Talk:Tenseness/Archive 1

Definition of Tenseness

Tenseness unfortunately does not have a consensus definition among linguists. I gathered some material from newsgroups and websites. Even Peter Ladefoged, in his A Course in Linguistics Third Edition, said once regarding tense and lax vowels:

  These terms are really just labels that are used to 
  designate two groups of vowels that behave 
  differently in English words.  There are phonetic 
  differences between the two groups, but they are not 
  simply a matter of tension...

  In each of these pairs, the lax vowel is shorter, 
  lower, and slightly more centralized than the 
  corresponding tense vowel.

This type of fuzzy definition is not up to today's linguistics standards. Mario Pei equally had a vague definition for lax vowels:

  1. A vowel sound pronounced with lesser muscular 
  tension in the speech organs ([I], [E], vs. [i], [e]). 
  Opposite TENSE. 

  2. A Sound produced with weaker breath pressure, less 
  vigorous action of the lips or tongue (i.e., less 
  muscular strain and deformity of the vocal tract), and 
  correspondingly lesser concentration of energy in the 
  spectrum and in time. 

In Pullum and Ladusaw's Phonetic Symbol Guide:

  Tense.  A problematic term phonetically, it is claimed by some 
  that there is an identifiable class of tense speech sounds 
  characterized by an articulation involving relatively more 
  forceful and extreme motions of the articulators, but there is 
  considerable controversy in the experimental literature about 
  such phonetic correlates. 

  As for "Lax": "The opposite of tense". 

This confusing situation has improved recently. For instance, Melissa A. Epstein, following Ladefoged 1971 and Gobl 1997, put forth the following definitions:

  Lax voice: Associated with loosely adducting arytenoid cartilages,
  loosely adducting vocal folds, and less tension in the vocal folds 
  themselves.

  Tense voice: Associated with tightly adducting artenoid cartilages
  and vocal folds, and greater tension in the vocal folds themselves.

Ladefoged, P. 1971, Preliminaries to Linguistic Phonetics, University of Chicago Press; Chicago. Ni Chasaide, A. and C. Gobl. 1997. Voice Source Variation. In W. J. Hardcastle and J. Laver, eds., The Handbook of Phonetic Sciences. Blackwell: Oxford, 427-461.

An analogy with party balloon may help to clarify the difference between tense and lax vowels. An inflated balloon is like a human lung holding some air pressure. If one holds the balloon's neck fairly tightly and only let open a minuscule gap, the air comes out slowly and can make some whistling sound. This kind of controlled release of the air pressure is analogous to the production of a tense vowel. On the other hand, if one fully releases the balloon's neck at once, one gets some puffing sound instead. That is analogous to the production of a lax vowel. In order to prevent the total loss of the lung's air pressure, human lax vowels must be followed immediately by some form of partial or full constriction. That is why lax vowels tend to occur only at checked syllables. Notice one crucial difference between tense vowels and lax vowels: tense vowels are sustainable, lax vowels are not. ~~~~ 07:28, 8 November 2005 (UTC)

Then GA English has no lax vowels, except maybe reduced schwa, for all others are sustainable. kwami 08:26, 8 November 2005 (UTC)

Not necessarily. English is a tenseness-assisted language, not a tenseness-strict language like Vietnamese. In colloquial speech, the English short vowels often start with lax phonation. But when lengthened, they are continued with tense phonation. The vowel qualities of these two ways of phonation are aurally different, but are allophonic to a native speaker. ~~~~ 17:10, 8 November 2005 (UTC)

Lax Vowel Training and Glottal Trill Technique

Because lax vowels are impossible to sustain and are short in duration by their very nature, it often becomes inconvenient to discern their vowel quality aurally and compare them with other vowels, short of using advanced electronic instruments. Speakers whose native language does not have lax vowels often have difficulty learning their proper pronunciation for a language that has them.

Fortunately, there is a vocal technique that can help to make lax vowels "longer" by repetition: the glottal trill technique.

Imagine taking a cold water shower in a freezing winter day. After it, you may say: "Brrr, it's COOOOOOLD!" with a trembling creaky voice. That trembling is the glottal trill. During glottal trill the glottis opens and closes intermittently.

