File talk:Extended IPA chart 2005.png

It's ... wrong!

Why is the chart so wrong? Gah. Look, by adding non-standard symbols you're clouding the issue quite a lot, for one thing, and confusing people who would otherwise learn the IPA. It's also a bit dubious to then label it the 2005 IPA at the top. The order of the consonants is probably important on the real chart, as they vaguely follow some sort of sonorant hierarchy.

Worst of all is the clicks. The IPA symbols do not specify whether a click is "fricated" or "abrupt" - this may come about in some people's speech, but it's not accurate for the symbols given. The descriptions given on the IPA are deliberately underspecified, probably mostly because when you string more than two long words together, it can get confusing! Describing these as simply 'Bilabial' and 'Dental' is more accurate too, simply because the descriptions given there are overspecified. For instance, it's perfectly possible to produce a dental click as apical, and perfectly possible to produce it without frication. These should not be specified by the IPA, they are, as with every other sound on there, deliberately open to interpretation so that different authors can make them mean different things. Plus, if you really want to learn about how clicks work, it's probably best to learn about them specifically. The chart does not need to describe this.

Just... at the end of the day, this is not the IPA, this is some randomer coming along and deciding that their theory of phonology must be better in some way. We don't need to see that on the IPA chart. We all know it's not perfect, and in some places it is indeed wrong... but it's the standard. Stop trying to alter it for superficial reasons. Finlay (talk) 12:59, 18 December 2009 (UTC)

Voiceless labialized velar... approximant?

I don't know if this is something they changed the definition of for the 2005 revision, but isn't ʍ a fricative and not an approximant? • 辻斬り^? 22:48, 25 December 2005 (UTC)

I think the difficulty probably comes from the fact that when an approximant becomes voiceless, it takes on the characteristics of a fricative. Due to this, I think some might even call voiceless approximants fricatives. As I understand it, [ʍ] is usually [w] without voicing. --j0no 20:37, 30 July 2006 (UTC)

The IPA still uses the term 'fricative', but for historic reasons. Phoneticians recognize that this is an approximant. (A labialized velar fricative is written xʷ.) One of the reasons for redoing the chart is to stop perpetuating such confusion. Likewise, lateral flaps get their own row, and epiglottals their own column here - people often think those aren't normal pulmonic consonants because in IPA publications they're lumped under 'other'. There is, however, one misconception that is perpetuated in this chart: h, ɦ are not glottal, nor are they fricatives or approximants, but rather bare phonation. They really should be placed under 'other'. kwami 07:22, 20 May 2007 (UTC)

I don't understand what "for historic reasons" means here. Could you explain, please? And "phoneticians recognize that this (ʍ) is an approximant"? - well, here is one phonetician who doesn't, and has regularly referred to it in publications as a fricative. I don't see any confusion. Also "h, ɦ are not glottal"? The former certainly is, because there is a noise-generating stricture between the vocal folds for this articulation; the latter involves phonation and usually some fricative noise, but could from the phonetic point of view be classed as a breathy-voiced vowel. There is, however, no such thing as "bare phonation", since the airstream that has passed through the larynx has to pass through the vocal tract and will thus as a minimum carry vowel resonances. RoachPeter (talk) 15:54, 13 July 2012 (UTC)

Take a look at history of the IPA. In earlier versions, there was no such thing as an approximant: [w ɹ j] were all "fricatives". The terminology dates from that time. If [ʍ] is actually a fricative, then it is equivalent to [xʷ] – and indeed some people use it that way. But most use it as an approximant, and write out ⟨xʷ⟩ explicitly for the fricative.

The comment on [h, ɦ] was overly simplistic. People have argued that in languages such as English, these are bare phonation, not glottal consonants, and the vowel resonances are just the allophonic effects of neighboring vowels. However, in Arabic [h] is said to be a true fricative. The IPA doesn't distinguish these possibilities.

