Talk:Finnish phonology

Voiced plosives

"However, it is common to hear these clusters eroded in speech ("resitentti") particularly, though not exclusively, by Finns who know little or no Swedish or English and who are not used to making sounds for letters such as d, c or x.

... The letters b and g do occur in Finnish in loanwords, but more often than not, they are pronounced voiceless, /p/ and /k/ respectively."

Perhaps I should just edit the original article but I'm afraid my English wouldn't be good enough... Nevertheless, I found those two sentences a little strange. In my opinion every Finn (exept for some of those who are more than 70 years old) pronounces the word "presidentti" (and most other consonant clusters) correctly... and in Helsinki region hardly anyone ever replaces b's, d's and g's with voiceless consonants. —Preceding unsigned comment added by 195.10.181.162 (talk • contribs) 12:01, 17 November 2004

(Well, apparently this unsigned comment has been hanging about here for ages and the article has been revised since, but I'll just make a quick response.)

Naturally, the typical Finnish pronunciation of presidentti is correct – by Finnish standards. However, even the "proper" /d/ in the Finnish word presidentti is not necessarily the same sound as the /d/ in the English word president. Neither is the /g/ in the Finnish gorilla exactly the same as the /g/ in the English gorilla, or the /b/ in bussi the same as the /b/ in bus. The difference may be difficult to recognize even by an educated Finn. A couple of years ago I learned from my English-phonetics teacher (an English gentleman who has been living in Finland for years) that detailed phonetic studies have found very little if any real difference between words such as bussi and pussi or gorilla and korilla, even when the speakers themselves think their pronunciation of the voiced plosives is proper (by English or some foreign standards). On the other hand, this is nothing to be ashamed about, as it's just a natural way of pronouncing these words in a Finnish context. – Simo Kaupinmäki (talk) 21:49, 9 September 2010 (UTC)

Terminology: "soft D"?

question:

"soft" D means fricative? or voiceless?

Thanks. — ishwar  (SPEAK) 22:29, 2005 Mar 22 (UTC)

hi again. any takers on this question?

what does the term soft in soft d refer to?

does it mean:

• fricative?
• voiceless?
• flap/tap?
• palatalized?

thank you – ishwar  (speak) 2005 July 6 15:43 (UTC)

I don't know where the writer got this term, but it is clear he meant a plosive. See the revised article for a better explanation. Malhonen 12:06, 14 August 2005 (UTC)

Well, it is rather common in Finnish to refer to b as "pehmeä p" (the soft p), and I think d can similarly be characterized as "pehmeä t" (the soft t). Also, d in the Finnish orthography originally stood for the voiced dental fricative [ð], so I'd guess the writer had just confused two phenomena linked to the same grapheme. – Simo Kaupinmäki (talk) 21:49, 9 September 2010 (UTC)

Phonetic realization of /d/

I thought I'd point this out because I'm unsure as to how exactly to best edit this. The realization of the Finnish /d/ indeed varies from dialect to dialect, but that is more of an underlying representation to surface representation variation, and /d/ isn't necessarily altered in place of articulation so much as manner. It pops up as either nothing or /d ð l r j/ and probably a couple other sounds. On the other hand, if there is so much variation, why does it need to be placed in the "dental" category with /t/?

On the other hand, Finnish does have a [d], particularly in Swedish-influenced dialects, and Helsinki slang that is much more back than /t/, and in fact back enough that I would almost say it's retroflex. What should the consonant chart include exactly? Those of you who are native speakers, try pronouncing it as dental if you can, and comparing that to what you might actually say, if you say something like duuni, diggaan, doka and ihQdaa. ;) So my main point is, is that if it's going to be in that chart, it needs to be further back, because that is (as far as I'm concerned) how it turns up in Standard Finnish, when it isn't produced as a flap, fricative, approximant, or anything else. --Ryan 23:50, 23 October 2006 (UTC)

I think the two columns should be joined anyway, since this POA difference is not distinctiv. If a good sorce can be found, we can discuss realization details separately then.

FWIW, I have dental /t l nt ntt/ but alveolar /d n r s tr rt ts st ln nl ls lss sl/; and /tn tr rtt rl lr/ are mixed-POA. Some of the clusters might have free variation, actually. This should serve as an example that the issue is far from trivial. (I tested those out just now — I was surprized especially by /rtt/ being [r̳t̪ː] but /rt/ [r̳t̳].) --Tropylium 17:30, 9 June 2007 (UTC)

Addendum: actually, my /l/ seems to vary between laminal and apical alveolar, not dental and alveolar; in somewhat random variation but generally laminal at least in the vicinity of /j i y/. --Tropylium (talk) 12:03, 9 June 2008 (UTC)

Doubled consonants

Shouldn't the pronunciation of double consonants be covered here? I don't think it's accurate to just gloss eg. the difference between mato and matto or kisa and kissa as phoneme length; the first has (as far as I understand it) a glottal stop, the second doesn't.

(And no, I don't want to plunge in and do it, because I don't know phonetics well enough to do it right.) Jpatokal 05:42, 11 Feb 2005 (UTC)

Well, actually, it is a plain difference of length. See any textbook on Finnish phonetics for references. Glottal stop is something very different: it's the small "pause" you hear between the words if you pronounce very clearly and slowly a sequence like "tien-este" (not "tie-neste"). I modified the text on the /b d g/ question and consonant clusters to better correspond to modern reality. Malhonen 12:06, 14 August 2005 (UTC)

On consonant clusters:

I don't know how to improve this, but these sentences are really strange, false and should be removed. Especially the text about presidentti is made up by the author and not true.