Using the glottal trill technique, one can discern various vowel-quality differences. For instance, the vowel quality of a short [ɪ] vowel in American English (as in bit) can vary from lax production to tense production. In normal speech, the word bit is often produced with lax phonation and has a short duration. However, when emphasizing, the same word can be followed with tense phonation, with a further retracted tongue root and centraling. Usage of the glottal trill can help to the recognition of these differences and also to the familiarizing of the proper production of the lax-vowel version. Similarly, there is a difference between the American English's [ɪ] and the standard High German's short [ɪ] as in the word bitte. The American version is RTR (retracted tongue root) whereas the German version is ATR (advanced tongue root). The usage of glottal trill can help to their comparison more easily. Glotal trill can also be used to elucidate the sound qualities of English's semivowels and vowel glides (in diphthongs). E.g., to help distinguishing [j] (as the semivowel in year) from [ɪ] and [i] (as the vowel in ear.) (~~~~ 17:10, 8 November 2005 (UTC)

American English

I have removed American English from the list of languages where tense vowels are long. To my knowledge, it is typical for many American English dialects that vowel length is independent from vowel tenseness and rather determined by the voice of the following consonant, e.g. long [bɪːd] vs. short [bɪt]. J. 'mach' wust 09:06, 9 Apr 2005 (UTC)

Although length may not be as salient as other acoustic features, there is still interaction between length and the other features. So, the vowel in bid is longer than bit but it is still much shorter than the vowel in bead. Likewise, beat is shorter than bead but longer than bit. Length and articulation change work together to make the composite feature of tenseness. peace — ishwar  (SPEAK) 07:34, 2005 Jun 13 (UTC)

"... the vowel in bid ... is still much shorter than the vowel in bead."

Not in my dialect; certainly not in my own pronunciation, nor any I'm familiar with. The tense vowels may be slightly longer on average, but to me one of the marks of a British (RP) accent is that the tense vowels are noticeably longer. Vowel length seems to be determined (more or less) entirely by the following consonant and suprasegmental features, such as the timing of stressed syllables and phrasing. — Damezi (tʰɑk tʰə mi 04:17, 22 September 2005 (UTC)

The GA "tense" vowels are diphthongs. They can be considered tense if approximants are considered tense vowels, or long due to the extra segment, but the length is almost insignificant and is not a primary phonetic cue, unlike for example the long vowels in Japanese or French. kwami 04:40, 22 September 2005 (UTC)

Even /i/, /u/ and /ɑ/? (PS: French has long vowels?? I thought it didn’t—but I don’t know French. Where'd they come from, and how are they spelt?) BTW, as I understand it, the shortest RP long vowels are shorter than the longest RP short vowels in any case, so even if AmE does that it doesn't necessarily mean anything on its own. —Felix the Cassowary (ɑe hɪː jɐ) 12:28, 22 September 2005 (UTC)

French long vowels are on their way out of the language. They're mostly due to an etymological /s/, an indicated with a circumflex: maître "master" (long vowel) vs. mètre "meter" (short vowel).

The question for RP is not just whether long /i/ is shorter than short /a/, although that's interesting (since open vowels tend to be longer than high vowels), but how much longer long /i/ is from short /i/. In languages with phonemic vowel length, the difference is seldom much less than 50%. I doubt my long vowels are much more than 10-20% longer, though I really should check. kwami 12:49, 22 September 2005 (UTC)

Sorry, about RP I meant that the /iː/ in ‘leap’ is (I’m told) shorter than the /ɪ/ in ‘live’, a phenomenon known as clipping (frex [1]). I know in my own Australian speech where apparently long vowels are 1.5–1.9× longer than the short versions, if there’s no phonetic short equivalent, long vowels are substantially shortened in this scenario (most noticeable with /iː/, /ʉː/ but still happens with /ɜː/ and /oː/); this is also noticeable with diphthongs. —Felix the Cassowary (ɑe hɪː jɐ) 13:03, 22 September 2005 (UTC)

Ultimately I agree with J. 'mach' wust's decision to remove American English from the list of linguages in which tense vowels are long. The relationship between the American free and checked vowels and either "tenseness" (however defined) or duration is just too messy to say anything definitive here. --Angr/_{tɔk tə mi} 12:55, 22 September 2005 (UTC)

