Joonas Vakkilainen provides some very interesting information about Finnish on Quora:
Written standard Finnish is an artificial construction which is based on a mixture of dialects, not on any specific dialect. There is no prestige language that would be the norm of formal written Finnish. Because of the constructed nature of the written language, there is an organisation that gives the norms for it. The board of Finnish language (suomen kielen lautakunta) consists of specialists of Finnish language, and they ponder the norms of written Finnish and can change them. These norms are followed in formal writing, such as newspapers and scientific writing. This is called language maintenance (kielenhuolto).
The norms of written formal English are called prescriptive grammar because they are man-made rules that are prescribed to be used in official English. Even though written Finnish is more man-made and artificial than written English, its norms are not actually prescriptions. They are called suggestions: the board of Finnish language suggests how official language be used. English-speakers think that their norms are meant to be used both in writing and speech, but in Finland, the prescribed norms are just meant for one register of language. Nobody speaks according to them (except for very formal situations such as TV news or public speeches) and nobody thinks that they even should be spoken. This is why I don’t want to call them prescriptions even though they actually are that; the linguistic culture just is different from the English-speaking world.
Furthermore, the “prescriptive” rules of written Finnish are more akin to spelling in English. Because Finnish is spelled phonemically, words look different when spelled according to different dialects. That’s why there are official forms of words. In English you can’t change the spelling according to your accent but in Finnish you can. The phonemic spelling of Finnish thus enables the use of dialects in writing. In English, it is not easy to write accents because you don’t have means to do it, but in Finnish you can do it easily. When Finns write on social media, text messages or chats, they often use dialectical language. In Finnish the spelling just change the pronunciation of the particular word if you read it aloud but it of course doesn’t affect how people normally speak. This is again why we should not see the standard Finnish as a prescriptive construction similarly as the norms of formal English.
Formal written English is like the language of the upper class and educated people. That’s why English-speakers try to speak according to its rules when they want to appear sophisticated. Formal written Finnish on the contrary is thought to be a tool for an equal society. The Nordic model wants to eliminate social classes. That’s why the language is not wanted to be the language of a certain group of people but a common form that nobody speaks natively, and this kind of form of language requires an organisation that does language maintenance.
Most of that is new to me. Thanks, Bathrobe!
“Formal written Finnish on the contrary is thought to be a tool for an equal society.”
Creating a rather artificial written standard might have worked as a great leveler in the 19th century, but then broadcasting happened, and the Helsinki spoken language is very privileged compared to other Finnish regional forms.
does the same description apply to Bokmål?
Very similar to the Croatian situation. The standard form “književni jezik” (literary language) is used in formal situations and in literature. For most everyday situations, Croatians use their local dialect – the register of which may be more or less influenced by the literary standard. The literary language is based on a dialect mixture developed over centuries, and finally codified in the 19th century during the national revival spearheaded by the Illyrian movement.
I guess I am having a bit of trouble following this without some actual examples. What is the sort of thing that people commonly do in writing standard and “maintained” Finnish that none of them regardless of dialect/region/class would think of doing while speaking (other than perhaps in a limited set of formalized/artificial circumstances)? Is the word order different from any spoken variety? Are inflectional endings different and/or different cases deployed differently? Are multi-word combinations that are always contracted or elided in speech “spelled out” in full? Something else?
leoboiko: does the same description apply to Bokmål?
It applies in the same way as it does to any standard language. But not on the specifics. Bokmål grew out of the pre-independence standard language of Danish, and its gradual norvagification was not in the direction of some supra-dialectal ideal, but in the direction of the spoken language of the southeastern towns (read: Oslo). When this process went beyond the language of the upper class (and the language ideals of the socially ambitious class), it became politicized and eventually reversed. But the resulting flexible norm does allow a good deal of dialect features — but not so much in spelling as in grammar.
The description of Finnish as a supra-dialectal construction is a better fit to Nynorsk.
@JWB: I can offer one possible example, from the tiny bit of Finnish I learned many years ago. In written Finnish, “I have a book” is minulla on kirja (literally, “a book is at/by me”). But as I recall, what everyone says is mulla on kirja, or something like it. So you would never say the former, in casual speech, but you would never write the latter, in formal prose.
I think that’s the general idea. Someone who actually knows Finnish may correct me.
The description reminds me of (what I heard about) standard Belarussian, which was also supposed to be a supra-dialectal standard (though in that case, IIRC, the aim was for a single Belarussian national identity).
Unfortunately; due to the sheer prestige of (fairly closely related) Russian, by the time anyone got around to standartization, just about nobody outside of remote villages actually spoke the dialects, either.
The Belarussian case reminds me of the Norwegian. Nynorsk (in its original form) is resurrected Belarussian, although adopted at a time when Belarussian dialects were still in everyday use in most of the country, while the Danish, Dano-Norwegian or Riksmaal of the better parts of the cities is the closely related prestige language. Bokmål is what would have happened if you gradually replaced Literary-Russian-with-a-Belarussian-pronunciation with more colloquial grammar and lexicon, mainly as found in Minsk.
In written Finnish, “I have a book” is minulla on kirja (literally, “a book is at/by me”). But as I recall, what everyone says is mulla on kirja, or something like it. So you would never say the former, in casual speech, but you would never write the latter, in formal prose.
Is mulla just an abbreviated form of minulla, making this sort of equivalent to saying “gonna” but spelling it “going to”?
Is mulla just an abbreviated form of minulla, making this sort of equivalent to saying “gonna” but spelling it “going to”?
Judging by Estonian (mina:ma, minu:mu, minul:mul), it seems so.
equivalent to saying “gonna” but spelling it “going to”
I think so. The difference, I guess, is that in English we don’t get very perturbed at gross differences between how a word is written and how it’s said, where as in Finnish there is supposed to be an almost perfect correspondence. So I was a little taken aback, after I learned some ‘book Finnish,’ to discover that people don’t always say words exactly as they are written.
PS: Also, English speakers will sometimes say “going to” rather than “gonna,” whereas “mulla on …” seemed to be standard, as best I can remember, for all registers of speech.
Maybe a better example than going to v. gonna would be “should have” (written) versus should’ve or shoulda in speech, where clearly saying aloud the written form w/o contraction or elision is as far as I can tell quite unusual/marked even for “prestige” speakers. Or maybe the minulla/mulla distinction is more like writing “boatswain” but saying “bosun.” In either event it does not strike me as dramatically different from the sorts of things that are common in English, so I am still looking for an example that would substantiate the claim made in the original piece linked by hat that written Finnish is different from spoken Finnish in ways that are unlike the ways in which written English is different from spoken English. (That Finnish orthography might not in practice be quite so perfectly ahistorical, transparent, and non-arbitrary as its supporters’ propaganda claims seems to me a separate point.)
I think “should’ve” is usually bisyllabic with a schwa in the second syllable. So isn’t that the most natural way to pronounce “should have” with unstressed “have”, at least when assuming that 1. unstressed TRAP reduces to a schwa as well, 2. /h/ disappears from unstressed syllable onsets when following a consonant? I see “should’ve” as more of a result of phonological rules than a separate lexical entry.
Others, like “wanna”, seem more tricky. I think it works in dialects that have a winner-winter merger, and where the vowel in unstressed “to” reduces to the same as final “-a”, but that’s obviously not all dialects.
As for Finnish, the post seems to implicitly assume that all dialects have more or less the same phonology, which can be expressed by the standard orthography, and I wonder if that’s really true. For example, standard Japanese and the Kansai dialect differ largely in grammar and to some extent in vocabulary, and those differences are easy to express in writing. But there are also phonological differences (like in pitch accent and vowel devoicing) that cannot be easily expressed. Other Japanese dialects vary a lot more, e.g. they might not have pitch accent at all, or they might have more phonemes that cannot be written.
As for English, I guess the poster’s point is that the biggest differences are in phonology (by which I mean the realization of phonemes), so eye dialects often only make sense for speakers of some dialects, and often not for speakers of the dialect they’re supposed to imitate.
@dainichi: I actually dislike the spelling should’ve, because in my view /ǝv/ is already the default value of have when spoken. I’d reserve ‘ve for cases that are truly non-syllabic, like I’ve.