"Originally, Finnish (outside the Southwestern area, roughly the triangle Helsinki-Turku-Kristiinankaupunki) had no initial consonant clusters" "More recent borrowings have retained their clusters, e.g. presidentti ← Swedish president ('president' as a head of state). In the past decades it used to be common to hear these clusters simplified in speech (resitentti), particularly, though not exclusively, by either rural Finns or Finns who knew little or no Swedish or English. "

The text doesn't include anything about the eastern dialects, nor about the numerous s-consanant clusters in the slang of Helsinki. Perhaps the introduction by Heikki Paunonen "Tsennaaks Stadii, bonjaaks slangii" has more useful information about consonant clusters in Finnish. —Preceding unsigned comment added by Hemuli (talk • contribs) 18:01, 25 January 2008 (UTC)

I've actually very much herd "resitentti" in use. You're right, tho, that "outside the SW area" is superfluous here & should be remoovd. The article Stadin slangi, BTW, does mention s-initial clusters briefly. See also Savo Finnish. --Tropylium (talk) 13:06, 28 January 2008 (UTC)

The nature of Finnish "s"

I doubt if Finnish s is the same as the sound described as voiceless alveolar fricative, which is known in English, German, French, Italian as well as in Hungarian (sz) etc. A native Finnish friend of mine once showed and explained to me that it's somewhere between [s] and [ʃ] (English s and sh). Would you please clarify and correct it in the article? Adam78 14:55, 9 December 2005 (UTC)

Was your native Finnish friend was drunk? The Finnish "s" is always pronounced as IPA [s], unless you're a comedian slurring on purpose. If anything, many Finns find š ([ʃ]) a little strange and pronounce it as [s] also. Ref: [1] Jpatokal 15:34, 11 December 2005 (UTC)

No, he was absolutely sober. I'm sorry that you don't perceive this sound as different. Maybe you speak a different dialect, I don't know, or you find it average, which is not surprising if it's your mother tongue and you haven't dealt with it from a foreign point of view. – A "proof" that it exists in standard Finnish is that I have a teacher who examines students at the leading Hungarian language examination centre ([2]) in Finnish language, and when she pronounced a Finnish name at the class, she pronounced that [s] sound differently. (I wouldn't say so if I hadn't noticed the difference, since she doesn't use that sound in her native Hungarian speech, so she must have acquired it specifically for speaking correct Finnish.) Aha, I said to myself, that was the sound I also heard from my native Finnish friend. I clearly remember when I practiced this sound for a few minutes, until this friend said my pronunciation began to approach the native way. Obviously, it wouldn't have been difficult for me if Finnish people were to use the very same sound as exists in English, German etc. Anyway, I'll try to reveal the character of this sound. Adam78 20:17, 11 December 2005 (UTC)

Perhaps this is ceceo or Helsinki lisp, part of the Helsinki slang. Furthermore, since Finnish doesn't really distinguish sh from s, the stridency of s is not as important in recognizing (and thus producing) the sound. It could be a difference between laminal and apical articulations, also. --Vuo 22:44, 11 December 2005 (UTC)

Thank you! Do you mean this sound is not part of the received (standard) pronunciation of Finnish (if there is such)? If it isn't, can the standard variation of Finnish be localized in the country? Adam78 18:32, 12 December 2005 (UTC)

Pretty much the only deviation from the standard language when it comes to pronouncing /s/ is the (feminine) Helsinki lisp. On the other hand, if you're going to compare to other languages, then the standard /s/ may sound "softer". This is because Finnish has only this fricative, and not much effort is required to distinguish from other sounds, unlike in e.g. English with both 's' and 'sh'. --130.233.243.228 01:42, 14 March 2006 (UTC)

This indeed is the case, Finnish s is softer than the English one, since Finnish doesn't need to distinguish s and š. The accurate pronunciation is between [s] and [ʃ]. --Fagyd 19:24, 27 February 2007 (UTC)

AIUI, in more technical terms this would be: Finnish, and generally languages with no /ʃ/, have a laminal [s̻], while languages that do have both sibilants, have an apical [s̺], assumably to maximize the contrast to the naturally laminal /ʃ/.

It could be relevant to note here that Germanic s-initial loans in Finnish that were acquired at a time when the sibilant situation was the reverse (one in G, two in F), indeed have nowadays an initial /h/ (deriving from a former /ʃ/). Two examples: huoma- "notice" from *sooman- (CF Old Norse sómi) hidas "slo" from *siiþaz (CF German seit). --Tropylium 18:14, 9 June 2007 (UTC)

Open front unrounded or near-open front unrounded?

The article says that the Finnish vowel ä is the Open front unrounded vowel. To me (I'm a native speaker but not a phonetician) that vowel's sound file in the article on that vowel sounds nothing like ä. The Near-open front unrounded vowel however sounds exactly like ä and even the IPA symbol looks much more like the one used in the phonology article. Has there prehaps been some kind of a mistake? Ossi 22:28, 21 March 2006 (UTC)

Artificial symmetry?

I removed this from the article:

Some phoneticians have also raised the question, whether it would be more appropriate to mark the open front vowel as /a/ instead of /æ/. This would make the Finnish phonological system seem more symmetric, since IPA [æ] is strictly speaking a little higher than the open front vowel [a]. Some acoustic studies seem to indicate that there is in fact no significant difference between the orthographical a and ä in terms of vowel closeness.

The perceived quality of /ä/ is [æ], and vice versa. I and the writer above can confirm this as native speakers. Trying to make /ä/ an [a] because it's more symmetric is prescribing, not describing. Real languages don't need to have perfectly symmetric vowel systems, and Finnish is a real language. --Vuo 22:09, 26 March 2006 (UTC)

Just thought I might throw in that you are absolutely correct, and I can confirm this. On the other hand in the sense of speech-production the finnish /ä/ sometimes bleeds into what would be described as [a] (as opposed to [ɑ]). This is perhaps what the previous writer was getting at, on the other hand this does not reflect the situation in 'Standard Finnish', which is, to my understanding, what is being talked about here. ;) --Ryan 11:17, 18 April 2006 (UTC)

Well, when I listen to Helsinki area people, women in particular, I often hear a slightly nasal sound [a] for the phoneme /ɑ/. So, /sɑlille/ may be realized as [salille]. But I Am Not A Phonetician, so I'll leave the article alone. --Vuo 23:32, 18 April 2006 (UTC)

On the other hand, many people from the Helsinki area, such as myself (didn't realise it until it was pointed out to me), do the exact opposite: they pronounce /ä/ as [a]. I think what you are describing is not a linguistic trait of the Helsinki area but a recent phenomenon of trying to sound 'posh', found in many Finnish localities. --Oghmoir 09:46, 26 August 2006 (UTC)