My addition

In American English, in fact, tenseness and duration are indipendent, (litter ˈlɪɾɚ, liter ˈliɾɚ, bidder 'bɪːɾɚ, leader ˈliːɾɚ) however these are not considered minimal pairs since the flap is allophone of two different phonemes (a phonemic transcription would be liter /ˈlitəɹ/ ~ leader /ˈlidəɹ/).

was completely deleted. However, it is true for at least some American dialects. However, I'm not a native speaker so I shouldn't be self-assured about such things, but I read somewhere that ladder was clearly distingushed from latter by the length (and pitch) of the vowel. (That site dealt particularly with Western Canadian English.) Maybe re-adding it, maybe starting it with "In some American English dialects", would be a good idea. However, having seen this discussion about the length contrasts, I prefer being cautious. --Army1987 16:12, 22 September 2005 (UTC)

Yes, I was the one who deleted your addition, for precisely the reasons mentioned. Even if we say "some American accents" (or better, "some North American accents") it's vague because we're not really specifying which North American accents. For my own (relatively General American) speech, I'm quite sure that bidder/bitter, leader/liter, rider/writer etc. are completely homophonous when I choose to flap. (Flapping is of course optional in North American English.) But I know there are other North Americans who do have a quantitative and/or qualitative difference in the vowels of those words. But until someone has pinned down exactly which accents do what in North America, I think it's preferable to stick to clear cases like RP (where tenseness definitely correlates with length) and Scottish English (where it definitely doesn't). --Angr/_{tɔk tə mi} 17:42, 22 September 2005 (UTC)

If the problem is just this one, we could write "In some (Northern) American English dialects, such as Western Canadian English, duration and tenseness are in fact..." etc. until we find some more precise info. As for "some North American accents is better than some American accents", well, in Centro-Southern America there aren't so many (native) English speakers as to make the latter ambiguous AFAIK...--Army1987 20:05, 22 September 2005 (UTC)

My objection to "some American accents" is that it is too narrow (omitting Canadian English, since Canadians generally do not self-identify as American), rather than too broad (including English as spoken in Belize, Guyana, and the Caribbean). Anyway, we already have two English accents identified on the page, RP and Scottish English, so we don't really need more. --Angr/_{tɔk tə mi} 20:27, 22 September 2005 (UTC)

My idea was to show that tenseness and duration are not necessarily related, and there are languages in which all the four combinations are possible. I thought of Northern American English because it's the only one I (and most readers of the English Wikipedia) know. --Army1987 20:55, 22 September 2005 (UTC)

I don't know of a language where all four possibilities are phonemically distinct, but that doesn't mean there aren't any. Icelandic has a phonemic tenseness distinction and phonologically predictable length that is quite independent of tenseness. --Angr/_{tɔk tə mi} 21:02, 22 September 2005 (UTC)

Tenseness and Spanish

Here is more information on tenseness and Spanish vowels: (1) In http://pegasus.cc.ucf.edu/~abrice/IALP04.pdf, we see the sentence: Only tense vowels are present in Spanish (Delattre, 1965), and reference is given for Delattre's book. Delattre, P. (1965). Comparing the Phonetic Features of English, French, German, and Spanish. Philadelphia, PA: Chilton Books. (2) In http://print.google.com/print?hl=en&id=yyqU_tXek1EC&pg=PA29&lpg=PA29&dq=lax&sig=QJO6ktcRgbupWmOtHM5B4UJgl20, we see the sentence: ..., primarily because of a tense/lax contrast not used in Spanish (where all vowels tend to be tense). (3) In http://www.lionelkaufman.com/vowels.doc, page 3, we see a table for English vowels and another table for Spanish vowels. Whereas for English we have the categories SIMPLE LAX, COMPLEX TENSE, for Spanish we have instead SIMPLE TENSE, COMPLEX TENSE. Given these evidences, it would be wise to re-write the paragraph regarding tenseness in Spanish. ~~~~ 08:47, 7 November 2005 (UTC)

In the artice about Bulgarian it is claimed there are only lax vowels in Bulgarian, which seems right to me, but IANAL. Not sure about under stress. —Preceding unsigned comment added by 87.121.52.119 (talk • contribs) 08:43, 21 January 2009

Paragraph on AmEng and Vietnamese

I removed this paragraph from the main article so we can discuss it here:

Complicating the matter is the advanced tongue root (ATR) vs. the retracted tongue root (RTR) distinction. For instance, the vowel in the English word "bit" has a lax RTR vowel in colloquial American English speech. One can artificially lengthen this vowel, but in the process the vowel would actually become a tense RTR vowel. There is debate as to what degree tenseness is a better/worse descriptor than tongue-root position for a language like English. For some other languages like Vietnamese, tenseness is clearly a better descriptor (e.g.: the lax-vowel letters ă and â are never pronounced in isolation, and even their alphabet names are borrowed from the tense-vowel letters a and ơ, only with a modified tone.)