J.W.B.: I think this information (from Wikipedia s.v. Peräpohjola dialects) should cast some light on the subject:
Peräpohjola dialects are forms of Finnish language spoken in Lapland in Finland, Sweden and Norway. The dialect group belongs to the Western Finnish dialects and it is divided to five more specific dialect groups.
Like the Northern Ostrobothnian (Oulu) dialects, Peräpohjola dialects are Western dialects that show features from Eastern dialects. For instance the reflex to standard -ts- in metsä is mettä according to a Western pattern, whereas the reflex of standard -d- [which is the only voiced stop, and was a fricative until recently] is deletion, ‘j’, or ‘v’, as in Eastern dialects, e.g. vedän – vejän. Epenthetic vowels (tyhjä – tyhyjä) are not common although found in southern Tornio dialects. There are no vowel or diphthong changes, a Western/standard feature, but there is a general gemination of consonants in short initial syllables (e.g. standard makaa is makkaa), as in Oulu and Eastern dialects. Palatalization is absent, which is a very Western feature, since palatalization occurs even in the (Western) Oulu dialect.
A unique feature is that unstressed syllables in e.g. the illative case are preceded with an emphatic ‘h’, e.g. talhon vs. standard taloon, menhään vs. standard mennään. This is highly distinctive but very difficult for outsiders to imitate correctly. The pronouns are also distinct: mie (minä, “I”) and sie (sinä, “you”(sg.)) are Eastern-like, while met and meän (me and meidän, “we” and “our(s)”) and tet and teän (te and teidän, you (pl.) and your(s)(pl.)) are unique to Peräpohjola. Also, the reflex to the standard third-person verb suffix -vat (“they”) is a simple -t, e.g. annoit vs antoivat.
Here’s a head-to-head comparison of Meänkieli (a Finnish variety that is a national minority language of Sweden) with standard Finnish, line by line, Meänkieli first, again from Wikipedia:
Ruotti oon demokratia. Sana demokratia
Ruotsi on demokratia. Sana demokratia
tarkottaa kansanvaltaa. Se merkittee
tarkoittaa kansanvaltaa. Se merkitsee,
ette ihmiset Ruottissa saavat olla matkassa
että ihmiset Ruotsissa saavat olla mukana
päättämässä miten Ruottia pittää johtaa.
päättämässä, miten Ruotsia pitää johtaa.
Meän perustuslaissa sanothaan ette kaikki
Meidän perustuslaissamme sanotaan, että kaikki
valta Ruottissa lähtee ihmisistä ja ette
valta Ruotsissa lähtee ihmisistä ja että
valtiopäivät oon kansan tärkein eustaja.
valtiopäivät on kansan tärkein edustaja.
Joka neljäs vuosi kansa valittee kukka
Joka neljäs vuosi kansa valitsee, ketkä
heitä eustavat valtiopäivilä, maakäräjillä
heitä edustavat valtiopäivillä, maakäräjillä
ja kunnissa.
ja kunnissa.
Unsurprisingly, Meänkieli has more Swedish loanwords than standard Finnish, but none happen to be on display here.
I actually dislike the spelling should’ve because in my view /ǝv/ is already the default value of have when spoken.
That’s why I prefer should of. 🙂
There are some dialectal differences in syntax as well:
tulla tehdyksi vs tulla tehtyä
In the pattern V1 V2, some V2 verbs take the 1st infinitive form, and others, the 3rd infinitive illative form, eg, alkaa sataa (1st infinitive) vs rupeaa satamaan “it starts to rain/raining” (Infinitives). Which form a particular verb takes may differ in various dialects.
Ruotti/Ruotsi o[o]n demokratia
Two things of note:
The Finns still call the Swedes the Rus’, as their ancestors did a thousand years ago (cf. Finnish venäläinen ‘Russian’, ultimately from Wend).
Minority languages serve very well to present nationalist propaganda.
I appreciate juha’s example of intradialect syntax differences, but I’m still thinking that the original post led me to believe that the standard written language has syntactic features (or at least some non-pronunciation features, not least because it is rather incoherent to talk about the “pronunciation” of the written-only form of the language) not shared with any spoken dialect, thus creating an artificial “level playing field” in preference to giving the native speakers of some prestige dialect a leg up. Can anyone point me to such features? Or did I misunderstand the import of the original post?
The standard form “književni jezik” (literary language) is used in formal situations and in literature. For most everyday situations, Croatians use their local dialect – the register of which may be more or less influenced by the literary standard. The literary language is based on a dialect mixture developed over centuries, and finally codified in the 19th century during the national revival spearheaded by the Illyrian movement.
Disagree.
The standard is used by very, very few people outside of Croatian studies (kroatistika) lectures, in any context. In formal situations taking place in their native area, people will just speak as they always do. Outside of their native areas, Shtokavians will usually adjust the way they speak just a bit (e.g. avoid words they’re aware might not be understood, if they know the standard equivalent). As far as yat goes, ikavians in non-ikavian regions will usually switch to a mostly ijekavian speech in really formal situations, though you’ll often hear people complaining that some (usually Dalmatians) don’t. Chakavian and Kajkavian speakers will, in formal situations outside of their native areas, speak the (Shtokavian-based) urban dialect of the big urban centre their native area gravitates to, usually with some telltale non-Shtokavian features.
No one, or close to no one, will use the standard. Especially the standard accent, which most wouldn’t even be able to pull off, and many are confused about or not very familiar with.
In lierature, many varieties are used. Authors will use the standard (for the entire work or just outside of dialogue) if they’re aiming for a neutral style. Chakavian and Kajkavian usually only appear in poetry, or sometimes as bits of dialogue (usually watered down for intelligibility).
The literary language is based on a dialect mixture developed over centuries, and finally codified in the 19th century during the national revival spearheaded by the Illyrian movement.
That’s the official line, but it doesn’t really reflect reality. It’s Shtokavian of the Eastern Herzegovinian variety all the way.
@gwenllian – what dialect are Croatian tv and radio personalities speaking?
More links:
https://en.wikipedia.org/wiki/Colloquial_Finnish
https://en.wikipedia.org/wiki/Savonian_dialects
News anchors and the main reporters aim for the standard, with various degrees of success and a lot of funny sounding overcorrection. Local correspondents are all over the place. I think they’re probably told to aim for the standard but don’t get the coaching or attention the former do, so they wing it or just get away with not trying much.
In less formal TV content it varies. Depends on the channel and the target audience. If it’s entertainment content they often speak naturally (again, for non-Shtokavians in such cases that means using their go-to urban Shtokavian-based variety, not their native non-Shtokavian one) or adjust their speech to some sort of intermediate variety. There’s been some controversy over different perceptions of whether some regions and dialects are overrepresented on air. It’s usually a bit of a Dalmatia vs Zagreb pissing contest, while other regions, which really could be said to be underrepresented, just shrug.
The public broadcaster’s 3 national radio stations always use the standard, I think. I don’t listen to them much, so I’m not totally sure. Other stations tend not to bother.
Much like in Ireland, then, except that Finnish and Croatian newscasters can actually pronounce the standard well. It’s not actually incoherent, pace JWB, to talk about a written-only standard being pronounced: that needs to be done when written text is read out loud in all diglossic languages. The “reading traditions” of classical languages are essentially the same thing.
As for Finnish, the post seems to implicitly assume that all dialects have more or less the same phonology, which can be expressed by the standard orthography, and I wonder if that’s really true.
Close enough. Most dialectal variation in phonology involves primary splits such as variation in length, vowel quality or voicing; diphthongizations, monophthongizations, metatheses, epentheses or elisions. All of these are immediately representable by conventions of the standard language: e.g. ‘coffee’ may vary as kahvi (standard) ~ kahve ~ kahavi ~ kahave ~ kaffi ~ kaffe.
Final palatalization in Savonian dialects is not captured this way, but has its own widely recognized orthographic device: -Cj (e.g. vesj /vesʲ/ for standard vesi ‘water’). I think the only thing going consistently unwritten might be /ɾ/ in some western varieties, which gets rendered simply as r, the same as /r/.
