Joel of Far Outliers has posted some excerpts of Mikołaj Gliński’s contribution to a newsletter from Culture.pl, How much Polish is there in Esperanto?; it begins:
The Slavic Esperanto lexicon is generally considered to derive from two Slavic languages, both of which Zamenhof knew very well – Russian and Polish. While Russian, as he later recalled, was his favourite during his early days, as well as the language of the larger part of his linguistic output, Polish was the language spoken in Warsaw, where Zamenhof would eventually spend most of his life. And it’s actually these two languages that make up the 1% of Slavic in Esperanto vocabulary. The problem is that sometimes identifying the original language is very hard if not simply impossible.
While Esperanto words like barakti (to flounder, from барахтаться barahtat’sja), gladi (to iron, from гладить gladit’), krom (except, from кроме krome), or vosto (a tail, from хвост) are easy to identify as of Russian origin (they simply don’t appear in Polish), and others like kolbaso can be considered as coming from Russian based on their phonetic build-up (Rus. kolbasa, compare Pol. kiełbasa), there are also some that can be really problematic. This goes for words like celo (aim), bulko (bun), klopodo (effort), kaĉo (porridge, from kasza / каша kaša), prava (right [in opinion], from prawy / правый pravyj), or svati (to matchmake, from swat / сват svat). Due to their almost identical forms in both languages, it seems hardly possible to tell from which language Zamenhof took the original root.
…
And yet the group of Esperanto words which are usually identified as Polish loan words is significant.
If that interests you, click through for more. And here’s a bit that amused me about a very peculiar feature of Esperanto:
While Zamenhof on the one hand made sure that addressing people in Esperanto would be as democratic as possible (Esperantists address each other by a simple vi (you), much like in contemporary English, and unlike in Polish), he also reserved a mode of speaking when addressing the highest royal authority. In finding the right word he somewhat naturally turned to Polish. A choice well justified, as Polish is perhaps the only Slavic language (along with Czech) to have developed an elaborate honorific system that distinguishes between who you’re talking to, based on your age and social status (historically, class of society).
That Polish is at the core of Esperanto honorific usage is best exemplified in the phrase via reĝa moŝto, which translates as Your Majesty, and can be seen as quite a direct calque from the Polish: wasza królewska mość…
There are also sections on “Is Ĉu Polish?,” “Is Esperanto’s accusative Slavic?,” “Semantic calques,” “Phonetics,” “Alphabet,” and “Accent” (meaning stress).
Re “really problematic,” it would seem to me to be more of a feature than a bug, given the overall ambitions of Esperantism, for a Slavic-origin lexeme to appear strategically ambiguous about which specific modern Slavic tongue it might have come from. And ditto for Germanic-origin or Romance-origin lexemes.
That looks like Hungarian cél, so I wondered if Hungarian got it from a Slavic language, but no, both Hungarian and Polish got it from German Ziel.
Re “really problematic,” it would seem to me to be more of a feature than a bug
I don’t think he means it’s problematic in the usual English sense — rather, that deciding which specific language is the source is a problem (i.e., not immediately obvious).
If the word is so similar in the two candidate donor languages, does it really make sense to ask where it is from? I mean, it could just have been a part of Zamenhof’s background knowledge of Slavic.
The “phonetics” section is too short. Here and in the next two sections, for example, is some puzzlement about how j and ŭ work and why they aren’t parallel to each other as /j/ and /w/ are in English. Solution: ŭ is not /w/ and not really [w]; rather, like Polish, Esperanto has exactly two diphthong phonemes, aŭ and eŭ, despite lacking any other diphthongs (aj, again as in Polish, is a vowel followed by a consonant). In Polish, the diphthongs are rather rare loanword phonemes and often actually pronounced as two separate syllables (I’ve heard variation from the same speaker within a minute); they just count as single syllables for stress assignment. (They’re also not graphically distinguished there from random vowel collisions, e.g. nauka [naˈuka] “science”; that must be why Zamenhof added the breve – once again like Polish, or even more so, Esperanto mercilessly stressed the second-to-last syllable of any word, so it matters what counts as one syllable.)
The “phonetics” section is too short.
Indeed; as also noted in the JBR version, Esperanto distinguishes (…or, was intended to distinguish; AFAIK it’s becoming increasingly nonphonemic, if it ever was phonemic in the first place) between /x/ and /h/, which is (JBR says and I have no reason to disbelieve) a thing in some Polish dialects but not in the standard language.
I know how to distinguish between cats based on THEIR colour.
I could distinguish between my selves based on OUR colour. If I were many.
But the idea of distinguishing between cats based on my colour hanged my mind for a minute.
Does he mean that in Polish and Chech terms of address depend on both who is talking and who she is talking to, two variables rather than one?
“a thing in some Polish dialects” – isn’t it a range of voiced fricatives for etymological *g, rather than [h]?
I think so, and I also wonder if it’s general eastern or just southeastern. However, the spelling (if any) used for this phenomenon is indeed h; there are people named Halina / Halszka out there.
I think this is about distinct pronunciations for written “h” and “ch” (e.g., in handel “trade” vs. chóry “sick, ill”). When I started speaking Polish, I pronounced them [h] and [x]*); respectively, and was told by our teacher that I sounded like someone from Eastern Poland, and that the Standard pronounces them both as [x].