I disagree with the claim about artificially lenghtening the vowel of "bit". I would say that artifically lengthening that vowel does not change anything about its articulation, and therefore whatever the definition of "tense" and "lax" are, a lax vowel does not become tense by being lengthened. As for Vietnamese, I don't know what "ă and â are never pronounced in isolation" means. According to Vietnamese phonology, the letters ă, â, a and ơ stand for /ɐ/, /ɜ/, /ɐː/ and /əː/ respectively, indicating that the primary distinction is length rather than tenseness. --Angr/_{tɔk tə mi} 21:53, 30 October 2005 (UTC)

"Long" and "short" are linguistic misnomers. Please read this: "...The tense/lax distinction correlates with laryngeal activity and concerns different voice qualities. Tense voice involves an increase in muscular tension and thus results in a constriction of the larynx and pharynx, heightened subglottal pressure and raised pitch and volume..." from http://www.ims.uni-stuttgart.de/phonetik/EGG/pagel3.htm A "long" vowel pronounced in 0.1 seconds won't make it short. Similarly, a "short" vowel contorted into a tense-RTR (retracted tongue-root) vowel can last for 10, 20, 30 seconds, but that does not make it long, either. Using "short"/"long" is really confusing, especially because they are not relative terms, either. Native speakers can always tell which is which, independent of their relative durations. If people want to be more precise, they should use the lax/tense plus the ATR/RTR classification, which is what Morris Halle has suggested. At any rate, it is really substandard to present a modern discussion on the issue of tenseness without making reference to related concepts like ATR/RTR, vowel length, etc. As for Vietnamese lax vowels, these vowels must always end with a consonant (to provide the necessary constriction). And more, these vowels don't even have their own names: they borrow their alphabet names from the long vowels, with a modified tone! A proper understanding of tenseness/laxity can only be achieved when a person understands why monophthongic lax vowels are always constricted at the end with a consonant. There is no ambiguity whatsoever on this point.

The page you cited relates to the contrast in German, not languages in general. There are certainly languages with a genuine long/short distinction in vowels, where the only contrast is one of duration. Ladefoged has argued that whatever the so-called "tense/lax" distinction in English is, it isn't ATR/RTR, though of course others may disagree with him. It is interesting that the Vietnamese "lax" vowels are checked vowels, just like the English ones are. --Angr/_{tɔk tə mi} 19:46, 3 November 2005 (UTC)

The paper is universal as far as the observation of tenseness being constricted production of voicing goes (and the opposite: unconstricted voicing for lax vowels). That is the best definition of tenseness/laxity and clearly explains why (monophthongic) lax vowels must be checked. The consonantal "checking" of lax vowels happens exactly because there is no "checking" (constriction) at the glottis. Lax vowels are produced exactly like semivowels (e.g. /w/ and /j/ in English) or glides in English diphthongs. All these sounds (whether one calls them semivowels, glides or lax vowels) are not sustainable since the glottal opening is motionful/widening and lung air pressure is lost all too quickly. That is the main difference with tense vowels. Tense vowels are produced with narrow glottal opening and can be sustained for as long as feasible (up to 1 or 2 minutes for singers/swimmers). English's short vowels have allophonic realizations. In colloquial speech, they tend to be lax+RTR vowels. However, when emphasized, they are continued as tense+RTR vowels. One should not use single dimension to describe English's vowel system. Rather, both tenseness AND tongue-root position must be used in conjunction. Spanish vowels are tense vowels by nature. They are produced without need of checking and they are always sustainable. ~~~~

Also, ATR/RTR is not an English opposition. kwami 21:58, 30 October 2005 (UTC)