In a sense this is not an accident as much as by definition: some Northern Finnic varieties that show more exotic phonology (e.g. Ingrian with a three-way length distinction and reduced vowels) tend to be excluded from the definition of “Finnish” without any especially well-defined linguistic reason.
the original post led me to believe that the standard written language has syntactic features (or at least some non-pronunciation features …) not shared with any spoken dialect
One component of language that is not pronunciation is of course lexicon, and it’s here that much playfield levelling has taken place, e.g. through the coining of native equivalents for learned vocabulary. Many have survived in use (sähkö ‘electricity’, muovi ‘plastic’, tutka ‘radar’, happi ‘oxygen’, kännykkä ‘mobile phone’), many others have not (sähkäle ‘electron’, sätiö ‘radio’, nimentö ‘nominative’, tytinä ‘aladobe’, hyrysysy ‘car’).
I cannot think OTTOMH any syntactic or morphological features created totally ex nihilo. But it may be worth noting that a few grammatical categories of standard Finnish — the potential mood, possessive suffixes — are receding or nonexistent in the colloquial speech of Helsinki, but can still find regular uses in rural dialects.
So I was a little taken aback, after I learned some ‘book Finnish,’ to discover that people don’t always say words exactly as they are written.
Same here, and the differences can be fairly major: a simple sentence like “I’m coming with you”, which I had learned as Minä tulen sinun kanssasi, is actually pronounced Mä tuun sun kaa (in Helsinki Finnish, I think). The differences in this case are all phonological except for the final -si of kanssasi, which is the possessive suffix which j. says above is receding.
I remember being told that the consonant written d is a voiced stop only in the standard language (and there it seems to be something like a retroflex).
I think the only thing going consistently unwritten might be /ɾ/ in some western varieties
There is some unwritten gemination in the standard, mostly word-initially after words formerly ending in a glottal stop, like imperatives (tule t[:]änne “come here”), but also medially in a few cases. I don’t know how this is reflected in spoken varieties.
Minä tulen sinun kanssasi, is actually pronounced Mä tuun sun kaa (in Helsinki Finnish, I think)
Estonian has grammaticalized this (or more precisely its own cognate postposition) to the comitative sinuga.
For German it’s so far from true that I’m sure this is a major part of why the dialects aren’t written more often.
Interesting. Standard German and many dialects literally use “to rain” there (zu regnen, infinitive), but my dialect avoids this construction completely and goes for *zum Regnen, “to the raining”, with “the” in the dative – the closest thing to an illative we can muster.
Ringe’s 2006 book says that something about infinitives and datives is a common Germanic feature mostly lost in the modern languages, but no details or examples are given, so I don’t know if that’s what was meant.
I’m surprised to learn that the official line tries to deny this. My understanding was that the Illyrian movement explicitly agreed on Eastern Herzegovinian because it had the coolest hero songs, except for the (AFAIK somewhat incomplete) restoration of /x/ which has apparently been mostly lost there. I think a translation of the passage in question is on the English Wikipedia somewhere (I’m way too tired to look for it now).
Ringe’s 2006 book says that something about infinitives and datives is a common Germanic feature mostly lost in the modern languages, but no details or examples are given, so I don’t know if that’s what was meant.
Oh, yeah… I haven’t read Ringe’s book, but now I seem to remember something about the dative of the infinitive serving as a gerundive. I think the pattern was said to be lost as an active process, but not without leaving a trail of adjectives in its wake. Providing I remember correctly, could that be it?
gwenllian says:
June 2, 2017 at 1:25 pm
News anchors and the main reporters aim for the standard, with various degrees of success and a lot of funny sounding overcorrection.
That’s broadly right. The national broadcaster Hrvatska Radiotelevizija retains an orthoepy unit and I remember seeing a documentary last year where they said that all TV reporters take pronunciation lessons. They still have editors (lektori) whose job it is to maintain the standard language in broadcasting.
The standard is still taught and reinforced in schools. It is used in interactions between a teacher and students.
David Marjanović says:
My understanding was that the Illyrian movement explicitly agreed on Eastern Herzegovinian because it had the coolest hero songs, except for the (AFAIK somewhat incomplete) restoration of /x/ which has apparently been mostly lost there.
That’s right in a way. The Illyrians were enamoured of Dubrovnik literature, which is witten in the Dubrovnik Ijekavian dialect (it always retained the /x/), and the Illyrians based their language norms on that. However, the Croatian standard is not the same thing as Eastern Hercegovinian (which is really another name for the ijekavian dialect with the so-called new accentuation): eg. the standard never accepted the “newest jotation” such as the use of “đe” for “gdje” (where), or “ćerati” for “tjerati” (to chase). The 19th & early-20th century standard also retained the separate case endings for dative, locative and instrumental plural nouns and adjectives (syncretised in E. Herc. and many other štokavian dialects). This was abadoned in the early 20th century, but since 1990, some signs of differentiating the cases are coming back, at least when it comes to adjectives. The differentiation was always present in the kajkavian and čakavian dialects, and some štokavian dialects. Many lexical items in the standard are not found in the E. Herc dialect: eg. “uporaba” (use) is originally čakavian, “kukac” (insect) is originally kajkavian, “kiselina” (acid) is an Illyrian creation… The wide range of lexical inputs into the standard dates from the centuries-old pre-Illyrian period of standardisation. So while the E. Herc. dialect may come close to the standard, it is not the same thing as the standard language. Any native speaker of Croatian can easily differentiate the two.
An ijekavian dialect was used by Vuk Karadžić in his attempted codification of Serbian around the same time as the Illyrian movement was taking place. However, his concept of a Serbian language was never adopted – Serbs by and large speak an ekavian dialect, which is what the Serbian standard is based on. Indeed, it was only after Karadžić’s death that his version of the cyrillic alphabet for Serbian was adopted in Serbia itself. He was successful and influential in two areas: (1) the notion that Serbs should abandon the literary language they used at the time, which was full of Russian and Church Slavic words, and use a pure “folk” (narodni) speech. Of course, for Karadžić, the purest folk speech was spoken in his native region; and (2) that all štokavian speakers, including Croatians, Bosniaks and Macedonians are really Serbs. This notion was adopted by early pioneers of Slavic studies, eg. Dobrovsky and Miklosich, but quickly became discredited as patently false. It is still trotted out by Serb nationalists, including some Serb linguists, from time to time.
Possible; no idea. I should look it up.
Fascinating.
Yes, sandhi is consistently mostly unwritten. (Most spoken varieties extend this to realizing final n before non-stops, some also t before stops, as gemination.) This is one thing about Finnish orthography that gets consistently prescribed: writing out sandhi in quoting someone is a common device for indicating or alleging that they are a simpleton.
Apical alveolar [d̺̳], in contrast to t being lamino-dental [t̻̪]. b d gcan also have some degree of implosion (in my idiolect at least). I wonder if some varieties will eventually settle on a new but nonstandard means of distinguishing t d
As an aside, for anyone interested in phonological or morphological variation within Finnish, Lauri Kettunen’s dialect atlas from 1940 can be found online.
sandhi is consistently mostly unwritten
What about unwritten medial gemination, as in sydämen? I’ve never understood why that isn’t spelled sydämmen, and I don’t know if it’s an isolated exception or part of a larger pattern.
Sydämen is a singular irregularity: the spelling is, I think, by the example of the more common inflection type -n : -me- (eläin : eläimen ‘animal’, avain : avaimen ‘key’, ydin : ytimen ‘core, marrow’, etc.), while ‘heart’ is the only word with /-n/ : /-mme-/. (The reason this unique inflection exists at all is poorly understood, though it’s attested all across Finnic.)
On the other hand, we do write out the contrast between -n : -mA- in caritive adjectives (koditon : kodittoman ‘homeless’) versus -n : -mmA- in superlatives (komein : komeimman ‘the most handsome’), so yes, the exception status for sydämen is not very well motivated.
FWIW two other similar cases of unexpected consonant lenght that are commonly met in colloquial Finnish are piirustaa /piirrustaa/ ‘to draw (by pencil etc.)’ and puolustaa /puollustaa/ ‘to defend’. These are explained, I think, by analogy with related verbs: piirtää : piirrän ‘to draw’, puoltaa : puollan ‘to side with’. (Formally the first two verbs are instead derived directly from the nouns piiru ‘mark, scratch’; puoli ‘side’.)
