I’ve been around, it’s just that I spend so much time reading here I have no time left over to post! Probably because of the new site format. Instead of jumping into a long thread near the end, I feel compelled to start at the beginning, because instead of discrete pages, it’s all one great big lump. Thanks to ACsenray for summoning me.
That’s remarkably accurate for a “guess”! You’ve posted the genuine Irish phonology. Let’s break down the IPA symbology for everyone to understand what you’ve said.
The superscript character indicates what’s called “secondary articulation.” The primary articulation in each example is the [s], formed by holding the tongue tip near the alveolar ridge and letting the air hiss over the front top surface of the tongue. While doing that, you can also rearrange other mouth parts to modify the basic sound, giving it an altered overtone series of frequencies. That’s the secondary articulation. The subscripts indicate which other mouth part is doing what. The symbols are called “co-articulated” consonants.
[j] is the y-glide we’re all familiar with; e.g. “yell” is [jɛl]. When it co-articulates with another consonant, that sound is said to be “palatalized.” The sound [j] is made with the middle top of the tongue approaching the hard palate. So a co-articulated superscript [ʲ] means while the main consonant is happening, the top middle of the tongue is also moving toward the hard palate.
[ɣ] is a consonant we don’t have in English; it’s the voiced velar fricative. (It’s found in Spanish, Arabic, Persian, et al.) Voiced means the vocal cords are humming along. Velar means the soft palate at the back of the roof of the mouth. Fricative means the air is pushed through a tiny space so as to cause turbulence, making turbulent-air noise. When it co-articulates with another consonant, that consonant is said to be “velarized.” The sound [ɣ] is made by raising the back of the tongue toward the soft palate. So a co-articulated superscript [ˠ] means while the main consonant is happening, the top back of the tongue is also raised toward the soft palate.
To make it easy for non-Irish natives to hear the difference between “broad” and “slender,” listen for the formant frequencies. Palatized consonants have higher frequencies than the plain version. Velarized consonants have lower frequencies.
When you palatalize [sʲ], you’re getting very, very close to the tongue position for [ʃ] (as noted by the OP, our familiar “sh” sound), which is technically “post-alveolar,” meaning articulated just back of the alveolar ridge. Well, you know what’s back there, right? The hard palate! There’s maybe a millimeter’s difference between the two articulation points. That’s why it’s so easy for [sʲ] to slip over to [ʃ]. In fact, in Irish phonology this “slender s” actually does become [ʃ]. So don’t force yourself to articulate [sʲ] too precisely. The good news is you can relax and just say the comfortable old familiar sound [ʃ] we already know and love from English. In Irish, the [sʲ] is purely theoretical.
You’ll need it for Russian, though, where the palatalized s really is [sʲ]. You can tell the palatal sibilants [ɕ] and [sʲ] by listening for the higher frequency formants. Think “white noise.”
By contrast, velarized consonants with their lower frequency formants are like “pink noise.” In English phonology, we simply don’t have anything corresponding to this. English speakers learning Irish, to make it easy on themselves as the OP suggests, may simply ignore it, as long as they concentrate on palatalizing the slender consonants, which is conceptually easier for English speakers because we’re already familiar with the sound [j].
Palatalized basically sounds like the consonant is mixed with [j]. But since [ɣ] is so unfamiliar, velarization will be hard for English speakers to even detect. Hint: listen for the pink noise effect.
If you can palatalize, you can learn to speak fluent, intelligible Irish while ignoring the velarization—but to Irish ears you’ll have a whopping great English accent! If you want to get the true Irish sound, you’ll have to velarize. Learning Irish means drilling yourself over and over, first velarizing and then palatalizing, until they become second nature.
Amazing!
What is “pink noise”?
And I’m interested in your view on the question of whether so many people learning Irish with an English accent is actually changing Irish phonology.
[quote=“Johanna, post:22, topic:918834, full:true”] That’s why it’s so easy for [sʲ] to slip over to [ʃ]. In fact, in Irish phonology this “slender s” actually does become [ʃ]. So don’t force yourself to articulate [sʲ] too precisely. The good news is you can relax and just say the comfortable old familiar sound [ʃ] we already know and love from English. In Irish, the [sʲ] is purely theoretical.
[/quote]
I mean, well, shit, the S is the one consonant in the bunch that’s easiest for me to palatalize in a way that I can hear the difference as well. What if I just keep doing it?
I suppose T and D are next in line in terms of what I have been able to master. Most guides say to pronounce slender T like English CH and slender D like English J, but I’ve been able to actually palatalize them in my mouth.
Sorry, that’s my electronic music background. You know how white noise is all the frequencies at once, like white light is all the wavelengths at once? Pink noise is when a bandpass filter removes the higher frequencies from white noise, leaving the lower ones (analogous with red light, which has the lowest wavelengths of visible light).
As for the main OP question, lemme think about it some more.
Okay, I found a pink noise track on YouTube. It kind of sounds like a rain shower. I’m not sure how to use that to learn the velarized consonants.
If you listen carefully while palatalizing, you’ll hear higher frequencies compared to the plain sound. Listen while velarizing and you can hear the lower frequencies. It just occurred to me that listening to the difference between white and pink noise could shed light on broad and slender sounds. No?
