distinct in certain positions, like d and t in German, which merge in final position, it is more satisfying to have three.) But you are right, there are not many minimal pairs, and if you think it appropriate you can certainly make your point somewhere in the detailed list further down the article. Though to be honest, I doubt if thou (=thousand) is standard enough to be worth bringing in - feels rather contrived, the kind of thing I would only cite for the sake of being cute. But if you really use it, well ok. --[[User:Doric Loon|Doric Loon]] ([[User talk:Doric Loon|talk]]) 21:04, 3 August 2011 (UTC)

==Refs==
If anyone is looking for some refs to back up a lot of this article you may want to check out the OED entry for th. In fact a good bit of this article seems to have been taken from the OED entry, including examples. [[User:pschemp|pschemp]] | [[User talk:pschemp|talk]] 22:14, 25 November 2012 (UTC)

== Ðanks ==

I (and most of my family members) pronounce "thank" and its derivatives with /ð/ – an exception to the rule that only function words have initial /ð/. Does anyone else do this? <sub>([[Special:Contributions/Phinumu|suoı̣ʇnqı̣ɹʇuoɔ]] · [[User_talk:Phinumu|ʞlɐʇ]])</sub> [[User:Phinumu|nɯnuı̣ɥԀ]] 20:28, 20 May 2014 (UTC)
:I've never heard that, and would be interested to know where you come from, or whether you belong to any particular sociolinguistic grouping. If you are a native speaker and it is part of your language, then that is highly interesting, and certainly not wrong. For the moment, I wouldn't add it to the article though.--[[User:Doric Loon|Doric Loon]] ([[User talk:Doric Loon|talk]]) 16:42, 26 February 2020 (UTC)

I also pronounce it like this and so do most other people who live in my area. [[User:Ledlecreeper27|Ledlecreeper27]] ([[User talk:Ledlecreeper27|talk]]) 23:16, 25 May 2021 (UTC)

== Alveolar sibilant ==

This article uses "[[alveolar sibilant]]" interchangeably with "[[alveolar fricative]]". The former term is maybe a bit more intuitive, but rather uncommon (as evidenced by the fact that it's a redlink). We should be consistent. Should we either use the former (and create a redirection to [[alveolar fricative]]) or the latter? &mdash; [[User:SebastianHelm|Sebastian]] 10:59, 16 January 2018 (UTC)
: Since nobody replied, I changed the one remaining instance of “alveolar sibilant” to “alveolar fricative”. ◄ [[User:SebastianHelm|Sebastian]] 13:48, 26 February 2020 (UTC)
::Well, "[[alveolar sibilant]]" may be a redlink, but "[[alveolar fricative]]" is a redirect, which is only marginally better. A [[sibilant]] is a kind of [[fricative]]. They are not interchangeable, but all sibilants are fricatives. Your change has made this slightly less precise. I don't think it makes much difference, though. --[[User:Doric Loon|Doric Loon]] ([[User talk:Doric Loon|talk]]) 16:30, 26 February 2020 (UTC)
:::I think the change, though not essential, is perfectly fine. English has no non-sibilant alveolar fricatives and "fricative" is indeed a much more widely recognized term. [[User:Nardog|Nardog]] ([[User talk:Nardog|talk]]) 18:12, 26 February 2020 (UTC)