Un- or less assimilated loanwords also tend to have a lot of unwritten length going on, typically in originally stressed vowels (ameba /ameeba/) or in voiceless consonants (golfata /golffata/ ‘to golf’), possibly both (pyton /pyyton ~ pyytton/). But in firmly established loans this gets usually sorted out eventually (moottori, ooppera, politiikka ‘politics’, poliittinen ‘political’).
I’m surprised to learn that the official line tries to deny this. My understanding was that the Illyrian movement explicitly agreed on Eastern Herzegovinian because it had the coolest hero songs, except for the (AFAIK somewhat incomplete) restoration of /x/ which has apparently been mostly lost there. I think a translation of the passage in question is on the English Wikipedia somewhere (I’m way too tired to look for it now).
Well, the practical reason for choosing it, at least for Croatia, was simply South Slavic unity. The hero songs were then praised and taught, to the exclusion or minimization.of the importance of things like, for example, historical Chakavian poetry. Since in Croatia the historically disproportionate emphasis on hero songs is viewed as part and parcel of Serbian dominance in the whole South Slavic unity idea, bringing up either that unity or hero songs to big up the standard wouldn’t exactly make the masses feel warm and fuzzy about it the way the official line (which emphasizes Croatian unity and overemphasizes the harmony between different Croatian traditions, regions and dialects) does. The rich tradition and literature of Dubrovnik does get brought up as an explanation of why Shtokavian was chosen, and specially as an excuse as to why an ijekavian Shtokavian was chosen, when Croatian ijekavians were something like 70% ikavian. But Dubrovnik wasn’t the actual reason for the choice, and the standard isn’t that close to Dubrovnik speech anyway.
The standard is still taught and reinforced in schools. The standard is still taught and reinforced in schools. It is used in interactions between a teacher and students.
It’s taught, but it isn’t really reinforced, and while, depending on the teacher, some adjustments might be expected in class (e.g. many will “correct” ikavian speech to ijekavian), nobody’s gonna insist on the standard. Things were certainly different in the past, though. The post-WW2 period of change and modernization was especially rough. That approach was dropped a long time ago.
In Chakavian and Kajkavian areas Shtokavian will be used or an intermediate variant, though not all teachers will insist on students changing things much. Croatian classes, of course, will emphasize the standard (or, again, the go-to urban Shtokavian-based variety, which is locally thought of as pretty much the standard), but (unlike in the past), students aren’t expected to try to imitate the standard accentuation (and many teachers of Croatian don’t either), or hassled about sounds they find tricky to pronounce.
For Shtokavian students, other than the ijekavianism (which, again, many won’t insist on), the whole thing is a non-issue. The textbooks are written in the standard, and students are expected to learn how to differentiate č and ć and je and ije in Croatian class (in writing, there’s no expectation to differentiate them in speech). That’s pretty much it.
However, the Croatian standard is not the same thing as Eastern Hercegovinian
The differences are minor, and many of them exist in Serbian as well. Croatia’ official line here is not based on facts, it’s just trying to paper over what is perceived as less than ideal about the actual history and reality of our standard.
So while the E. Herc. dialect may come close to the standard, it is not the same thing as the standard language. Any native speaker of Croatian can easily differentiate the two.
Again, same goes for Serbia, yet you won’t hear them denying Eastern Herzegovinian as the basis.
An ijekavian dialect was used by Vuk Karadžić in his attempted codification of Serbian around the same time as the Illyrian movement was taking place. However, his concept of a Serbian language was never adopted – Serbs by and large speak an ekavian dialect, which is what the Serbian standard is based on.
The Serbian standard, unlike the Croatian, allows two realizations of yat. In recent times there’s been an uptick in the use of ekavian in writing among Serbs in Bosnia, Croatia and Montenegro, stemming from a bit of a perception of it as a better, more clearly Serbian option. But, again, ijekavian is not in conflict with the Serbian standard.
overemphasizes the harmony between different Croatian traditions, regions and dialects
Sorry, should’ve made clear that I was using the word dialects here for Croatian narječja, the word used for Shtokavian, Chakavian and Kajkavian (although they should more accurately described as languages), which then further divide into many dijalekti (although the differences between those could again be described as different lanuages, except in the case of dialects of Shtokavian, a relatively uniform so-called narječje compared to the others).
Well, I’m no expert on this but wikipedia claims that there was a Štokavian expansion into Croatian areas from the east due to Ottoman-era migrations so it looks like there is a grain of truth to this (not that all of those people still identify as Serbs but that a significant part of Croats may stem from Serb migrants retaining their dialect but not religion; “original” Croatian Štokavian was only/almost only in Slavonia according to this). However, it’s only wikipedia. Comrie & Corbett write: “Štokavian, the most widespread group, covers Serbia, Bosnia-Hercegovina, Montenegro and much of Croatia. Since the 1400s refugees from Turkish rule have carried it north and west, into earlier Kajkavian and Čakavian territory. The pre-migration landscape presumably had smooth transitions to Čakavian and Kajkavian, but the present map shows abrupt boundaries and islands of older phenomena amid larger homogenous areas.”
Also, Eastern Hercegovinian is apparently Eastern Hercegovinian only in name and historical origin as it covers large swaths around the north-western Bosnian nook prodding into Croatia, too. So the sheer extent of it was likely not the smallest of the reasons that led to its adoption as the basis for the literary standard. C&C: “Eastern Hercegovinian, (i)jekavian with new accents and neutralized plurals, is the most widespread type, carried far from home by migrations. As Karadžić’s native dialect, it formed the basis for standard Serbo-Croat.”
There’s no need for speculation, these weren’t prehistoric migrations, and the historical record is pretty clear.
not that all of those people still identify as Serbs but that a significant part of Croats may stem from Serb migrants
The ancestors of today’s Shtokavian Croats who took part in those expansions of Shtokavian by and large never identified as Serbs. Some identified as Croat (in whatever meaning the word had to those people at that time), many didn’t. The modern identities of all South Slavic nations are a much more recent thing.
retaining their dialect but not religion
The ancestors of today’s Shtokavian Croats who took part in those expansions were, by and large, Catholic. The ancestors of today’s Croatian Serbs who took part in those expansions were, by and large, Orthodox. Conversion from Orthodoxy to Catholicism or vice versa was pretty rare in the Western Balkans.
So the sheer extent of it was likely not the smallest of the reasons that led to its adoption as the basis for the literary standard.
Its extent in Croatia was limited compared to the other countries where it was spoken. Even in a united Yugoslavia, Eastern Herzegovinian is a bit of an odd choice for Croatia… That is, if a Croatian standard separate and different from the Serbian one was to co-exist with South Slavic political unity (so no total and practical language unification anyway), Younger Ikavian would’ve been the more obvious choice for it.
Still, the choice of E. Herzegovinian does make more sense within the context of a Yugoslavia than without. But, like I said, that explanation isn’t really a crowd pleaser in Croatia. Also, traditionally few Croats spoke Eastern Herzegovinian, whereas pretty much all Croatian Serbs, most of whom wouldn’t describe themselves as speaking Croatian or participating in the Croatian language community, always have. So when the choice of the standard is discussed, usually Dubrovnik is brought up as an (not very convincing) explanation for it, and contributions from dialects other than Eastern Herzegovinian exaggerated greatly.
Not so much farther east, apparently: my name seems to be equally common on both sides, and the -rj- for mainstream -rij- should indicate it didn’t originate in many different places (as frozen patronymics are of course expected to do).
Not so much farther east, apparently: my name seems to be equally common on both sides, and the -rj- for mainstream -rij- should indicate it didn’t originate in many different places (as frozen patronymics are of course expected to do).
