Links: Prequel about PIE *(s) and laryngeals: • Super challenging Prot... Timed link for Umlaut: • Summarizing Germanic s... Simon Roper’s video about PIE having two vowels: • Did Proto-Indo-Europea...
Bulgarian still has a very active Ablaut system and it's funny because nobody ever questions it or is thought about it we just intuitively know when, in both spelling and speech, e becomes goljam er or just gets shadowrealmed. I probably would've never even knew about the concept had I not studied OCS.
I'm always super stoked when I see a video from this channel on my feed. Another smashing video! I'd love to see a video explaining why a distinction between inanimate objects and animate objects in PIE means I need to remember that 🍎is masculine, 🍌is feminine and 🍞is neuter.
Thank you for this highly informative video. You answered a few nagging questions I had. On another note, if you have enough information on this, would you be willing to make a video about the development of the feminine gender in Indo-European, and how it relates semantically to collectives/plurals and abstracts? It’s peculiar, but it seems to mirror the evolution of *-(a)t in Afro-Asiatic, which, despite being the foremost feminine suffix now, does not seem to have been the original or at least primary feminine suffix, instead seemingly coding for roughly the same semantic categories [and more] as the feminine in Indo-European. I have seen a fair bit of information that this may have been influenced by patriarchal attitudes espoused at some point, but that seems to a slightly less sought-after conclusion now.
Sylvia Luraghi has published some excellent papers on the rise of the feminine gender in indo european, her conclusions are based on cross language family studies and in a nutshell is based on animacy vs inanimacy or rather the ability of an agent to participate in the reproductive process (hence why children or young animals are often neuter) and when a suitable morphologically suitable suffix became available (via perhaps collectives) then its natural that the next derivation seperates the two genders. Luraghi explains it much better than I do so check her out. Matasovic is worth a read too.
I love your theory about the isolated reduplicative prefix becoming Germanic past participle prefix *ge-. I want to say that despite inherting the finite perfect verb well, Germanic doesn't inherit the reduplication regularly for those perfects that now form the past tense. Nor is the IE perfect active participle retained in Germanic, since its function might be replaced by the present active participle now that there is no aspectual difference between the present and past ( < IE perfect) tenses.
As a german teacher, I was NEVER able to explain why the "ge" prefix was used in the present tense like "gehören" and this video that wasn't even supposed to be about it, just answered it. Thanks mate!
Can you do a language overview video about Lithuanian, please? It's the oldest surviving IE language and has a complex grammatical system. As far as I heard they still use dual forms.
It isn't the oldest, though. Remember, all languages are equally old, meaning English is just as old as Lithuanian, and if you go by when it was first written down, it's only been written down since about 1500 AD. It is, however, considered to be the most conservative modern Indo-European language. The dual does persist in some dialects, though.
I think the phrase you're looking for is "most conservative." At least, that's how people often describe it -- it's not the "oldest" language because that's not really a meaningful concept, but it is one of the surviving languages which retained the most features directly from PIE
6:36 so the ā here is non-phonemic due to the Stang's law which states that if a semivowel (i.e. y or w) or a laryngeal (h1, h2 and h3) is followed by a nasal, then it is deleted with compensatory lengthening. this means that the original form is *h1dteh2m, where the e has an allophone of a before the h2. so the whole chain of sound changes is *h1dteh2m > *h1dtah2m > *h1dtām also in transcription dt should be tst or something similar due to the no geminates law in PIE
Thanks for the new answer regarding "ge-" (that was my question you answered). Makes total sense - after all, "going to" is used to express the future - why not have the same verb in a different form express the past?