I hope I’m not derailing this conversation (as I find it fascinating, as well). But when I (as a native American speaking Midwesterner) listen to Irish folk who speak English as their first language, I hear alot of breathiness. As If there is air leaking out all over the language (which I assume is what your talking about with palatization). But I don’t hear that same breathiness in the native Gaelic speakers, as in GreenWyvern’s video. Is it there and I’m just not able to hear it or is it unique to the Irish accent?
Thank you for your great post, Johanna!
Noting that the theoretical phoneme /sʲ/ turns out to be /ʃ/ in actuality, you really have to wonder if that drifted in from English phonology. I doubt that it did, since the other Irish phonemes seem unaffected by proximity to English, but it does give pause for thought.
If there were a more general pattern of English phonemes influencing Irish ones, we could expect to see /tʲ/ > /tʃ/ and /dʲ/ > /dʒ/. That hasn’t happened.
“I’ve been working on learning Japanese for years now (admittedly at a very slow pace) and can see why this would be the case. I’d be interested to know if the situation is the same for native Japanese speakers who are learning Spanish. From what I can tell, vowels in Japanese and Spanish sound the same, or at least are a lot more similar to each other than English vowels. Maybe this would eliminate the need for such a process when a native Japanese speaker is learning Spanish instead of English.”
Japanese does sound rather like Spanish, but without the emphasis on the vowels. Japanese has short vowels in the syllables, (e.g. pa, pe, pi, po, pu) but not as short as the English equivalents, which BTW are a major problem for learners of English. Foreigners have problems with the long syllables in Japanese, such as the ‘o’ of Osaka, and with the doubled consonants, such as Nippon or Hokkaido. Japenese has less stress than Spanish, so a Japanese person speaking Spanish would probably sound rather flat. That said, I never encountered such a case while I was there.
Japanese does of course have a huge number of English loanwords and some bizarre coinages that sometimes combine a Japanese word with an English one.
That’s how they’re taught by people on line, though. That’s what made me wonder. It’s interesting to note that this hasn’t changed the “true” phonemes. It’s hard to find sources online to confirm this.
Ah, I found a 2018 draft thesis that seems to address this topic directly – https://academicworks.cuny.edu/cgi/viewcontent.cgi?article=3758&context=gc_etds
I’m still going through it, but I thought I’d share it among those who are experts in the field to take a loog.
Ah, yes. That is a good example. People in the United States learning Irish as a second language (“L2”) change the palatalized consonants into a sequence of consonant + [j] — like in “cute” /kju:t/ — instead of merging the quality of palatalization into the consonant itself, as the superscript indicates.
Irish is never going to be more than an L2 in America. In Ireland, though, is where your hypothesis might be put to the test, where it’s the official language and is in active community use in a few spots. I’ll make a prediction that the core Gaeltacht is not going to be subjected to depalatalization or develarization anytime this century.
But in the majority English-speaking areas, where Irish citizens have to L2 the Irish language, could it become depalatalized as in America? I think that question hinges on the L2s of Ireland becoming integrated into a living (spoken in the home, on the street, in meetings, etc.) Irish-speaking community or just learning it as a school subject in an otherwise English-dominated milieu.
Here’s a key point from that thesis:
Nearly all Irish speakers are Irish–English bilinguals and therefore all have knowledge of both sound systems making it unnecessary for the phonemes of English words to assimilate to the Irish.
I’m not saying you’re wrong. I just don’t understand what this means in a practical sense. Do you have any examples I could listen to?
Try these: Category:Irish terms with audio links - Wiktionary
The velarization can perhaps be most clearly heard in the first syllable of Taoiseach.
What the OP question most reminds me of is the concept of superstratum, where the dominant language influences the subaltern language when the two are both spoken in the same area.
My learning is more about deep history of languages, where the coordinate term substrate is important. Substrate is when speakers of a new language come into an area, take it over, and subordinate the language that was already there. Over time, the subjugated speakers of the native language come to speak the dominators’ language, but bring with it influences from their own languages. This concept is important in explaining the pre-Greek, non-Indo-European elements found in Greek. The old languages vanished without a trace, except for how they affected Greek.
A famous Celtic example of a linguistic substrate is Gaulish vis-à-vis French. The odd pattern in French numerals of quatre-vingt (fourscore) for eighty and quatre-vingt-dix (fourscore and ten) for ninety (replacing Latin octoginta and nonaginta, respectively) was preceded by the Gaulish system of counting by twenties. Post hoc ergo propter hoc? In this case, probably yes!
All I’m saying is that superstratum, while definitely a thing, is less studied than substratum. They also thought up something called an “adstratum” where two languages in contact are on the same level of prestige, but I object to it on the grounds that the same level cannot be a different stratum. Never mind that. I recommend you read up on linguistic superstrata starting with the Wikipedia article references or something like that, to see how well that tracks with the situation of English over Irish. This is the answer I ought to have given in the first place; sorry I took so long to get to it.
I listened to the clip of taoiseach on that page. It is transcribed as a dental:
But I don’t hear it as a dental in that clip. It sounds like the speaker is pronouncing it with an alveolar.
This is fascinating, Johanna! I learned so much from it. I had been somewhat familiar with the concept of substrate, but it’s great to learn that what I was thinking about has a name, superstratum. Thanks!
I suspect the failure to distinguish the aleph (glottal stop) from the ayin (pharyngeal stop) is an example of this. I knew an Israeli whose native language was Palestinian Arabic who told me that he did make that distinction even when speaking Hebrew. But he was an outlier.