Well, it was far from non-existent, just rare enough that it wasn’t a phenomenon of demographic significance. People, usually individuals or individual families, would “cross over” occasionally or marry into the other community bringing the surname with them, as you’d expect with two denominations speaking the same language living so close to each other for such a long time, but it was basically a drop in the ocean of the masses (I’m leaving out the region’s other religion, since the circumstances were completely different and consequently conversions from both Christian denominations to Islam and back again were obviously very much significant demographically).
farther east
I was actually thinking of the whole ex-Yu territory when I used the term. And I also definitely used it unwisely, as I didn’t even think about the fact that it also includes Albania, a country I don’t really know a whole lot about, especially when it comes to its history.
the -rj- for mainstream -rij- should indicate it didn’t originate in many different places (as frozen patronymics are of course expected to do)
Way out of my depth here, but would -rij- really be expected here? Marjan isn’t that rare of a name, especially among Serbs. For what it’s worth, Marjanović (by far most popular in Serbia, then Bosnia, then Croatia) is a much more popular name than Marijanović (pretty rare in Serbia, somewhat more popular in Bosnia, and most popular in Croatia – but still less popular even here than Marjanović is.)
Days later I’m surprised nobody challenged the idea that the English word have is generally pronounced /ǝv/ ! I hear that H in there in my own and many people’s speech. I even feel the little puff of air! There’s definitely a distinction between should’ve and should have.
That’s what I was told. I may be wrong.
At least in the US, many barely literate people write “should of”, “would of”, and the like for the pronunciation with /ǝv/.
@Lucy Kemnitzer: I’m not saying that AmEng have never takes /h/, just that I think /ǝv/ is the form that shows up most commonly in real speech (as an auxiliary, that is). In non-initial position and preceding another verb, as in “should have gone”, I can’t imagine any context where I would plausibly pronounce the h.
What bugs me about the spelling “should’ve” for [ʃʊɾəv] is that none of its users, as far as I know, follow their own logic and write something like [ˈɛml̩i ən ˈdʒoʊ əv əˌɹaɪvd] as “Emily and Joe’ve arrived”.
That includes half of the Internet. I’ve often seen people who write a fair amount use it; and Google finds 10,700,000 hits for “should of”.
English spelling doesn’t give a lot of alternatives to write the cline should have gone ~ should’ve gone ~ shoulda gone. Also, the cline doesn’t manifest in all positions. For “Emily and Joe have arrived”, there is possibly only Emily and Joe have arrived ~ Emily ‘n Joe’ve arrived. The way that these things are represented in normal orthography, it’s hard to represent them in a fine-grained sort of way. Perhaps Finnish is better….
It appears where gwenllian and I agree is in the importance of the ijekavian dialect (called “Eastern Hercegovian” by some linguists, also called the “southern” dialect by Vuk Karadžić). The followers of Vuk Karadžić appear to suggest that there was nothing before he based his grammar on his own dialect. This is somewhat true, but only if viewed from the Serbian language perspective. Before Karadžić’s time, Serbs used Church Slavic or a Russified version of Church Slavic as their literary language. Karadžić’s grammar made a clear break with the past. I have seen a number of English language anthological publications on Slavic languages that seem to adopt this view, especially where there is a chapter on the historical development of “Serbocroatian.” Such publications seem to conflate Serbian and Croatian and then disregard the separate development of Croatian.
From the Croatian point of view there is a long continuity and evolution of the use of štokavian as a literary language. The first štokavian grammar by Bartol Kašić was published in 1604. Since then there has been a long line of continuity in the evolution and standardisation of the literary language based on the štokavian dialect. Most grammars published before Della Bella’s (1728) and Voltiggi’s (1803) were based on both the ikavian and ijekavian dialects. Della Bella and Voltiggi were from Dubrovnik, so the basis of their grammars was ijekavian Dubrovnik dialect, as did Stulli in his 1801 dictionary. These were influential publications, and found favour with the Illyrian movement. The continuity and evolution of grammatical norms is evident throughout the 19th century. The Illyrian movement was both political and cultural in nature. The Illyrians strove for political unity of all southern Slavs, including the Slovenes, Serbs, Macedonians and Bulgarians. To make their message widely accessible they consciously chose the štokavian ijekavian as being the most widespread dialect, and as having the benefit of an established literature – namely that of Dubrovnik. Just about every issue of the Illyrian flagship publication – the “Danica ilirska” literary magazine – contains some extract from the Dubrovnik poets or writers.
It appears where gwenllian and I differ is whether the ijekavian dialect is the same as the Croatian literary language. Ijekavian does come close to the standard, but it is not the same. In support of the view that it is not the same, here are some examples of phonological differences:
(a) Jekavian has soft consonants ś and ź as a result of the “newest jotation” while the standard has sj and zj. As an aside, standard Croatian lacks ś and ź, while standard Montenegrin contains these consonants.
(b) Initial consonant group zr- is realised as zdr-, eg. zdrak vs zrak “air”.
(c) Word final fricative+plosive combination is realised without a plosive, eg. groz vs grozd “bunch of grapes”.
(d) The combination hv is realised as f: eg. falim vs hvalim “I thank”.
(e) Sometimes, p is realised as f, eg. klufko vs klupko “ball of wool”.
(f) /x/ has disappeared and is sometimes realised as /k/ or /g/ eg. trbuk vs trbuh “stomach” and moig vs mojih “my, mine.”
Source: Pavle Ivić (chief editor) “Fonološki opisi srpskohrvatski/hrvatskosrpskih slovenačkih i makedonskih govora obuhvaćenih opšteslovenski lingvističkim atlasom”, [Phonological descriptions of Serbocroatian/Croatoserbian, Slovene and Macedonian dialects covered by the Panslavic Linguistic Atlas] Sarajevo 1981. I have used the chapter on the ijekavian dialect of Jasenik, in eastern Hercegovina, and have mentioned only those peculiarities which make it different from the neighbouring ikavian dialects of Guber, near Livno in Bosnia and Otok near Sinj in Dalmatia.
There are of course numerous grammatical and lexical differences between ijekavian and the standard too.
What an informative comment! I’m learning a lot from this learned (and friendly) disagreement. (I checked my ancient copy of R. G. A. de Bray’s Guide to the Slavonic Languages—so old he spells it “Croätian”—and sure enough, he takes the “it all started with Vuk Karadžić” line.)
If so, then I stand corrected.
What about the Croats in Bosnia? Are they a remnant of the local pre-Islamic Christian population who started to identify as Croats due to their denomination? Are they, historically, the source of the Catholic Štokavian migration into Croatia?
Oh! Question: is Standard Montenegrin in actual use now? It was only developed a few years ago.
(The proposal I saw, also a few years ago, didn’t contain those consonants as far as I remember. But it did have a lot of dz in very odd places; apparently /z/ has turned into [dz] in a lot of environments, even between vowels…???)
Looking at the various sound differences among the different varieties that fall within the Croatian/Bosnian/Serbian complex, I can’t help but wonder how much impact this has on mutual intelligibility. Just looking at some of the examples (including on the Internet), the impression I get is that many of these differences are minimal and systematic, and don’t pose a great problem for intelligibility. What is the situation on the ground?
They are about as mutually intelligible as British and American English: perhaps 100 differences of vocabulary and a very few systematic distinctions of spelling, which do actually affect pronunciation, unlike say honour/honor. I’m speaking of reading other groups’ written standards here.
David Marjanović says:
Oh! Question: is Standard Montenegrin in actual use now? It was only developed a few years ago.
It’s been a while since I visited Montenegro now. A Montenegrin news portal, http://www.vijesti.me/ does not appear to use ś, ź and dz. eg. it has “sjutra” as opposed to “śutra” (“tomorrow”). However, as I recall it, the Montenegrin official orthography that introduced the new letters ś, ź and dz in to the alphabet made them optional rather than obligatory. That is, the Montenegrin standard allows both “sjutra” as well as “śutra”. The President’s website has the spelling “predsjednik” rather than “predśednik”. The Parliament website uses both “predsjednik” and “predśednik” (“president”).
My guess on the use of the new spelling is that it all depends on how far the Montengrins persist with it. If the new spelling is being taught in schools now it will take a few years for it to spread throughout society into general use.
If we were to disregard politics, Serbs, Bosnians and majority of Croats speak the same language which used to be called Serbo-Croatian.
Some Croats also speak two other closely related smaller languages which are distinct from Serbo-Croatian.