I would actually like to take a moment now that you mentioned a possible alternative for the origin of the prefix *ga-, since I believe it’s precise origin is key for understanding the phonology of Proto-Germanic. This is something I’ve given much thought to, so I hope you can give it a read and hopefully your other viewers find it interesting as well: Although *ga- being derived from a very productive *ghe- prefix from the stative in Pre-Germanic is an interesting proposition, I’m actually more inclined to believe the theory deriving it from PIE *k’om via Grimm’s+Verner’s Laws, of course Grimm’s Law alone wouldn’t suffice to explain this development, but taking in consideration the following…: - Verner’s Law states that Pre-Germanic fricatives become voiced following an unstressed vowel - Syntax from PIE all the way to Proto-Germanic most likely followed an SOV order - Most nouns in Pre-Germanic would have had an unstressed ending (*-az, *-ō, …) …it’s not a stretch to think that a prefix inclined to follow an unstressed vowel would have been subject to Verner’s Law. Now, here’s why I stick so firmly to this idea, which I find the most interesting and I’ve never heard anyone or read any articles taking about this: if we picture the development of PG from PIE while sticking to the phonological structure of a modern language, a logical development of the PIE stops would be bh -> b, b -> p & p -> ph, just as it happened in Armenian, this makes up for the rarity of *b (now interchanged with a rarity of *p, although still found after *s and in geminated context) and the odd lack of voiceless aspirates in a system that allows for aspirates, and voiced aspirates at that, all issues that different IE languages dealt with in different ways. KEYNOTE: voiced aspirates becoming voiced stops, not fricatives; this would go against a recent theory suggesting Pre-Germanic voiceless obstruents (aspirates + *s) became voiced given the conditions of Verner’s Law before the voiceless aspirates became fricatives, merging with the PIE voiced aspirates which, in this model, don’t loose their aspiration, only to then develop into voiced fricatives altogether. I don’t particularly agree with this model for a number of reasons: - This would suggest that, at some stage of Pre-Germanic, the stop consonant repertoire of the language would have consisted of voiceless and voiced aspirates, voiceless stops and a complete lack of voiced stops, which seems like a trade of an irregular system for another, disregarding the motive that would have caused the chain shift in the first place - Fricatives are more likely to become voiced in certain phonological contexts, by assuming their fricative nature is what allowed them to become voiced, this makes the development of *s into *z fit more neatly with the other consonant developments of Verner’s Law - Finally, and what I’ve heard no one argue, clearly the voiced fricatives would have existed in an allophonic relationship with the voiceless fricatives, and for a phone to be an allophone of a phoneme, they have to be perceived as the same sound by the speakers of the language, having the voiced fricatives exist outside of this relationship (sentence-initially, for example) would break it completely. Because of this, the assumption of many has always been that the voiced fricatives went through fortition word-initially in order to make way for the allophony of the fricatives, but this doesn’t solve the issue completely, it doesn’t account for all the word-medial voiced fricatives that followed a stressed vowel and the fact that *g remained a fricative word-initially In order to make sense of what would be the Proto-Germanic consonant repertoire the following model gives the best explanation: - PIE stops go through a chain shift (Grimm’s Law) bh -> b, b -> p & p -> ph - Voiceless aspirates become fricatives ph -> ɸ, th -> θ, kh -> x & khw -> xw - Fricatives develop voiced allophones following unstressed vowels (Verner’s Law) ɸ -> β, θ -> ð, x -> ɣ & xw -> ɣw, [ɣ] in the prefix *ga becomes the only word-initial voiced fricative - Stress becomes fixed in the root syllable, phonemicizing the voiced fricatives, although their mostly word-medial position would have made them relatively unstable as independent phonemes - Further changes in the language draw the voiced stops and fricatives closer together into an allophonic relationship: e.g. voiced stops lenite word-medially and voiced fricatives go through fortition after nasals, [g] from PIE *gh and [ɣ] from Verner’s Law still co-exist word-initially preventing the allophony between the two phones - The productivity of the prefix *ga- (being able to be attached to almost every verb) makes [ɣ] win over [g] as the word-initial realization of the phoneme /g/ - Labio-velar fricatives [xw] and [ɣw] debuccalize to [ʍ] and [w] respectively. [gw], having collapsed its allophony with [ɣw] (given that [w] could already occur word-initially), merges with [b], except after /n/, this neatly explains the development of gw -> b, but kw -/-> p - [x] debuccalizes to [h] in some environments, but remains unchanged in others, as does [xw] - [ɣ]‘s allophony with [g] prevents it from debuccalizing, this doesn’t happen for [ɣw] since its debuccalization is what kickstarts this last chain shift in the first place So, in theory, hadn’t Pre-Germanic taken PIE *k’om to form its past participles, today we’d say jesterdadge instead of yesterday and jenough instead of enough
First of all, WYL seems to have brought up his ga- proposal to the wiktionary editors, so I suggest you do the same, if only to help the dictionarians get a balanced perspective. Even better if you have a source that agrees with this analysis. There nevertheless exist fairly convincing examples of the exact opposite of this analysis: Verner's law applying prior to Grimm's. There are borrowings into Germanic that seem to require a fairly late date for Grimm's law (hemp comes to mind), yet there exists evidence of Germanic languages having undergone the Verner shift prior to that point. What's your perspective on such things, and how do you square it with your [admittedly impressive] theory here?
@@Copyright_Infringementhello my friend, first of all I’m sorry it has taken so long for me to answer, I’m quite a master procrastinator myself, I appreciate your kind opinion of my theory for which I must say I couldn’t rely on any particular source to come up with, so I doubt I would be taken seriously if I were to present it on wiktionary, and in regards to the possible contradictions I must say I haven’t found any particular example of a specific word of a specific Germanic language that would suggest it might have gone through Verner’s Law before Grimm’s Law, although I would always welcome sources to give me a more complete picture. I think in the case of the word ‘hemp’ (of which I didn’t previously know its etymology, so I extremely appreciate the little bit of insight) we have to make a lot of assumptions, first regarding the estimated time period in which PIE would have been spoken, which has recently been contested, then there’s the time period in which hemp would have first been used and where, for which new discoveries can come up, and finally the amount of time it would have taken for one sound change to occur following the other one, which there might be no way of knowing how much it was. Either way it’s fun to speculate, specially when we can think of the changes languages go through as logical and natural developments, instead of random occurrences that make the language unnecessarily complex.