== Homophonous pairs ==
The list at {{pslink|th-alveolarization}} was recently expanded to contain some rather unusual words, such as “Louth”. Now the question arises: Where do we draw the line? Is the list intended to be complete or as a list of examples? ◄&nbsp;[[User:SebastianHelm|Sebastian]] 13:48, 26 February 2020 (UTC)
:No, the list is not intended to be complete. We should have as many examples as are necessary to illustrate the point, not more. One of my constant complaints with linguistics articles is that people feel the need to keep adding other examples or details that occur to them without thinking about how it helps the whole text. The result is a clutter that becomes less and less clear to the reader. So I would encourage you to be bold and revert additions that you think aren't really helping.--[[User:Doric Loon|Doric Loon]] ([[User talk:Doric Loon|talk]]) 16:36, 26 February 2020 (UTC)
::This is a problem with almost every one of these types of lists used in language articles. The de facto rule is to allow any number of examples, as long as they're valid. These lists can never realistically be complete, so there will always be more possible examples to add. Any limit will be arbitrary and will differ from article to article (unless it gets incorporated into a manual of style for linguistics or something). I'm slightly for introducing a limit, but I think the hassle of reverting additions that go over the limit is more than any benefits gained from the limit in the first place. That's not to mention figuring out which examples to keep and which to throw away.--[[User:Megaman en m|Megaman en m]] ([[User talk:Megaman en m|talk]]) 16:45, 26 February 2020 (UTC)
:::Those lists of homophonous pairs have been a source of OR and [[WP:INDISCRIMINATE]] and only IPs seem to ever edit them. I'd totally be down with axing them, though I'm sure seeking a wider consensus at [[WT:LING]] or [[WT:WPENGLISH]] or the like is preferable. [[User:Nardog|Nardog]] ([[User talk:Nardog|talk]]) 18:18, 26 February 2020 (UTC)
::::I would agree with axing it. Put two very obvious examples into the text, if they are sourced, and leave it like that. Though I am not convinced we even need that.--[[User:Doric Loon|Doric Loon]] ([[User talk:Doric Loon|talk]]) 19:56, 10 March 2020 (UTC)

== Three strengths of "th" not two?? ==

It seems that the IPA only recognizes two strengths of "th" but I can hear three. Let's consider the words "the", "father" and "thought". I think that everyone would agree that out of these three words, "the" has the weakest pronunciation of "th" and "thought" has the strongest. However, does anyone agree with me that the word "father" has a middle-strength pronunciation of "th"? If yes then does anyone know why the IPA does not recognize three strengths of "th"? [[Special:Contributions/95.172.233.137|95.172.233.137]] ([[User talk:95.172.233.137|talk]]) 15:12, 18 November 2020 (UTC)
:I am not convinced that there are three strengths in my dialect (UK near RP). I think that "this" and "father" are essentially the same, bar the usual additional emphasis on stressed syllables.  -- [[User:Q Chris|Q Chris]] ([[User talk:Q Chris|talk]]) 11:05, 19 November 2020 (UTC)
::Thank you, Q Chris, very much for your reply. For me, the "th" in "this" is definitely weaker than the "th" in "father". Imaging saying "far the" (not pronouncing the "r" like Americans and Irish people do) when you intend to say "father". To me, that would sound wrong. What do you think? [[Special:Contributions/95.172.233.137|95.172.233.137]] ([[User talk:95.172.233.137|talk]]) 12:43, 19 November 2020 (UTC)
:Remember that the talk page is [[WP:NOTFORUM|not a forum]] to discuss the topic at hand, rather it's used to talk about how to improve the article.--[[User:Megaman en m|Megaman en m]] ([[User talk:Megaman en m|talk]]) 13:33, 19 November 2020 (UTC)

:::There are two "th"-phonemes in English. Each of them has a variety of allophones. So the voiced sounds in "the" and "father" are the same phoneme, but obviously there will be a slightly different realization depending on whether the sound is in initial or medial position. But try experimenting with minimal pairs. You can hear two different sounds in otherwise identical words if you try saying "thy" and "thigh". You can't get a third one in the same position. --[[User:Doric Loon|Doric Loon]] ([[User talk:Doric Loon|talk]]) 22:28, 19 November 2020 (UTC)

::::The 'th' in 'the' and the 'th' in 'father' are exactly the same, as are the all the listed examples. There are only two: eth and thorn [[Special:Contributions/2601:182:4381:E60:244F:964C:E05F:5BA2|2601:182:4381:E60:244F:964C:E05F:5BA2]] ([[User talk:2601:182:4381:E60:244F:964C:E05F:5BA2|talk]]) 17:57, 24 January 2022 (UTC)