If we were to disregard politics, some Pakistanis and some Indians speak the same language that used to be called Hindustani.
However, it is not called Hindustani any more. The Pakistani standard is called Urdu. The Indian standard is called Hindi.
It is difficult to see the utility in using obsolete or obsolescent names for languages. Especially where the vast majority of the native speakers do not use those names.
@zyxt: Are either of those terms (fully) obsolescent, though? I have the impression that “Hindustani”, at least, is still current when referring to the common speech of north-central Indians and Muhajir Pakistanis (the kind used in Bollywood films – as opposed to the sanskritizing or persianizing national standards), though I could be wrong.
Then it’s better than the situation with Mongolian, where pronunciation differences (including stress, cadence, and delivery) and vocabulary differences can be so large that people from the east of Inner Mongolia cannot be understood at all by people in Ulaanbaatar. Part of this is due to lack of familiarity (lack of communication between the two areas), but even just taking pronunciation differences you would be hard put seeing them as the same language, were it not for the strained consensus that they all speak something called ‘Mongolian’. (Strained because people in Mongolia don’t usually include Inner Mongolian dialects in their world of ‘Mongolian’.) Since the traditional script is still only gaining ground very slowly in Mongolia, even the bond of a written language is tenuous.
The situation with “Serbo-Croation” sounds like pure sectarianism by comparison. However, I’m still curious about mutual intelligibility with the ‘minor languages’ (which I assume are Kajkavian and Čakavian).
Let’s call it “Former Yugoslav Serbo-Croatian language” or FYSCL.
Re: eastern Inner Mongolian
Lack of comprehension is one-sided. Khorchin Mongols understand Khalkha Mongolian just fine (since it differs very little from Inner Mongolian standard which is taught in schools and broadcast on TV)
That’s precisely my point. If the speakers of Kajkavian and Čakavian didn’t learn the Štokavian-based standard, what would intelligibility be like?
Bathrobe says: However, I’m still curious about mutual intelligibility with the ‘minor languages’ (which I assume are Kajkavian and Čakavian).
You are absolutely on the money when you talk about a lack of familiarity. When Yugoslavia was one country, there was much more contact between the different peoples and the languages they spoke. Since the independence of its constituent states, there is less familiarity. Mutual intelligibility has suffered as a result. In Croatia for example, the younger generations are not familiar with Cyrillic now. Even though I grew up in Yugoslavia, I admit to struggling when faced with Serbian words like namćor, ugursuz, aratos, budžaklija, gedžo etc. or Bosnian words like mešihat. 25 years after independence there are now publications like Marko Samardžija’s “Srpsko-hrvatski objasnidbeni rječnik / Српско – хрватски експликативни речник” [Serbian-Croatian Explanatory Dictionary] to explain some of these differences to Croatians who might have an interest in Serbia.
With respect to the mutual intelligibility with the ‘minor languages’: The minor “languages” would include kajkavian and čakavian dialects, as well as the Burgenland Croatian standard language and the Molise Croatian dialect. In the history of developing the Croatian standard, grammarians and lexicographers would freely use and incorporate elements from štokavian, čakavian and kajkavian (eg. dictionaries of Vrančić (1595), Habdelić (1670), and Belostenec (1740), which suggests that the educated Croatian speakers were at ease with all 3 dialects in that period. From the 18th century, štokavian begins its domination over the other two dialects (eg. Della Bella’s grammar and dictionary of 1728). However, in Croatia, the standard was always open to the influences of all three – štokavian, čakavian and kajkavian. This is understandable because the main cultural centre – Zagreb – is in a kajkavian speaking area. Contrast this to the Serbian experience, where all the Serbs are štokavians, and have no familiarity with, or use for, the čakavian and kajkavian dialects.
A lot has been written about Croatian “purism” in language. The Croatian standard has been “puristic” in trying to substitute Slavic words for internationalisms eg. “odbor” for “komitet” (committee) – and this has been going on for centuries. However, Croatian has been relatively open in admitting čakavian and kajkavian words into the standard, as well as words from other Slavic languages – Czech, for example, had an important role in the formation of the standard – eg. “plin” (gas) and “vodik” (hydrogen). Serbian, on the other hand, has also had “puristic” tendencies in the sense that it rejects čakavian and kajkavian forms – understandably so, since Serbs don’t speak these dialects. In forming the Sebian standard, Serbian language purism was evident in the rejection of all words except those that are “spoken by the people.” The result of this is that the Serbian standard is more accepting of non-Slavic words, eg. “špilzohne” from German for a type of stocking, “ćebe” (blanket) from Turkish.
The familiarity of individual speakers with the non-standard dialects depends on their individual background. I would guess that a person from Zagreb would be very familiar with the kajkavian dialect. My own dialect (ikavian štokavian) has many similarities with the southern čakavian dialects. On the other hand I would not be very familiar with the northern kajkavian dialects. Similarly, the Burgenland Croatian standard language, would not be that familiar to me. Having said that, when travelling through Czech Rep., Slovakia and Poland I spoke Croatian and had little trouble communicating.
In conclusion, where there is a will, there is a way to communicate. But the importance of a standard language is as a vehicle for communication and cultural expression within a particular speech community; and the mark of an educated speaker is the ability to master the particular standard language. The standard sits separate from the dialects as a neutral norm – just like in Finnish (to bring it back to the opening topic of this thread).
If we were to disregard politics, some Pakistanis and some Indians speak the same language that used to be called Hindustani.
However, it is not called Hindustani any more. The Pakistani standard is called Urdu. The Indian standard is called Hindi.
It is difficult to see the utility in using obsolete or obsolescent names for languages. Especially where the vast majority of the native speakers do not use those names.
I don’t think anybody’s insisting on using the obsolete names, but it’s important to make the point that regardless of politics and the nomenclatural preferences of native speakers, the linguistic facts are what they are. Urdu and Hindi, at the conversational level, are in fact one language (setting aside the mutually unintelligible high-culture level, larded with Perso-Arabic and Sanskrit loan words respectively), as are Serbian, Bosnian, and Croatian; conversely, “Arabic” and “Chinese” consist of a number of mutually unintelligible languages. It wouldn’t be necessary to insist on those facts if people didn’t insist on denying them. Needless to say, the current names used should be those preferred by native speakers.
Isn’t it rather strange to insist that čakavian and kajkavian are mere dialects, but Serbian is a different language altogether?
I would have thought that you either can be a splitter or a lumper, but not both at the same time.
For what it’s worth, there are separate Wikipedias in Serbo-Croatian, Serbian, Croatian and Bosnian. (Also in Slovenian and Macedonian, which belong to other subgroups of South Slavic.)
There aren’t any Wikipedias in Kajkavian or Čakavian, unfortunately (not quite sure why, since, e.g., a lot of German “dialects” made it).
I would have thought that you either can be a splitter or a lumper, but not both at the same time.
You can be a splumper (if your choice to split or lump depends on expedience).
SFReader: Isn’t it rather strange to insist that čakavian and kajkavian are mere dialects, but Serbian is a different language altogether?
The limits of languages are defined by sociolinguistics. Some sociolinguistic situations ære møte messy than others.
Another way to say it is that what constitutes a language sociolinguistically is not decided by the appllicability of a reference grammar.
And some spelling chequers ære møte stupide than others.
Honour meeting!
I made that comment from my phone. I hate spellcheckers, but I have to use the one installed on my phone just to have easy access to all the letters I use. I also hate commenting from my phone, but I have to do it just to have easy access to all the letters I use while half asleep on the sofa.
I have a box full of letters right here in my office, from the 1950s and ’60s. Nobody writes letters any more!
And still we drown in paper! Nothing worth keeping but I have no way of getting rid of anything.
And I thought it was deliberate….
That would have been so much better.
>I’m not saying that AmEng have never takes /h/, just that I think /ǝv/ is the form that shows up most commonly in real speech (as an auxiliary, that is). In non-initial position and preceding another verb, as in “should have gone”, I can’t imagine any context where I would plausibly pronounce the h.