SOWL says, the shape of the vocal tract during h or ɦ is often simply that of the surrounding sounds. [English example] In such cases it is more appropriate to regard h and ɦ as segments that have only a laryngeal specification, and are unmarked for all other features. There are other languages which show a more definite displacement of the formant frequencies for h, suggesting it has a constriction associated with its production. [glottal constriction in Arabic and Hebrew]

The voiceless counterparts of the central approximants j and w occur ... in some dialects of English. [In most Scottish dialects, whether begins] with a non-fricative ʍ. ... [a labio-velar fricative is impossible, so] if it is a fricative, it is better described as a voiceless labialized velar fricative. (That is, as [xʷ].) — kwami (talk) 20:12, 14 July 2012 (UTC)

This is a simple matter of logic and the definition of scientific categories, and it doesn't matter in this context what symbols people might choose to represent, say, a Scots 'wh'. Here we go:

As the WP article on the subject says, an approximant involves an articulation that does not become close enough to generate noise. Therefore an approximant does not produce noise.

If an approximant is voiceless, then there is no vocal fold vibration going on to produce audible sound.

Therefore the [ʍ] as you classify it in your modified IPA chart is noiseless and voiceless, and therefore completely silent.

Of course, people can and will use ⟨xʷ⟩ to represent the "voiceless w" in English, and good luck to them. What I am concerned about is that the WP change to the IPA Chart has rendered the symbol [ʍ] vacuous, in that it now represents only silence.

I will leave the issues concerning [h, ɦ] for now, because these have not been changed in the Chart, but nothing you have written justifies the claim that these sounds are not glottal. RoachPeter (talk) 09:18, 15 July 2012 (UTC)

As the WP article on approximants goes on to say:

Because voicelessness has comparatively reduced resistance to air flow from the lungs, the increased pulmonic pressure creates more turbulence, making acoustic distinctions between voiceless approximants (which are extremely rare cross-linguistically) and voiceless fricatives difficult.

This explains why we can make reference to voiceless approximants. Something similar may occur with voiceless vowels, which are a thing in some languages. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 14:11, 15 July 2012 (UTC)

I cannot conceive of any way in which the reduced resistance to air flow from the lungs in a voiceless segment could result in increased pulmonic pressure. It would, however, make sense to claim that the reduced resistance to air flow from the lungs results in a higher rate of air flow through the vocal tract. The passage quoted could then go on to say that as a result of this higher rate of flow, sounds which when voiced are approximants become voiceless fricatives. RoachPeter (talk) 15:46, 15 July 2012 (UTC)

So the defining difference between approximants and fricatives is turbulence? I'm a little skeptical of this, though the source cited for the above claim (Ohala 1995) describes the situation in a way similar to how you have:

"From these equations we see that turbulence can be increased by decreasing the crossdimensional area of the channel. This is the usual view of how fricatives differ from approximants. But I don't think this is what is involved in the cases cited. Rather, another way to create turbulence is by increasing U, the volume velocity and this, in turn, can be effected by increasing P_Oral. In the case of [ɬ], the P_Oral is increased by virtue of its voicelessness: this reduces the resistance at the glottis to the expiratory air flow. The upstream pressure is then essentially the higher pulmonic pressure. Thus the fricative character of the [ɬ] need not result from its having a narrower channel than the approximant [l] but simply from being [-voice]."

It seems that Ohala's position is that voiceless approximants simply don't exist. Presumably, then, crosslinguistic descriptions of voiceless approximants have more to do with differing in terminology used by linguists than actual phonetic differences from languages describe as having voiceless fricatives. I'm not sure how representative this is of the position of most linguists, though Ohala's wording suggests that his view wasn't common 17 years ago. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 17:27, 15 July 2012 (UTC)

RoachPeter, rather than us explaining it to you, it would probably be easier if you were to read the SOWL sections on voiceless approximants. There is a description under laterals of how voiceless approximants differ from fricatives (Tibetan has [l̥], for example, while Navaho has [ɬ], though no known language contrasts them). [w̥] is not the same as [xʷ], and ⟨ʍ⟩ is defined as the former. Basically, it's similar to how voiced approximants and fricatives differ, in the degree of turbulence in the airflow. — kwami (talk) 04:02, 16 July 2012 (UTC)

Italics

Could we remove italics and replace them by straight up ones? ---85.27.32.51 13:45, 8 May 2006 (UTC)

Which italics? kwami 07:22, 20 May 2007 (UTC)

Error

The Primary Articulation box at the bottom shows a dental 'b'. Commander Nemet 17:23, 4 August 2006 (UTC)

Yes, a labio-dental plosive, for those who don't wish to use the grey non-standard symbol in the chart. kwami 07:22, 20 May 2007 (UTC)

Queries regarding reference sources for certain symbols

Hello everyone, I'd like to check a bit into the gray symbols on this chart. As is standard on talk pages, my queries are directed to everyone but I suspect it is Kwami who would be in the best position to answer them.