If it involves language, it's linguistic. People trying to sound 'posh' is definately a sociolinguistic thing, too.  ;) Anyway [a] isn't necessarily more nasal than /ɑ/ unless you all mean something like [ã] (though Finnish wouldn't have things that nasalized). In Helsinki I noticed the sort of change in quality of the vowel represented by 'ä' to be not so much [æ], as [a]. Nasalization indeed does occur (at least in Helsinki Finnish) in nasalizing environments, although not all changes with these vowels are going to mean that nasalization is involved. If you do hear some sort of nasalization in non-nasalizing environments, please do mention it. --Ryan 17:46, 26 August 2006 (UTC)

Ryan, you're right. I wasn't saying what Vuo was talking about isn't linguistic, I was expressing my doubts about whether it is characteristic to the Helsinki area. :) And yes, slight nasalisation is quite common in the environment of nasal consonants, but also, nasalisation can be heard in non-nasalising environments, too, in the speech of the aforementioned people, especially girls (I see myself as neutral in gender issues, that is a neutral observation!), sometimes to annoying degrees. That is part of the stereotype about those types, but it is based on reality, I can verify that, hehe. --Oghmoir 18:00, 26 August 2006 (UTC)

Diphthongs as sequences of vowel phonemes or as independent phonemes

When I renewed the section about diphthongs I introduced the view that diphthongs should best be viewed as sequences of vowel phonemes rather than phonemes of their own. This is the view I learned at the lectures on the phonology course I attended during my studies of linguistics at the university, and as far as I remember, it is based on the fact that a) it is sufficient to analyse the diphthongs as sequences of monophthong phonemes as it can be done; for all diphthongs there can easily be found a respective sequence, which in standard speech even reflects the pronunciation, and b) it is preferable to do so according to the principle of economy: the less phonemes you need to represent the phonology of a language, the better. (There might have been additional reasons but my mental lecture notes are not that accurate after a few years.) Hence, the diphthongs are best analysed as sequences of vowel phonemes.

However, User:Vuo edited the section to present the view that the diphthongs should be viewed as their own phonemes. In his edit summary he wrote, "This is a very dangerous statement because it implies an insertion of a glottal stop," concerning my statement about diphthongs being phonologically vowel sequences. Unfortunately, I don't understand at all what he means. I am not familiar with a phonological model of Finnish that dictates that between every sequence of two vowels there is automatically inserted a glottal stop (as his wording implies), and I don't see it as being useful. I have contacted the user Vuo on the matter and I hope he can clarify his view so the matter can be settled. --Oghmoir 17:15, 26 August 2006 (UTC)

In the context of several other languages, the "separation of vowels into monophthongs" unambiguously means separation with a glottal stop, hiatus or stress. In particular, I'm thinking Japanese. Phonologically, dipththongs are defined as phones that glide from one vowel position to another during articulation, and this is what the Finnish diphthongs are. Calling them something different for grammatical reasons is going to mislead a lot of readers. In phonemics, it is appropriate to analyse the diphthongs as phoneme sequences. This does not mean they are physically speaking diphthongs. --Vuo 17:27, 26 August 2006 (UTC)

When you say, "Phonologically, dipththongs are defined as phones that glide from one vowel position to another during articulation," you are wrong. Phonetically that is true. You seem to be confusing phonology with phonetics (in phonology there are no phones but phonemes, phones are the units of phonetics). It is exactly my point what you say here: "In phonemics [=phonology], it is appropriate to analyse the diphthongs as phoneme sequences. This does not mean they are physically speaking diphthongs." Phonology is specifically about analysing sounds as phonemes. Physically, ie. phonetically speaking, the beginning of a diphthong does gradually glide to the end part, so there doesn't seem to be a sequence of two sounds, but what I'm talking about is that they don't constitute their own phonemes, ie. phonologically speaking, they are not single sounds but sequences of vowel phonemes.

I am going to make this clear in the article. If you have something to add to that, please do so. --Oghmoir 17:41, 26 August 2006 (UTC)

O

It's definitely an open-mid back unrounded vowel.. try for yourself. The IPA [o] sounds more like "ooh" than the "real" O to me.. I'm not even going to argue, I speak Russian as my native tongue: there the O is realized as a close-mid back rounded vowel (IPA: [O]). In layman terms, comparing Russian vowels to Finnish vowels, Finns pronounce it more deeply, lower voiced, in the back, whereas Russian pronoun it more in the front. Seriously. Also, I'm not really sure about the e, either. Sounds more like an open-mid front unrounded vowel. --84.249.253.201 00:21, 23 December 2006 (UTC)

I'm a nativ speaker and I agree with these observations, at least as far as my own idiolect goes. Close-mid [o] is a sound I don't think Finnish uses at all; proper mid [o̞] is common, but so is open-mid [ɔ], especially in the /uo/ difthong and in the speech of peeps like me who have a fronted /ɑ/. This results in a slightly raised /e/ too, however; but for those with a backed /æ/ insted, I think this might be reversed. --Tropylium 17:05, 9 June 2007 (UTC)

http://de.wikipedia.org/wiki/Finnische_Sprache#Phonologie - The German Wikipedia seems much more accurate to me, and even ö doesn't sound like the IPA [ö] exactly (when I heard some German, the ö the person pronounced sounded more like [y] to me, but it's the [ö] ö, not the [œ] ö I thought was the "real" one as a Finn". The Finnish phonology involves a more open mouth (less rounded) from what I can tell. Since OR is not allowed, I suggest using the German article's sources on phonology. :) --nlitement ^[talk] 02:09, 28 June 2007 (UTC)

I wouldn't use any of them, neither the English nor the German. Instead I would do the same as Suomi, Toivanen and Ylitalo in Finnish Sound Structure,* p. 21, where they treat the Finnish mid vowels as being between [ɛ], [œ] and [ɔ] used in the German Wikipedia and [e], [ø] and [o] used here in the English Wikipedia, respectively, thus [e̞], [ø̞] and [o̞]. It is also worth noting that [ɑ] has the same height as [æ], thus [ɑ̝]. I'm not a native speaker, but I have grown up in various Finnish environments, and [e̞], [ø̞], [o̞] and [ɑ̝] are definitely how I would pronounce the vowels. -- Llonydd (talk) 08:52, 31 December 2009 (UTC)

voiceless velar fricative? voiceless palatal fricative?