I would sometimes pronounce the h in an auxiliary if the auxiliary was the last word and I was being emphatic. For instance:
I might’ve given you the book vs. “Would you’ve given me the book?” A: I might have
I suspect that the situation of Serbo-Croatian may prove that presence or lack of mutual intelligibility does not depend only on the actual linguistic distance involved, but also on the willingness of the people to entertain the possibility in the first place. If enough people decide to play dumb when encountering words in a neighboring variety that differ only marginally from their own ones, that’s going to be quite effective in putting a stop to ongoing spreading of linguistic innovations — thus establishing a language boundary instead of a dialectal one; and eventually, most likely, also creating a gap of more solid mutual intelligibility.
By contrast, if there is will, mutual intelligibility can stretch very far indeed. Bilingual Northern Sami – Finnish speakers report to this day an intuitive understanding of many words in the two languages being “the same”, no matter if they diverged phonologically from one another already something like 2500 years ago. A classic example recorded in Finno-Ugric research in the early 1900s was a preteen boy noting the identity of NS gahppir ‘hat’ (< *kɪpērē < *küpärä) and Fi. kypärä ‘helmet’ in passing to a field researcher. (It’s not “full-scale” mutual intelligibility, though: pretty much all non-trivial consonant isoglosses are treated as opaque, and monolingual speakers have trouble communicating across even the Northern – Inari Sami boundary, or between Southwestern and Eastern Finnish.)
…Although, on second thought, Sprachbünde make it hazardous to argue for a language/dialect boundary just on the basis of “permeability” rather than the theoretical possibility of mutual intelligibility. Perhaps it would be a good idea to add dimensionality in our understanding of language borders, instead of treating them as a simple cline from “identical” to “foreign”.
Perhaps it would be a good idea to add dimensionality in our understanding of language borders, instead of treating them as a simple cline from “identical” to “foreign”.
That’s a good point; I’m sure it would.
So from my layman point of view, one question which can be asked is whether
1. I’ve/you’ve etc. are lexicalized forms, and should’ve is ambiguous, since arguably it can’t be distinguished from its non-lexicalized form “should have” (at least in some situations). Or…
2. /v/ is a separate stress-less form of “have” which is phonemically different from /hav/, and the schwa in should’ve is inserted for phonotactical reasons. Personally, I like this analysis, and Lazar’s “Joe’ve” supports it. I feel like there’s an aversion towards non-syllabic words in Western European phonemic analysis, but I’m really too much of a layman to back that up with anything.
Having phonemically different strong(stressed) vs weak(unstressed) forms of words is quite common in Danish e.g.
vil, will/want to: /vel/ vs /ve/
jeg, I: /jɑj/ vs /jɑ/
dig, you (sing.obl.): /dɑj/ vs /dɑ/
skal, must/will: /skal/ vs /ska/
Some (but not others) of the weak forms can even be used in stressed positions, in my mind not unlike the way e.g. (American?) English “because” can be pronounced with /ʌ/ instead of /ɔ/ in stressed positions.
Rumor has it that when the political boundary between Baden and Württemberg in southwestern Germany was abolished in the 1950s, it became a dialect boundary, as people started to assert their identity by preferring words they thought were more common on their side of the former border and avoiding the converse.
That’s right, 1950s, within the Federal Republic of Germany, in the absence of any separatism from it.
Interesting idea. Could hold for Standard Average European.
Now that Danish was brought up, there are lots of nonsyllabic words in allegro speech. [d̥ekd̥ekg̊œː] (fairly broad) for det kan du ikke gøre is a bit sloppy but acceptable, though [d̥ek̚ d̥ɤˈekg̊œːɐ] is ‘nicer’.
Some (but not others) of the weak forms can even be used in stressed positions…
http://langevo.blogspot.com/2013/02/the-secret-ways-of-weak-forms-here.html
Fi. kypärä ‘helmet’
Is that a wanderwort related to Hittite kaupagi > Hebrew kova’, and therefore related to head?
Some Croats also speak two other closely related smaller languages which are distinct from Serbo-Croatian.
Yep, that’s pretty much it. I’d just use several instead of two.and many instead of some. It’s still around 40% of the population (around 30% Kajkavian varieties and around 10% Chakavian varieties), though constantly decreasing. And some Serbs (not really sure how many) speak Torlak.
I would have thought that you either can be a splitter or a lumper, but not both at the same time.
You can be a splumper
The entire concept of narječje vs. dijalekt in Croatia is kind of a bizarre exercise in splumping absurdity.
However, I’m still curious about mutual intelligibility with the ‘minor languages’ (which I assume are Kajkavian and Čakavian).
Kajkavian and Chakavian both consist of many different varieties, with different levels of intelligibility between them. The amount of diversity is huge. In contrast, all Shtokavian speakers can understand each other pretty well even without prior exposure to each other’s varieties. And due to the standard being Shtokavian, Shtokavian is so ubiquitous that really the only people who have a limited understanding of it are very, very small non-Shtokavian children. Without prior exposure, Kajkavians and some Chakavians would find Slovenians (and each other) easier than Shtokavians to understand.
Most Shtokavians’ experience with non-Shtokavian ranges from zero to some, though of course there are also many who come from non-Shtokavian families or live in majority non-Shtokavian areas. So as zyxt says, it depends on the individual Shtokavian. But it also depends on how interested or how good at getting the hang of languages in general the person is. Excepting those with non-Shtokavians in their immediate family, there just isn’t enough exposure for all Shtokavian locals to come to understand an area’s non-Shtokavian varieties well just through osmosis. Non-Shtokavians basically always switch to Shtokavian or at least “water down” their speech when interacting with people outside of their native area (even with speakers of another Kajkavian /Chakavian variety), and non-Shtokavian varieties have been going through different extents of Shtokavization for a long time. Some have been mixed for hundreds of years, but the bulk of this has occurred following standardization, and then later post-WW2 urbanization and cities continuing to devour their surroundings into the present day.
I’d say a conversation between a group of modern speakers of a not significantly mixed non-Shtokavian variety would be a bit more intelligible to a Shtokavian than Slovenian. To many people, Shtokavians and non-Shtokavians alike, many non-Shtokavian varieties are so foreign that they aren’t even able to identify them as some sort of modern variety spoken in the country, let alone understand them. Or to quote parody group Vatrogasci and their cover of this Chakavian mid-90s cover of an American song I can’t remember right now, one of the few modern non-Shtokavian songs that have had success nationally:
Katedomolovo uzenta i ajngemahtec štrudl gendu hacijulage
Ajlagede štrudl ojlzenta, uzenta de ferencvaroš gdl t nje
Ajndi kalunga štimje vuga nam tu, serte rokta nalala tanja lavu
Latanje ferencvaroš primja tugu, samo nedajda dada išvalja!
Ok, that’s obviously very much an exaggeration for humorous purposes, but honestly, it’s not that far off from the way people would sing along to the original lyrics back then, even with them being in the mixed Chakavian-Shtokavian dialect of Western Istria, which should be relatively easy.
The usual etymology is metathesis < *käpürä (perhaps for root-structural reasons: *ü was originally restricted to the 1st syllable), borrowed from Baltic (Lith. kepurė, Latv. cepure ‘hat’). None of the Leiden etymological dictionaries at least seem to equate these with head, caput etc. (or with anything at all in fact), but I could buy this idea if someone wants to actually make the argument.
If anyone’s looking for a form without the annoying third consonant, I can offer German Haube, meaning “knit hat” and the like. But even so we still have, in the first syllable, *au in Germanic, *a in Latin and *e in Baltic…
No. (arch. + dial.) huve, Da. hue “knit hat”.
perhaps 100 differences of vocabulary
It feels like there should be more of them, if only because specialized vocabulary is very different for most areas. I’m no good at estimating these things, though.
It appears where gwenllian and I differ is whether the ijekavian dialect is the same as the Croatian literary language. Ijekavian does come close to the standard, but it is not the same. In support of the view that it is not the same, here are some examples of phonological differences:If we were to disregard politics
My position isn’t that Eastern Herzegovinian and the standard are identical, it’s that they’re extremely similar, and input from other varieties present in Croatia extremely limited. Again, standard Serbian is not identical to Eastern Herzegovinian either, yet people don’t go out of their way to point it out there, because there’s no political or sentimental impetus to distance the standard from E. Herzegovinian or to throw a bone to speakers of other varieties. That there is a number of differences between the standard and E. Herzegovinian is definitely a valid thing to point out, but I just don’t see it as relevant in this particular discussion.