At least one person seems to think it is - see this paper (which I have no opinion on). On this subject, though nothing to do with improving the article, I'd like to know what the distinction I make between e.g. cut and coat really relies on. I'm reasonably confident that it's not diphthongisation (I'm from the north of England, where the FACE and GOAT vowels usually don't have much of a glide), and none of the other explanations (except possibly ATR, which does seem to fit with what my tongue seems to be doing in coat) seem very convincing. Not being a linguist myself, I'm not really able to be confident about what's going on. I'd make a recording, but I don't have a microphone.--JHJ 18:22, 31 October 2005 (UTC)

Lip rounding?--Army1987 18:38, 31 October 2005 (UTC)

They're both rounded. (Rounded cut is common in northern England; it's often not distinct from the vowel of put, though in my somewhat RP-influenced accent cut is a bit more open. See foot-strut split.)

According to Ladefoged, who may never have worked with your dialect, ±ATR is not a useful parameter for European languages, including English. Some vowels may be argued to be ±ATR with regards to each other, but this doesn't correspond to the tense-lax distinction: some of the vowels that would be +ATR are "tense", but others are "lax", even for a single speaker of a single language. Diphthongization, even if slight, might contribute a noticeable tenseness to a vowel, I suppose. kwami 18:58, 31 October 2005 (UTC)

Can you give a reference (which I might get round to looking up some day) or quote some examples? It's intuitively clear to me that the four vowels FACE, FLEECE, GOAT and GOOSE share something (which I might call "tenseness"), but I'm not really sure what it is, and the apparent confusion about what "tenseness" is doesn't help.--JHJ 18:40, 2 November 2005 (UTC)

No one really knows what it is. See checked and free vowels. --Angr/_{tɔk tə mi} 19:16, 2 November 2005 (UTC)

Well, no one can give an account that works for all English dialects. It's pretty clear what it is in my dialect, diphthongization (very broadly [ej, ij, ow, uw]). kwami 19:37, 2 November 2005 (UTC)

That's right, JHJ, you wanted a quote from Ladefoged? This (next section below) is from SOWL. Bolding is my own. kwami 09:05, 8 November 2005 (UTC)

Tense/Lax and ATR (quoted from Ladefoged SOWL 1996)

The Akan vowels in figure 9.17 also differ in the height of the tongue in the front part of the oral cavity. This leads us to consider whether the differences between [+ATR] and [-ATR] vowels are the same as the differences between so-called Tense and Lax vowels, which may also differ in both the height of the tonge and the position of the root of the tongue. There are differences of this kind in Germanic languages, as exemplified by pairs of English words such as heed-hid and bait-bet. Following Jones (1956) and a long British tradition, we regard the members of these pairs of vowels as being distinguished by variations of the major vowel qualities, Height and Backness (and perhaps Rounding). We note, of course, that the differences may involve diphthongization implemented through variation in Height and the members of each pair also differ in length. But we do not find it necessary to consider any additional parameters such as tenseness.

We recognize, however, that there is also a long tradition in which these vowels are considered as being distinguished by the feature Tense (e.g. by Bloch and Trager 1942, and by Chomsky and Halle 1968). This leads us to consider two related questions that might be asked at this point. Firstly, are we correct in our phonetic characterization of these vowels as differing only in the regular vowls dimensions of Height and Backness (and Rounding), plus Length? Secondly, are ATR variations the same as Tense/Lax variations? We can get a partial answer to these questions by comparing the vocal tract shapes shown for th Igbo vowels in figure 9.16 and the Akan vowels in figure 9.17, with the pairs of English vowels shown in figure 9.19 or the pairs of German vowels in figure 9.20. In Igbo and Akan the tongue height is not correlated with the tongue root position. In English the position of the tongue root is correlated with the tongue geight (more so for the back vowels than for the front). In German the same is true of the back vowels, but in the front vowels what difference there is in root position would favor the so-called lax vowel having a more advanced tongue root. There is no common setting of the tongue root for the so-called lax vowels that distinguishes them from the so-called tense vowels. This conclusion is supported by statistical analyses of tongue shapes, in which it has been shown that the sagittal position of the tongue in sets of Englsih vowels containing the Tense/Lax pairs can be specified very completely by reference to only two variables (Harshman, Ladefoged and Goldstein 1977, Ladefoged and Harshman 1979). Using similar techniques, Jackson (1988) has also substantiated the finding that in English there does not appear to be a separate control of the root of the tongue; but he did find that there were three independent parameters of tongue shape in Akan. [That is, English only has Height and Backness, while Akan has Height, Backness, and ATR. –kwami] Tiede's (1993) three-dimensional study of pharynx volume using MRI shows further differences between Akan and English. In the pharyngeal region below the epiglottis, Akan shows a positive correlation between the transverse width of the space and the tongue root advancement, whereas in English transverse width is negatively correlated with advancement. Accordingly it seems that the situation in English (and other Germanic languages) is not the same as that in West African languages. Although there may be some increase in the height of the tongue accompanying the advancement of the tongue root in Akan, the changes in tongue height are small in comparison with the expansion that occurs in the pharyngeal region. Furthermore, on some occasions there may be virtually no increase in tongue height for [+ATR] vowels, as is shown in the case of the Igbo vowels in figure 9.16. We conclude that the advancement of the tongue root is a separable tongue gesture in languages such as Igbo and Akan. In Germanic languages, however, it is simply one of the concomitants of vowel Height.