The article and its consonant chart mention neither the voiceless velar fricative nor the voiceless palatal fricative even though those pages have "lahti [lɑxt̪i]" and "vihko [ʋiçko̞]" . --Espoo (talk) 14:56, 21 January 2008 (UTC)

fixed. --Tropylium (talk) 13:06, 28 January 2008 (UTC)

Could you please also add them to the consonant chart?

Since these are allophones only they don't belong in the consonant phoneme chart.

Please also explain what happens between the other vowels and a consonant, "intermediate" is no help for non-experts like myself.

It's in the literal sense. I don't think there are any appropriate technical terms. Nothing "happens" tho, it is entirely standard for /h/ to take on features from the adjacent sounds (vowels, in this case). Eg [ʍ] is just the closest consonantal counterpart to a voiceless vowel [u̥], and [xʷ] is the same except with more friction. The reason, I think, why the article should be concerned with the exact articulation of Finnish /h/ anyway is exactly the friction appearing in some environments, not the general "voiceless approximant" quality. Maybe this should be stated more clearly.

Did i correct the pronunciation of Pohja accurately?

[h] is probably still the best approximation. The velar place of articulation corresponds to close back vowels only, of which Finnish only has the basic /u/. I suppose uvular [χ] might have some link to [o], but that consonant usually has a very harsh, strongly fricated articulation, and here the folloing /j/ 1) keeps the articulation soft (I think, I seem to notice friction occuring mostly before /t/ and /k/? might need a quote here, going to OR territory otherwise...) and 2) nudges it forwards - but along the middle, ie. the vowelspace portion of the mouth, not the palate. (I don't think "vowelspace" is an official linguistic term but I hope the meaning is clear enuff.)

And are [ʍ] and [xʷ] really correct? I thought the former is the sound of "wh" in UK English "whine", which doesn't seem to exist in Finnish, and the article on voiceless velar fricative uses /x/ after /u/. --Espoo (talk) 16:58, 9 February 2008 (UTC)

Far as I gather, it is the same sound, just obfuscated by the fact that it only appears syllable-initial (before a vowel) in English, and syllable-final (before a consonant) in Finnish. Try dropping the /l/ from "juhlat" but keeping the articulation of the /h/, and see if you can hear the resemblance with "you what?" :)

I was using slightly wider transcription on the [x] article, rounded vowels such as /u/ and /o/ very commonly labialize adjacent consonants but this usually left untranscribed unless the distinction is phonemic or one wishes to draw special attention to the fact. --Tropylium (talk) 11:17, 11 February 2008 (UTC)

raa'an?

Gradation of single k leads sometimes into long vowel groups (e.g., raaka → raa'an, ruoko → ruo'on). Could someone explain how they're described in IPA? At least in the colloquial speech there is no glottal stop in the normal speech. A change of tone, or what?

An unrelated comment, shouldn't it be shown in the IPA chart that t is normally dental [t̪]? And secondly, are you sure that maha is really [mɑhɑ] instead of [mɑɦɑ]?

Oh, and please do add some references. — JyriL ^talk 20:32, 2 April 2008 (UTC)

I can't add some references, but my own feeling is that raa'an at minimum has a clear syllable break and duration associated with syllable lengths. It's otherwise clearly distinguishable from a word like *raan. A glottal stop may be rare, and in existence in slow and careful speech (or in existence when you ask a Finnish speaker how to pronounce it, for instance). I feel like it's comparable to ruoan, anyway.

As for maha, in Helsinki Finnish (and probably other Finnishes), it is really pronounced [mɑɦɑ].

In terms of gradation of single k, I discovered some other new and fun words that may delight and amuse: rei'ittää 'preforate'; which leads to rei'itin 'preforator' and so on. It comes from reikä 'hole'; rei'issä 'in (the) holes'.

Pyry (talk) 17:01, 19 August 2008 (UTC)

vaan 'but', vaa'an 'scales (sing.gen.)' (the weighing kind), van (in surnames of foreign origin), va'an 'venerable (sing.gen.)' is a fairly impressiv minimal quadruplet of this sort. Yes, it's an interesting issue. Note that a "syllable breik" is a theoretical construct and not any actual phonetic element. The distinction seems to rely on tonal factors in addition to length. I bet it's been reserched more than a few times by now.

BTW while this always seems to originate in a lost */ɣ/, also note the defaulting to a plain long vowel in taakse, teen, reen, täällä, hään, rään (not *ta'akse, *te'en, *re'en, *tä'ällä, *hä'än, *rä'än, …)

Heterovocalic clusters are easier to explain, at least in phonetical terms — they're (long) semivowels. kiuas [kiwːɑs] (or, if you will, [kiu.wɑs], taiat [tɑi.jɑt] and so on. (Some prescriptivists may try to tell you that there is a distinction between eg. haltia and haltija, but that'll be a hypercorrectiv spelling pronunciation at best.)

--Trɔpʏliʊm • blah 22:06, 19 August 2008 (UTC)

I'm also tempted to label rei'ittää just a quirk of spelling: most people pronounce, and occasionally misspell, it as reijittää. Jpatokal (talk) 09:52, 20 August 2008 (UTC)

Reijjittää. Or kind of "rei'jittää", so that the glottal is assimilitated with the following j. I assume this is very common way saying it... --62.248.174.190 (talk) 16:56, 24 December 2013 (UTC)

The L sound

Isn't the Finnish L sound more like [ɫ], unless the previous and next vowel are both front vowels?