In preparation for asking here I tried to do a little bit of research in advance. I looked at my copies of Ladefoged and Maddieson (Patterns of Sounds), Pullum and Ladusaw (Phonetic Symbol Guide), and the IPA Handbook. This yielded a reference source (Ladefoged and Maddieson) for the labiodental stop symbols, but none of the others. I probably hunted too quickly and missed others; sorry if so.

In any event, I would appreciate seeing reference sources for the following symbols.

• retroflex lateral fricative
• palatal lateral fricative (Ladefoged and Maddieson use a different symbol here)
• velar lateral fricative (Ladefoged and Maddieson use a different symbol here)
• epiglottal trill
• retroflex lateral flap (Ladefoged and Maddieson mention this but use no symbol)
• retroflex implosive
• subapical retroflex click (Ladefoged and Maddieson allude to this but say it is not a contrastive possibility, which is the criterion for IPA chart inclusion)
• lower high central unrounded vowel (barred cap I)
• lower high central rounded vowel (barred "horseshoe")

Thanks very much, to Kwami or whoever else can help.

PhoneticsPhonology (talk) 03:55, 11 July 2012 (UTC)

Most of these are iconic, and therefore obvious symbols and have probably been used or proposed more than once. ⟨ᵻ ᵿ⟩ are used by the OED; that's the reason they're included in Unicode. ⟨⟩ was used by Ladefoged in his Toda stuff, and therefore added to Unicode recently. ⟨ ⟩ and ⟨⟩ are the obvious choices; SIL added them to to their fonts because linguists have told them they would use them, but have held off proposing them to Unicode until they are more widely attested in the lit. (What were the symbols that L&M used?) For ⟨ᶑ⟩, people have pointed out that this is already supported in Unicode, though AFAIK there haven't been accounts of languages with this sound since then. ⟨‼⟩ has been used for various KS languages. AFAIK ⟨ǃ˞⟩ has not been, but it is the obvious composition if you wanted pure IPA. I've come across ⟨ᴙ⟩ in the lit, but for the life of me can't think of where. — kwami (talk) 08:07, 13 July 2012 (UTC)

Confusing caption on vowels

The caption under the vowel chart is unnecessarily confusing. When trying to assimilate a visual chart like that, the caption should agree visually, thus: "vowels at left & right of bullets are unrounded & rounded." This means that a quick reminder glance at the end of that sentence places the terms "unrounded" and "rounded" in the same order as on the chart, which is far more helpful. Hammerquill (talk) 18:52, 10 February 2013 (UTC)

SVG version

I'll be happy to create an SVG version if someone can tell me how to work out the bugs in Adobe Illustrator CS3. I can import a PDF (all fonts embedded) and export to SVG, but the fonts get screwed up. Gentium works, but not Doulos, Charis, or Gentium Basic, or even Gentium if it's in italics. Kwamikagami (talk) 08:50, 4 August 2008 (UTC)

Now the SVG version is 2.36 MB while the PNG version is only 162 KB. --Mahmudmasri (talk) 02:12, 1 September 2010 (UTC)

j (voiced approximant) aligned on left-hand side

It has come to my attention that the j (w:palatal approximant) is on the left-hand side, suggesting it is not voiced. Is this a mistake or is there some background to this? 84.75.8.21 20:27, 24 March 2014 (UTC) (lKj)

It's not. It's on the right, directly below voiced ɲ and ɟ. Kwamikagami (talk) 02:15, 25 March 2014 (UTC)

Oh, because there's no separation there, so it seemed like j was just below ç. My bad. Kwamikagami, thank you very much for the clarification! 84.75.8.21 12:22, 26 March 2014 (UTC) (lKj)

I wonder if there shouldn't be a separation, though, as [j] is presumably always palatal. Kwamikagami (talk) 01:08, 27 March 2014 (UTC)