If the advancement of the tongue root is an independent gesture that can be learned as part of the sound pattern of a language, then it must have observable acoustic consequences that distinguish it from all other possible ways of achieving similar acoustic effects.

Lindau (1979) has also pointed out that there are differences between ATR and Tense/Lax characterizations of vowels in the acoustic domain. Figure 9.21 shows the acoustic characteristics of ATR differences in a number of languages. The Akan data is from Lindaw (1979), the DhoLuo from Jacobson (1978), and the remaining languages are from our own files. It may be seen that in virtually all the cases the [+ATR] vowel appears to be raised and advanced in the acoustic space. The only exception is the Ebira lower mid back vowel which is raised, but only very slightly advanced. Among front vowels, paris of vowels differing in ATR have formant frequency characteristics that are reminiscent of so-called tense-lax pairs of vowels in Germanic languages, such as Englsih bead-bid; bade-bed; both retracted tongue root vowels and the lax vowels are lowered and more central in the acoustic space. Among front vowels there is this parallel between [+ATR] and [-ATR] tongue root vowels on the one hand, and Tense and Lax vowels on the other, but among back vowel pairs there is no such parallel. The high back retracted tongue root vowel is always further fack than its counterpart, rather than further forward, as is the case for the traditional lax back vowels. Lax vowels of all kinds are normally taken to be more centralized. Retracted tongue root vowels do not always have this characteristic.

[end section] kwami 09:05, 8 November 2005 (UTC)

Thanks for posting that. I'm still not very sure what's going on, though. Part of the problem is that my GOAT and GOOSE vowels feel central, not back (except the allophones before /l/), so I'm not sure about how the last paragraph applies. But there's no way of telling what's really going on without acoustic analysis, which I don't have the equipment for, so let's leave it here.--JHJ 21:49, 8 November 2005 (UTC)

Comparison between tense and lax vowels

This section needs re-work in light of the new definition.

"In general, tense vowels are more close (and correspondingly have lower first formants) than their lax counterparts. Tense vowels are sometimes claimed to be articulated with a more advanced tongue root than lax vowels, but this varies, and in some languages it is the lax vowels that are more advanced, or a single language may be inconsistent between front and back or high and mid vowels (Ladefoged and Maddieson 1996, 302–4). The traditional definition, that tense vowels are produced with more "muscular tension" than lax vowels, has not been confirmed by phonetic experiments. Another hypothesis is that lax vowels are more centralized than tense vowels. There are also linguists who believe that there is no phonetic correlation to the tense-lax opposition.

In many Germanic languages, such as RP English, standard German, and Dutch, tense vowels are longer in duration than lax vowels; but in other languages, such as Scots, Scottish English, and Icelandic, there is no such correlation.

Since in Germanic languages, lax vowels generally only occur in closed syllables, they are also called checked vowels, whereas the tense vowels are called free vowels as they can occur at the end of a syllable."