E.g. halla [haɫːɑ], helle [helːe]. --85.156.230.132 (talk) 10:32, 17 June 2008 (UTC)

"More like", possibly in some contexts, but that doesn't necessarily imply that it should be transcribed (or, is actually transcribed by anyone) as such. To my ears it's much less velarized than in languages where the contrast is phonemic, or even in ones like English, if at all. This seems like one of those way minor phonetic variations that "nobody cares about"; things like [sɑt̪ɑ] - [sɑt̪ʷu] and from there on down. But hey, maybe you can find some paper that describes this variation too. --Tropylium (talk) 11:52, 17 June 2008 (UTC)

The wrong quality of L reveals the foreignness of the speaker. --85.156.228.123 (talk) 19:50, 17 June 2008 (UTC)

It's NOT velarized (and I'm saying this with Russian and Finnish being both my native languages), but the halla/helle difference does exist. The L in halla is closer to being a retroflex lateral approximant, while in helle it's alveolar. This is one of those things that phonologists fail to note and if you tried to speak Finnish exactly according to the IPA table in this article, you'd sound rather foreign. The German Wiki article of this is better. --nlitement ^[talk] 23:43, 14 February 2009 (UTC)

I agree with the OP, /l/ is definitely velarised when in proximity to "back" vowels and pronounced without velarisation when around "front" or "neutral" vowels. I actually came here to write that but found that someone already did. 122.148.127.225 (talk) 23:08, 13 October 2014 (UTC)

Geminates

What about the participle tehtykään /tehtykkään/? --Vuo (talk) 12:52, 2 August 2008 (UTC)

Smacks of the dialectal gemination processes a la "ossaa vallaa tinnaa", but don't quote me on that. --Trɔpʏliʊm • blah 07:30, 8 August 2008 (UTC)

Too difficult to understand

Am I the only one who finds this difficult to understand?

Can we perhaps have a more layman like introduction with perhaps pointers to other articles that may help us to understand the rest of what is to come? More wiki-links would be helpful when some special phonetic characters or terms are used.--Hauskalainen (talk) 20:47, 12 February 2009 (UTC)

Are you being sarcastic? There are a number of ways to improve the intro but making it more laymanlike is not one of them.

What terms are/were you confused by? — Ƶ§œš¹ _{[aɪm ˈfɻɛ̃ⁿdˡi]} 20:50, 12 February 2009 (UTC)

I'm sorry you received such a rude answer, User:Hauskalainen, but it seems it was not meant that way. When people become experts in a field, they often lose the ability to talk about it in a way that is comprehensible to nonexperts, but this is of course completely unacceptable in an encyclopedia meant for the general public. This article is indeed much too difficult for almost all Wikipedia users who haven't studied linguistics. --Espoo (talk) 13:08, 11 October 2016 (UTC)

Why should readers believe any of this?

Aside from one very minor and peripheral point, the whole article is unsourced. It reads as calm, reasoned and informed; but the native or near-native intuitions of a variety of phonologically informed and well intentioned editors aren't enough. Where can one find verification for all (or much) of this? -- Hoary (talk) 00:10, 12 July 2009 (UTC)

I've added a mention of and link to Suomi, et al., Finnish sound structure. NB my description of it as "further reading" is deliberate; it's not a reference, as I have not checked the content of the article against it. -- Hoary (talk) 23:35, 13 July 2009 (UTC)

The recently online-publish'd Iso suomen kielioppi ("The Large Finnish Grammar") contains a fair amount of information for anyone wishing to verify the contents of this article. --Trɔpʏliʊm • blah 13:59, 15 July 2009 (UTC)

Gradation

Okay, so "t → d (lato - ladot)" is a change caused by gradation, which happens when "If the onset of the last syllable is a plosive, it is subject to consonant gradation" Okay, that's fine... except the onset of the last syllable in "lato" is a plosive, yes? 75.180.45.139 (talk) 02:00, 18 May 2010 (UTC)

Good catch. Someone seems to have been confused about the conditioning: syllable closure is the general condition, and exceptions like pue! result from secondary loss of syllable-final consonants (*pukek > *puɣek > *puɣe > pue) --Trɔpʏliʊm • blah 16:08, 19 May 2010 (UTC)

Two recent additions

Per the request in the edit summary of this edit, the two edits are redundant at best but more likely self-contradictory.

The first addition says that the mid vowels are "true mid vowels" (something the vowel chart shows already) but then says that they're open-mid. Which is it? Is there a source?

The second addition says that /i/ and /e/ behave as front vowels in the absence of back vowels. But the paragraph it's attached to says that they are "neutral to harmonic processes" meaning they also behave as front vowels in the presence of back vowels. — Ƶ§œš¹ _{[aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi]} 23:45, 14 August 2010 (UTC)

Well it's clearly self-contradictory to claim that they both "behave as front" and "behave as neutral" (and it doesn't tell anything new over the latter to say they "behave as front in the vicinity of other front vowels"). The attempted point here, I think is that they're phonetically front vowels (but that's also indicated previously).

Something ought to be said on what suffixes they do take, however: while derivational suffixes vary quite a bit (mene- >meno 'a going; an errand'; > menetys 'a loss') , grammatical suffixes generally default to front (mennä ' to go'). --Trɔpʏliʊm • blah 16:46, 6 September 2010 (UTC)

Wiik's data: http://www.helsinki.fi/puhetieteet/projektit/Finnish_Phonetics/vokaaliakustiikka_eng.htm --vuo (talk) 18:52, 7 September 2010 (UTC)

Letters vs pronunciation

Two points:

1. "letters indicate pronunciation, they are not themselves pronounced" does not make any sense. "Z" is a grapheme that represents multiple phonemes, including t͡s; or, to render that in English, Z is a letter that can be pronounced multiple ways.
2. The statement "Z is often used in foreign words and names such as Zimbabwe to indicate the pronounciation [t͡s]" implies, entirely incorrectly, that Zimbabwe is spelled in Finnish that way in order to be pronounced t͡simbabwe. This is naturally not the case; the spelling is imported wholesale and it can be variably read as "simbabwe", "t͡simbabwe", etc.

Discuss. Jpatokal (talk) 06:16, 27 December 2011 (UTC)

I'm not sure how much clearer I can be than "letters aren't pronounced." In the word ship, the "sh" isn't pronounced /ʃ/, it represents /ʃ/. That's the only thing I'm trying to change with my edits. It wasn't my intention to change any other meaning.

How about this: "‹Z›, found mostly foreign words and names such as Zimbabwe, may also represent [t͡s] as is the case with German orthography." — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 16:21, 27 December 2011 (UTC)

I remain puzzled by your assertion. How do you pronounce the word "ship"? And what are you pronouncing if it's not the phonemes dictated by the letters "s", "h", "i", and "p"? (Sure, the first two letters are a digraph, but you're still pronouncing the sound of that digraph in English.)