*) Or whatever voiceless velar fricative I have.
Interesting. Yes, they’re both [x] in the mainstream (which you have at the very least in, say, Buch).
I’m confused.
On the one hand:
– Czech, Belorusian and Ukrainian have fricative G (spelled h in Czech and г in Belarusian and Ukrainian). I expect some Polish dialects to have same pronunciation.
On the other hand:
– if your G is voiced fricative, that doesn’t mean you will be tempted to read Polish h in a book borrowing* as (your) g and not as others in Poland read it, that is as Polish ch).
For this you need a different tradition of reading Latin. (one way to have it is to shift from literary Czech or Cyrillic to literary Polish).
* handel is not a book borrowing, but other words are and if they learned their pronunciation of handel from German speakers, I don’t understand how those who have velar fricative G can hear it rahter than [x] in German [h].
Yes, you absolutely need to have some connection to a tradition of *g as [ɦ] (Czech, Slovak, mainstream Ukrainian, and AFAIK southeastern Polish).
I can’t remember a language where /ɦ/ and /h/ both are phonemes. I wonder why.
And it is not that they are necessary allophones of each other. English /h/ is normally [h], Slavic [ɣ~ɦ] is voiced.
It’s very rare because they’re just too similar. One Basque dialect apparently has both.
Supposedly it has turned into voiceless [h] in southwestern Ukraine. Unfortunately, the only evidence I’ve seen is a map I can’t find anymore; I was sure it was in a paper by Henning Andersen, but apparently not. If it’s real, I wonder about Hungarian influence, but the [h] zone was much larger than the area that ever belonged to Hungary and also didn’t extend into Slovak…
Nawdm has a contrast of (written) h and ɦ, but unfortunately Nawdm ɦ actually represents /ʔ/. (Presumably they adopted the symbol ɦ to make the uppercase/lowercase distinction easier.)
a language where /ɦ/ and /h/ both are phonemes
Lots are, if the databases are to be believed. Those include many Indic languages and Southern Bantu ones.
At least Sanskrit and Punjabi, as far as I know, only have one of the two, pace the database.
“too similar”
@DM, subjectively (sound so to you) or…? For me they are very different.
Vibrating vocal chords are vibrating vocal chords.
I was expecting that database’s report on Dagara to be wrong (its source is extremely unreliable on Oti-Volta), but Alain Delplanque’s Dagara grammar actually does support the h/ɦ distinction, with /ɦ/ classified as a voiced glottal fricative. Actually, the voiceless /h/ is not the regular outcome of any proto-Oti-Volta consonant in Dagara, but (as in Kusaal) it turns up in well-established loanwords and phonaesthetic vocabulary, so it would be churlish to deny it residency.
There’s some very odd stuff going on with Dagara phonology. It has a ɓ/b contrast, too, which is unparalleled in Oti-Volta. Dunno how it originated. No good cognates to shed light on it all.
Dagaare, the Ghana-side part of the same dialect continuum, doesn’t have any of these weird unOtiVolta consonants. Maybe it’s some sort of substrate thing. (But the local Burkina-side Grũsi languages, the obvious culprits, don’t have these odd consonants either. Mande? Something else?)
Well, Sanskrit has both – [h] is an allophone of… uh… well… /s/ or /ʂ/ or… it might have to be considered a separate phoneme after all; I’ll need to reread a certain paper.
The IAST transcription is backwards because it takes frequency into account: the very common h is [ɦ], the rare (word-final only) ḥ is [h] (…including [x] and [ɸ] in Vedic).
Subjectively to me and everyone with a similar phonetic background…
@DM, I think here we must rememeber that Arabs have 3 [x]’s (from the Russian perspective) and Russian has 7 s-sounds (as some Finnish speaker said, they say), [p] and [b] are allophones in some Arabic idiolects and perhaps about p in English spV.
Are there voiced allophones there, in your phonetic background?
No, but then there aren’t voiced allophones of /b/ or /d/ either… However, [ɦ] has been spotted in some Englishes as an allophone of /h/, and also in the kinds of Dutch that retain voiced fricatives, AFAIK.
I remember how I was trying to test my superficial understanding of Arabic phonology by transcribing speech and was perplexed by the first h in ẓahra-ha.
(https://vocaroo.com/1jntMMYyRI3n She’s speaking very quietly, that’s her voice is not much louder than her noice)
I knew that Arabic h can be voiced, likely I recognised possessive ha “her”. But I didn’t think the sound in ẓahr “back” could be h. I was thinking “it is not ghayn… it is not ‘ayn… what’s it?!” Then I noticed that the second h is voiced too. I could accept it as a /h/ with some hesitation, but not the first sound.
I actually hear the ẓ as [n], consistently; I blame my current lack of headphones…
The second h is definitely voiced, though it took me a while to be sure. The first – I can’t really say how I’d have identified it if I didn’t already know, especially given the fact that I’m only used to [h] in prevocalic positions; it does seem to be the same thing, though.
(i mean, that was long ago, I was a beginner in Arabic and the fact that I still remember that illustrates the point that perceptually voice changed a lot to me)