4tildes 05:56, 16 November 2005 (UTC)

And why have you decided on a single definition as the truth, when there is no such agreement in the literature? I didn't revert it immediately to give you a chance to respond, but I think you should justify your edit. kwami 06:15, 16 November 2005 (UTC)

All Wikipedia items involve some judgement call on which definition to choose, otherwise each item would have an endless list of definitions. I chose the one that I personally considered to be the most precise. It certainly is fair to include alternative points of views. Two would be OK, three is already a bit too much. If you can find an alternative definition of tenseness that allows you to explain it with several supporting topics (like in my case, I have included balloon analogy, glottal trill, semivowels, tenseness assisted, etc. etc.), we can figure out a way to better organize the article, like moving my part into a separate subsection. If this still does not work, it is OK to move my part entirely to a totally different article, too. The whole point is, it is very disappointing to see people not understanding tenseness and making statements like Spanish vowels cannot be meaningfully described as tense or lax and take it as true. I have seen all too many people (including linguists with academic positions) being confused about tenseness, which really is not that complex if you focus on the glottis area. It is just sad. Wikipedia is not my life. I will be gone soon enough. You guys that stay are the ones that need to look into yourselves and ask: have I really understood tenseness? 4tildes 15:55, 21 November 2005 (UTC)

I agree with kwami. Your edit does not cite its sources (whose definitions are these? whose is the party balloon analogy?), and it's unencyclopedic to claim one definition of a term to be used here when there is no consensus among experts. You added the Ní Chasaide and Gobl reference but you didn't cite them anywhere in your edit, so it's impossible to know what exactly they argue for. --User:Angr/talk 07:09, 16 November 2005 (UTC)

I agree with Angr agreeing with kwami. The recent additions also have an obvious contradiction (or poor explanation), that being that if lax vowels and semivowels are unsustainable, then the glottal trill technique should be impossible (because how you repeat the loss of all the air from your lungs if you have no more air in your lungs?), or how could you produce a vowel to follow the semivowel? (Even if the latter is possible, I think that this physical arrangement would result in semivowels being very marked as onsets—yet that's their preferred placement!) This could just be the result of a poor explanation or understanding, though; kwami and Angr's points are more than enough to disapprove of the additions. —Felix the Cassowary (ɑe hɪː jɐ) 08:51, 16 November 2005 (UTC)

Well, there are at least four vocal techniques with various degrees of faithfulness to real lax vowels. (a) Tense phonation, like the long-duration it in Michael Jackson's song Beat It. (b) Vibrato phonation, used by opera singers, where the tension of the vocal cords changes intermittently, but not enough to fully close the glottis, (c) Glottal trill, where glottis closes intermittently, (d) real lax phonation, which is impossible to sustain. The degree of faithfulness increases from (a) to (d). My point is, glottal trill is only a technique to produce a sequence of vowel-like sounds close enough to the actual lax vowels produced in colloquial speech. It's an approximation, not the real thing. If you want the real lax vowels of colloquial speech, then use actual colloquial speech, and live with the difficulties of not being able to compare them easily to other vowels or explain them easily to another person. As for your question how it is possible not to lose all the air pressure or how it is possible to follow up with another vowel, remember the balloon analogy. If you let open the neck completely, at some point you will lose all the pressure. But if you constrict the neck (tensify your vocal cords) soon enough before all that pressure is gone, you surely can (a) repeat the same process again to achieve glottal trill, (b) follow up with a tense vowel. Tense vowels themselves are a form of partial constriction. It is OK to consider English's semivowels as on-sets and glides as off-sets. But in a language like Vietnamese, they do recycle the lax vowel â for diphthongs, notationally. And phonetically, many of the phonetic lax vowels in Vietnamese can appear (a) in monophthongs, (b) as on-glides (equivalent to English's semivowel) in diphthongs/triphthongs, (c) as off-glides in diphthongs/triphthongs. There is no reason to consider them as separate entities, just because they are monophthongic, on-glides or off-glides. 4tildes 15:55, 21 November 2005 (UTC)

Still smacks of original research, and is therefore inapropriate for Wikipedia. We need to describe the concept as it actually exists in the lit. If this is an alternate view expounded by a particular university or linguist, then we can include it too. Otherwise it doesn't belong. Either way, the other common uses should be there, even if they're nonsense. (In that case, we can state that many think these ideas are nonsense or useless.) kwami 01:59, 22 November 2005 (UTC)