And once more: the word "Zimbabwe" is not Finnish and it's not originally pronounced with a t͡s. It's thus incorrect to say that the "z" represents t͡s, because it doesn't; the intention of the sentence is to state that, when confronted with a foreign word containing Z's, Finns may pronounce those Zs (yes, the letters) as t͡s. Jpatokal (talk) 22:21, 27 December 2011 (UTC)

Orthography is a reflection of pronunciation, not the other way around. Thus, the word ship is a reflection of the phonemes in speakers' heads. So the letters aren't dictating the phonemes; if anything, it's the other way around. The sequence of phonemes determines the spelling. This is even true of words learned through writing and words pronounced with what's called spelling pronunciation. Speech acts occur independently of orthography, which is why illiterate people can still speak perfectly well. — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 23:04, 27 December 2011 (UTC)

I'm not sure how much simpler I can make this.

1) The written word "Zimbabwe" is an English transcription of a word in Shona. In this case, for English, the orthography does indeed reflect pronunciation.

2) The spelling of word does not conform to native Finnish orthography, which does not use the letters 'z', 'b' or 'w'.

3) The word was nonetheless imported wholesale into Finnish, in the original non-Finnish spelling.

4) The Finnish speaker thus has to decide how to read out loud (that is, pronounce) a word that is not manifestly Finnish. They can either approximate the English pronunciation and read it "Simpapve" (which is really how the word should be spelled in Finnish), or they can use the 'default' German pronunciation of the letter Z (see?) and read it "Tsimpapve" (which doesn't sound anything like the original).

5) But neither is intentional: that is, the Finnish word "Zimbabwe" is not spelled that way because it's supposed to be read "Tsimpapve". The "Z" is not meant to represent the Finnish phoneme t͡s, it's meant to represent the English phoneme /z/.

More generally, for any language, the letters most certainly do dictate the phonemes if you are not familiar with the word. Can you take a stab at pronouncing "Fin-tim-lin-bin-whin-bim-lim-bus-stop-F'tang-F'tang-Olé", despite never seeing or hearing it before? Jpatokal (talk) 01:40, 28 December 2011 (UTC)

As you agree with my general sentiment, I don't feel the need to pedantically rehash the nuances about why, in general, it's a bad idea to say that letters are pronounced. You have explained your position to such a degree that it calls me to question whether this isn't original research. Is the issue of what sorts of guesses speakers make when encountering foreign letters even relevant to a phonology article? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 02:23, 28 December 2011 (UTC)

I'll take that as a grudging admission of defeat ;)

And the article contains one paragraph about possible pronunciations of foreign fricatives, so a sentence about how 'z' is treated seems pretty relevant. Jpatokal (talk) 22:22, 28 December 2011 (UTC)

At this point, I'm ready to delete the whole paragraph. It's uncited, marginally relevant, and most of it is challenged. Do you have any sources for it? — Ƶ§œš¹ _{[ãːɱ ˈfɹ̠ˤʷɪ̃ə̃nlɪ]} 23:20, 28 December 2011 (UTC)

The only bits that are challenged are the claims about "educated speakers" doing or not doing various things, which are indeed dubious and should probably be nuked. The rest seems quite straightforward and easily verified, see eg. [3] for the case of ʃ and its buddies, plus large slabs of [4] for Finnish orthography and the phenomenon of "citation loans" (sitaattilaina). Jpatokal (talk) 02:49, 29 December 2011 (UTC)

/i/ ending diphthongs

Are you sure i in the latter part of a diphthong is pronounced /j/??? For example, aika (time) is definitely /ˈɑikɑ/, NOT /ˈɑjkɑ/! However, /j/ often appears between /i/ ending diphthong and vowel, at least in the spoken language: reiän /ˈreijæn/.

Phonotactics

The section on phonotactics seems to contradict other parts of the article. It says that we can't have "nasal + non-homorganic obstruent (except /nh/)" word-medially, but a little further up we've got /ŋk/ in kenkä; it says we can't have liquid + liquid word-medially, but there's the example of kiellot below kenkä. Is this analysis done at a different level or something? — Lfdder (talk) 17:07, 29 April 2013 (UTC)

Is /ŋ/ not homorganic with /k/? And I think long consonants are not analysed the same as sequences of consonants. In fact, historically, earlier sequences of -ln- have developed into -ll-, which you can still see in the past active participles of many verbs like olla. CodeCat (talk) 17:24, 29 April 2013 (UTC)

It is, I don't know what I was thinking. Fair enough. We should probably have a paragraph on how long consonants/geminates are treated. — Lfdder (talk) 17:38, 29 April 2013 (UTC)

Nearly all of the comments on consonant clusters are actually incorrect, if we're considering modern Finnish.

• Via loanwords at least /m/ is somewhat established word-finally, e.g. islam, kalsium)
• Initial clusters:
  • /s/ + consonant is about as well-established as stop + liquid. E.g. skeittaaja, slummi, snapsi, sponsori, stereot.
• Final clusters:
  • indeed not found in standard Finnish, but hella frequent in colloquial speech in all the major cities (if highly limited; mostly /lt st ks/).
• Medial clusters:
  • CCCC: ekstra, hamstrata
  • stop + nasal: etninen, tekno, hypnoosi
  • labial stop + non-labial stop: adoptoida, helikopteri, Raptori
  • non-dental stop + semivowel: ekvivalentti
  • nasal + non-homorganic obstruent: limsa, jamssi (admittedly nasal + non-homorganic stop basically does not occur)
  • nasal + sonorant: anjovis, invalidi, kanveesi, kamreeri, timjami
  • liquid + liquid: jaarli, kurlata

It really sounds like this is based on a source actually discussing the inherited Finnish lexicon (given the claim that /eː oː "œː"/ can be long "in cases of contraction") but the claim that stop+liquid combinations are allowed is throwing me off. --Trɔpʏliʊm • blah 23:50, 29 April 2013 (UTC)

It probably does, but these restrictions did still apply to Finnish up to maybe a hundred years ago, and may even still apply in some rural areas. People pronouncing "presidentti" as "residentti" is a notorious example, although I don't know how frequent that still is. "jaarli" and "kurlata" are older loanwords, so I think -rl- has been permitted in the language for quite a long time by now. Most of these consonant clusters are clusters that were either never present in the Proto-Uralic lexicon, or were eliminated through sound change. -tn- changed to -nn-, -ln- to -ll-. I don't think there are any restrictions on -nj- though, are there no inherited words that have it? CodeCat (talk) 02:01, 30 April 2013 (UTC)

I'm not sure if this level of analysis represents any general sort of Finnish circa 1900 as much as Finnish circa 1800 minus all transparent loanwords. Dialects are indeed another case entirely.