The Illyrians were enamoured of Dubrovnik literature, which is witten in the Dubrovnik Ijekavian dialect (it always retained the /x/), and the Illyrians based their language norms on that.
The Croatian standard that was finally settled on isn’t based on Dubrovnik speech, though. I’m not saying the Illyrians weren’t appreciative of Dubrovnik. Just that if that had been a primary motivation in choosing an ijekavian Neoshtokavian variety to standardize (as is often put forward when standardization history is brought up in Croatia), surely the ijekavian Neoshtokavian variety would’ve been the Dubrovnik one.
However, in Croatia, the standard was always open to the influences of all three – štokavian, čakavian and kajkavian. This is understandable because the main cultural centre – Zagreb – is in a kajkavian speaking area.
A lot has been written about Croatian “purism” in language. The Croatian standard has been “puristic” in trying to substitute Slavic words for internationalisms eg. “odbor” for “komitet” (committee) – and this has been going on for centuries. However, Croatian has been relatively open in admitting čakavian and kajkavian words into the standard, as well as words from other Slavic languages – Czech, for example, had an important role in the formation of the standard – eg. “plin” (gas) and “vodik” (hydrogen). Serbian, on the other hand, has also had “puristic” tendencies in the sense that it rejects čakavian and kajkavian forms – understandably so, since Serbs don’t speak these dialects. In forming the Sebian standard, Serbian language purism was evident in the rejection of all words except those that are “spoken by the people.” The result of this is that the Serbian standard is more accepting of non-Slavic words, eg. “špilzohne” from German for a type of stocking, “ćebe” (blanket) from Turkish.
We’ve already disagreed on the other points in previous threads, but I definitely second the importance of Czech.
which used to be called Serbo-Croatian.
It was called Serbo-Croatian (and, of course, still is internationally and occasionally domestically), but I find that the name thing is often not fully understood by outsiders. In everyday usage the vast majority of Serbs would have always in recent history said that they speak the Serbian language, and the vast majority of Croats would have always in recent history said that they speak the Croatian language. The name used isn’t any indication of what someone thinks about language issues or whether they have anything against the term Serbo-Croatian. It’s just the default and referring to the language as Serbo-Croatian when the context doesn’t demand it would sound a bit odd.
So there used to be a whole lot more people who would’ve answered affirmatively when explicitly asked if they speak Serbo-Croatian, but in most everyday contexts most of those people would’ve just talked about the Serbian language or the Croatian language even back then. That sort of everyday usage of the term was most common in Bosnia. There was also some disagreement on the name used in official contexts, i.e. hrvatskosrpski, hrvatski ili srpski,srpskohrvatski ili hrvatskosrpski and vice versa were also used (e.g. the 1974 Constitution of SR Croatia defines the langauge of Croatia as Croatian, a standard form of the language called Croatian or Serbian). I’m probably missing a name or two. Informally, in a context that calls for a term that encompasses all the Shtokavian varieties, people most often simply used, and continue to use, naš (jezik).
I suspect that the situation of Serbo-Croatian may prove that presence or lack of mutual intelligibility does not depend only on the actual linguistic distance involved, but also on the willingness of the people to entertain the possibility in the first place. If enough people decide to play dumb when encountering words in a neighboring variety that differ only marginally from their own ones, that’s going to be quite effective in putting a stop to ongoing spreading of linguistic innovations — thus establishing a language boundary instead of a dialectal one; and eventually, most likely, also creating a gap of more solid mutual intelligibility.
I don’t really think this fits the situation. Few people really play dumb in that way. For the vast majority the continued existence of separate standards and the standards’names are non-negotiable, but few would claim to have trouble communicating and most are perfectly happy to converse openly, happily and at length with speakers “from the other side”. Even those who really are offended by any mention of there existing a single language in any context rarely actually claim significant mutual intelligibility problems. When it comes to what differences exist, people will actually often use the “other side’s” word(s) if they think the other person might not know theirs. The spread of innovation hasn’t quite stopped either, though it has slowed down, due mostly to a decrease in shared mass media.
@David,
Interesting idea. Could hold for Standard Average European.
Could you elaborate? In my case, I found that French makes more sense to me when I think of de, je, que, le as /d/, /j/, /k/, /l/ instead of /dǝ/, /jǝ/, /kǝ/, /lǝ/. I.e. ǝ is a sound which needs to be inserted in certain phonotactical positions rather than a phoneme which has a zero-realization in certain places. Unfortunately neither my phonemic analysis skills nor my French is strong enough to check the soundness of my intuition.
@The original Lars:
allegro speech
Many things can happen in allegro speech, but I feel English /v/ and Danish /ska/ are categorically different from that, since they are possible in even very slow speech. Some time back I discussed with a Danish phonologist about whether Danish has separate /a/ and /ɑ/ phonemes. His opinion is that it doesn’t, while I think it does, not only because the complementary distribution rules has exceptions, but also because I want to distinguish /jɑ/ (for me the default pronunciation of “jeg”, I, even in stressed positions) from /ja/ (ja, yes).
I’m afraid not; I didn’t think about it much further. In fact, I like your analysis of French; a point in its favor is the fact that ours blanc gets an epenthetic vowel in the middle of its four-consonant cluster about as often as carte bleue does. However, now that I think of it, completely regular epenthesis – so that le can end up seeming to have its vowel on the wrong side – is a specifically Ch’ti feature.
Fun fact: Slovak has a length distinction between ja “yes” and já “I”…
Regarding French function words spelled with “e”: “le” at least can occur in phrase-final “stressed” position, where it is not possible to delete the vowel. Wikipedia gives the example of “dis-le ! /di lə/ → [di.lø] (‘say it’)”
Belatedly:
2. /v/ is a separate stress-less form of “have” which is phonemically different from /hav/, and the schwa in should’ve is inserted for phonotactical reasons. Personally, I like this analysis, and Lazar’s “Joe’ve” supports it. I feel like there’s an aversion towards non-syllabic words in Western European phonemic analysis, but I’m really too much of a layman to back that up with anything.
I think you may be misreading me. I don’t say [dʒoʊv]; for me there’s always at least a short schwa there – [dʒoʊǝv]. I was using this to argue against the tendency to render [ǝv] as ’ve after should, would, could. Myself, I would favor your choice #1 of I’ve, you’ve, etc. being lexicalized forms.
“le” at least can occur in phrase-final “stressed” position, where it is not possible to delete the vowel. Wikipedia gives the example of “dis-le ! /di lə/ → [di.lø] (‘say it’)”
The schwa also stays if you add another clitic, as in dis-le-moi! /di lə mwa/ → [di.lø.mwa] (‘say it to me, tell me’)”, dis-le-nous! /di lə nu/ → [di.lø.nu] (‘say it to us, tell us’). A less standard version (although most common in children’s speech) is dis-moi-le, dis-nous-le, again with stressed schwa.
I could buy this idea if someone wants to actually make the argument
By chance I have by now run into an explanation tackling the *au / *a discrepancy: one Dimitri Pisarev proposes *u-epenthesis in Germanic between unstressed *a and a labial, e.g. *augōn < *agʷā́n ‘eye’, *lauba- < *labá- ‘leaf’. Non-Vernered *haufud would have to involve the levelling of an earlier alternating paradigm with *haub´- ~ *háf-, though. On the other hand, he notes ON hafuð, OE hafudland, seemingly testifying for the non-epenthesized variant required.
> dis-le ! /di lə/ → [di.lø]
Yes, this is a problem for my analysis, but as far as I can see, the schwa-phoneme analysis has the same problem, i.e. why are some terminal schwas (as in /di lə/) pronounced, but not others (as in porte /pɔʁtə/). Analyses that have /pɔʁt/ instead have to explain why a schwa can be pronounced in “porte fermée”.
And in this case, I have no qualms pretending that this “le” /lø/ is different from the non-final “le” /l/. Personal pronouns as objects of positive imperatives already act a bit strange.