Re -nj-, I don't think Finnish has had any nasal+semivowel until the recent centuries actually? It's curiously absent from inherited vocabulary despite inherited -lj- and -rj-. Vanja "Russian" or linja (first attested mid-1700s) might be the oldest cases. Iso suomen kielioppi indeed lists -nj- among the clusters found in nativized vocabulary (though there is no clear boundary between foreign/nativized/native clusters, beyond the confines of what phonemes can be counted as native). I've not checked the source for this edit but I'd be tempted to fix the section to go along with ISK firsthand. Sources can probably be dug up to establish various details of the nativity cline. --Trɔpʏliʊm • blah 15:18, 30 April 2013 (UTC)

Labio-velar approximant?

There is no mention of the labio-velar approximant in the middle of words such as [ˈkiuwːɑs], which is especially interesting and noteworthy because it's one of the few sounds not shown in Finnish spelling. In fact, it's even missing from the article on the labio-velar approximant. --Espoo (talk) 15:20, 16 May 2014 (UTC)

I don't believe this is an independent phoneme in Finnish. You're describing a glide between two vowels, not something that can appear independently. Furthermore, I haven't seen a single source that would support this claim. --vuo (talk) 21:29, 16 May 2014 (UTC)

I don't think Espoo is claiming [w] to be an independent segment? And yes, optional epenthetic [w] ~ [ʋ] / {u y}._V is a well-recognized phenomenon in Finnish. Cf. e.g. Iso suomen kielioppi § 25.

Also note that there is a contrast with /ʋ/: ruu^[w]in 'with foods' (comit.) vs. ruuvin 'screw's' (gen.), valu^[w]a 'to flow' vs. valuva 'flowing', etc. --Trɔpʏliʊm • blah 16:13, 17 May 2014 (UTC)

Unclear section

In the section Consonants, there reads:

The rest of the foreign fricatives are not. 'š' or 'sh' [ʃ] appears only in non-native words, often pronounced 's', although some educated speakers make a distinction between e.g. šakki 'chess' and sakki 'a gang (of people)'.[citation needed] The orthography also includes the letters 'z' [z] and 'ž' [ʒ], although their use is marginal, and they have no true phonemic status. For example, azeri and džonkki may be pronounced aseri and tsonkki without fear of confusion. The letter 'z', found mostly foreign words and names such as Zimbabwe, may also be pronounced as [t͡s], thus 'Zimbabwe' /tsimpɑpʋe/, given that Finnish lacks also /w/ and the /b/–/p/ contrast is inconsistently followed.

These somewhat strong claims which might be true, but I'm somewhat skeptical, especially as it seems to lack citations. I am somewhat surprised to see that šakki should sometimes be pronounced with 's'. It is often written with sh, but still, as far as I know, pronounced š (English sh). The part about z seems to be true (though sources are needed), z is often pronounced ts or s, but Zimbabwe being pronounced /tsimpɑpʋe/ is something I have hard to believe. Though Finnish doesn't have the letter b, it is so common in loanwords (and used in slang) that people are familiar with it, and I don't think many people seriously say tsimpɑpʋe. K9re11 (talk) 01:11, 1 July 2015 (UTC)

I have found that even some very educated and urban Finns have an incredibly strong accent even when speaking English, so the replacement of [ʃ] with [s] in Finnish doesn't sound all that incredible since [ʃ] is, after all, not a native Finnish phoneme. Rural Finns can even be heard to say resitentti instead of presidentti (although it may be that with the advance of education in foreign languages, this is now limited to the oldest generations). Perhaps your experience is somewhat limited (/b/ is not much of a phoneme outside Helsinki or bilingual regions, I think) or you miss the fact that Finns do not necessarily actually say [ʃ] when they intend to say it and even believe they do so. It can be difficult to detect your own native-accent-related pronunciation mistakes (or those of your fellow speakers), especially when you don't have thorough phonetic training. People tend to have a lot of trouble detecting or producing contrasts which are not part of their native phonology. I believe you strongly underestimate the extent to which Finns have trouble with sounds that are alien to or marginal in spoken Finnish varieties, but imported from foreign languages. --Florian Blaschke (talk) 13:02, 12 July 2015 (UTC)

I see you are bilingual, and apparently Finland Swedish, from Uusimaa/Nyland. So it is unsurprising that you personally, and most people you know, socialise with or have met personally (many of whom Finland-Swedes), or hear in the media, have no trouble at all pronouncing [b] or [ʃ]; they are part of your native phonology, after all. But don't generalise from that biassed and unrepresentative sample to all five million Finns. --Florian Blaschke (talk) 13:11, 12 July 2015 (UTC)

Almost a decade late, but... I definitely agree with K9re11 on this. I know the bare minimum of Swedish, have literally the lowest possible education (barely got through middle school), lived my whole life in Roihuvuori, etc. so it's not like I'm some posh Swede or whatever. The idea that Finns can't pronounce /b d g ʃ/ as distinct from /p t k s/ and that we just imagine we can is honestly insulting. Like, they may not be the exact same as English or Swedish /b d g ʃ/, but they're absolutely not the same as /p t k s/ for most speakers. Even my parents generally pronounce them as distinct; my mum does sometimes accidentally pronounce /b/ as [p], but that's not the default even for her and she's almost 70.

As for what the usual pronunciations are and how they differ from English/Swedish/whatever, I'd say the biggest difference is that Finns don't labialise /ʃ/ like it is in English and German (and I think the Swedish sj-sound), but it's not even supposed to be labialised in Finnish. Of course speakers who have a retracted /s/ will have a harder time differentiating /ʃ/, but speakers who have a dental /s/ easily keep them distinct, ie. at least in Helsinki and I guess most of Uusimaa. /b d g/ are "softer" than /p t k/ but also more "tense" in a way; they don't have to be consistently voiced to be distinct from /p t k/ because there are other phonetic details, but they are usually voiced. (Ironically /b d g/ are partially devoiced in many dialects of English, too, but no one would say that implies the speakers can't pronounce those sounds. It's just that the phonetic details allow them to be partially devoiced. In English it's that /p t k/ are aspirated, in Finnish it's something else... it might be tongue root position or mild glottalisation or something else, I don't know, I just know something happens that keeps them distinct even when whispering!)

I know it might just be an Uusimaa thing or whatever, but that doesn't make it wrong. If anything, a prescriptivist stance would be that rural people who can't pronounce them as distinct are speaking incorrectly... which is officially the case, so I think it's pretty interesting that the article about Finnish phonology reflects the (mis)pronunciation of standard Finnish by some dialect-speaking people in the countryside more than the pronunciation of standard Finnish by people in the capital city.

Don't get me wrong, I don't think it should be prescriptivist (then my own speech would be "wrong" in regards to some other features, and I'm pretty sure no one speaks 100% standard Finnish), but as it is, it feels like it was written by a rural boomer who's upset that other Finns can pronounce those sounds. I mean, we might be able to pronounce them because we know English, but if the bar for how Finnish pronunciation should be described is 80-year-old 100% monolingual people in the countryside, that doesn't make sense because they wouldn't speak standard Finnish in any case but rather their respective rural dialects... and a lot of younger rural people (especially those who move to cities) are able to speak standard Finnish and/or English, and as such can more likely pronounce those sounds.

If the article had whole sections on dialects, in those sections it would make sense to say a lot of rural people can't pronounce those sounds that are in standard Finnish. But if it's standard Finnish that's being described, it doesn't, especially not as a weird statement that almost reads like it implies it's somehow better to not be able to pronounce them. I think it might be good to include if worded in a way that makes it clear it's non-standard to not distinguish the sounds.

But yeah, that's a lot of rambling with no citeable sources. That's the unfortunate thing about Finnish in general, that it's so variable and research on its features doesn't really take that into account. I mean, I think it's great that Finnish still has a lot of variation, but... VHGW (talk) 19:30, 12 February 2024 (UTC)

Help:IPA for Estonian and Finnish

Either the claim about the pronunciation [ˈherˈne.keit.to] on Help:IPA for Estonian and Finnish is nonsense or it needs to be explained in this article too. I have never noticed anything like that in speech or read anything about it.

I'm not even sure whether that article's claim about two-syllable imperatives is correct, but it does seem to be audible in some situations. Its explanation is also missing in this article.

That article has no explanation about the different sounds of h. The current claim of h always sounding like English h is of course nonsense and shows it was written without thinking at all of the needs of non-Finnish speakers.--Espoo (talk) 13:26, 11 October 2016 (UTC)

I've added the various sounds of h from Suomi et al. to the IPA help page. I don't know anything about the rest of the things you mention. — Eru·tuon 18:18, 11 October 2016 (UTC)

Thanks. I was going to quickly add info about the sound of h in kahvi being like the ch in Scottish loch and some young people's pronunciation of yech, but got stuck in the swamp of different but very similar symbols for the voiceless_velar_fricative and the voiceless uvular fricative and apparently other similar sounds. It seems i understand the differences between these sounds better now that i made those articles less confusing for normal readers, but i'm still very confused and surprised by the IPA's senseless use of different symbols that look almost identical in most browsers and completely identical in some (f.ex. Android Chrome). And i still don't know if this edit is correct. It is in any case senseless and confusing without an explanation and links to the different sounds.

Since you're a linguist, please explain the difference between the sounds of h in head and ahead because i can't really hear or feel it.

I was very impressed by these words of yours: My philosophy is to write understandably. If an article, even on a specialized subject, is impossible to understand, that's a problem. Language is communication, and if it fails to communicate, it needs to be rewritten. The current style and content of this article on Finnish phonology is completely incomprehensible and useless for almost all readers who haven't studied at least some linguistics, so maybe we could try to write a less intimidating summary of at least the article's beginning (on vowels). --Espoo (talk) 23:14, 12 October 2016 (UTC)

Stress?

"Thus, if secondary stress would normally fall on a light (CV.) syllable but this is followed by a heavy syllable (CVV. or CVC.), the secondary stress moves one syllable further ("to the right") and the preceding foot (syllable group) therefore contains three syllables."

In most current foot-based theory, feet can only contain 1 or 2 syllables; the second unstressed syllable would be unfooted. Generalizing away from feet, we could just say that there would be a lapse (two adjacent unstressed syllables). There currently isn't a source for any of the secondary stress analysis so I dunno how to edit the section without original research. Thoughts? P~M (Talk) 00:04, 18 December 2017 (UTC)

ɑj/ɑi

Perhaps aika can indeed be described by /ɑj/ as on Help:IPA/Estonian and Finnish, but then we need a second example with /ɑi/ such as airo, where /ɑj/ is at least confusing or should at least have the added explantion when it is pronounced [ɑi], i.e. where [ɑj] would be wrong. See also https://en.wiktionary.org/wiki/airo --Espoo (talk) 08:51, 17 June 2018 (UTC)

Glottal stop and gemination

They are related and need a better connection as well as sources. --Per W (talk) 19:55, 12 December 2018 (UTC)

Old post but I don't agree with the connection. (Not just me but it's disputed in many references, though I can't remember which ones I read off the top of my head.) The way I interpret this is that all Finnish vowel-initial words start with an underlying glottal stop which may or may not be realized depending on speaker/dialect, but is always realized when geminated. This way the boundary gemination has no connection with the glottal stop. E.g. in my opinion the interjection "noni" ends in a glottal stop (do others agree?) but it cannot trigger boundary gemination. Stockhausenfan (talk) 12:39, 7 June 2024 (UTC)

Flag

I don't know how to fix it, but the image at the top of this page on the mobile app is the Israeli flag, not the Finnish flag. 83.250.74.158 (talk) 12:21, 15 September 2023 (UTC)

It's vandalism, it'll be fixed once the cache expires. Nardog (talk) 18:55, 15 September 2023 (UTC)

@Nardog Thank you! 83.250.74.158 (talk) 16:42, 16 September 2023 (UTC)

Allophone of /h/

In my opinion /h/ can have the allophone [ħ] between /æ/ and a consonant. I'd say that [t̪æħt̪i] feels like the most normal pronunciation of 'tähti'. I don't have a reference though so I won't add it, but if anyone has a reference that discusses this then it would be useful to include. Stockhausenfan (talk) 12:21, 7 June 2024 (UTC)