Wow, great stuff! Glad I found your channel. So you mentioned in the video that ablaut doesn't explain the irregular past tense in Germanic languages. Then what's the origin of it? Like sing sang, eat, ate etc. Thanks in advance.
Because PIE is simply the common ancestor of all the IE languages, i.e. it's the last time when all the ancestors spoke the same language, after PIE, we get divergence. In order to reconstruct PIE to an older point, we would need to find other languages or families that are related, but split off at an earlier point. We can theorize about what PIE's ancestors sounded like, but without other descendant languages to use the comparative method on, there's no way to determine if that was the case.
I don't feel like ga- (or bi- for another example) has to follow grimms law perfectly, since both suffixes had the unique feature that set it apart from the rest of the verb/noun prefixes that it was unstressed in BOTH verbs and nouns (unlike uz-, that was unstressed in verbs but stressed in nouns) This is important, because that means that for both of them the stress was on the syllable after it, and it isn't an uncommon sound change for consonants in unstressed syllables before a stressed one (i.e. in pretonic position) to become voiced (like old irish "to-", which was pretonically in verbs was "do-" instead) Assuming this, it would also explain gothic dis-, which seems to be used for the same meaning as twiz- in cognate words in other germanic languages. The fact this is just in gothic would indicate that this is a later pretonic voicing only it had, but it would've been for the same reasons This is of course just another theory, but it is worth noting that bi- and ga-, on top of being always unstressed, are also the only simple CV- prefixes, so while we got CRV- prefixes like fra- that show unvoiced fricative, there is no *fi- or *ha- equivalent with an unvoiced fricative followed only by a single vowel Also, while the reduplicated ge- might make sense if we were just taking modern german, it doesn't explain why it is in fact ga- in gothic, with an "a" vowel, and seems from the oldest sources in other germanic languages that vowel is the original one, while "e" is just what it changed into in unstressed position
@@SchmulKrieger 𐍃𐌻𐌴𐍀𐌹𐌸 (he/she sleeps) 𐍃𐌰𐌹𐍃𐌻𐌴𐍀 (he/she slept) 𐌲𐌰𐍃𐌰𐌹𐌶𐌻𐌴𐍀 (he/she slept/fell asleep). German ge- is descended from old high german gi- which is related to Gothic 𐌲𐌰-, both used as perfectivity markers and as the meaning "together" as in PIE *ḱom which appears to be the origin for Gothic 𐌲𐌰𐌼𐌰𐌹𐌽𐍃, Latin communis (both meaning common, German still has the word gemein, with the ge-, though the meaning has shifted somewhat)
The Germanic ge- (or gi-, ga-) is actually the beginning, the end or the progressive one of an action. Like in some German dialects and some small amount of verbs in German that uses ge- in the present tense for durational aspects, like gebären (ybe bearing) as in die Mutter gebirt ein Kind (she is bearing a child). Or ich geleite dich über die Brücke - I am leading you over the bridge.
Can i find a table of all the vowels of your dialect? Because you keep reffering to it here and there in your videos and it made me curious how exactly /ɞu̯/ and /ʏu̯/ fit in
It’s not a table, but I list all of them at 1;09 in the Future American video. I don’t think I’ve ever made specifically a table of my dialect for RU-vid, and it would be hard to find one elsewhere online because typical literature about American English gets this wrong. So I don’t know, sadly
Can you also explain the upper case Hs with number that sometimes appear in PIE words? Like H1 H2 and H3. Also doesn't anyone think that this language with so many hs and aspirated sounds sounds a bit artificial, and something you would rarely hear in the real world?
There was this video (one of the ones mentioned at the start): ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-7My8EnkJEDU.htmlsi=OmsBECYJI6LhMYsV Also, as with any reconstruction, it's not supposed to sound like how we absolutely think they sounded like, but approximations that explain the similarities & differences with daughters. One video played with the idea of people in the future reconstructing English & with it it's not "much like what you'd hear in a language" - that isn't really what reconstructions aim for. Hope that helps! :)
upper-case with numbers? Usually you have lower-case with numbers, and upper-case without number if you don't know which laryngeal should be reconstructed, but want to show there should be one there.
Due to influence from IE languages beginning from thousands of years ago to the current day, Turkish also has ablaut,but oddly only in the dative form of the first person singular and second person singular pronouns.(ben -> bana sen ->sana)
Das stimmt für das moderne Deutsch in 2. Pers. Plural, aber früher sagte man auch Dinge "es ward ihm prophezeiet", "ich ward der Gefahr gewahr", in diesen Fällen (1. und 3. Pers. Singular" ist "ward" durch "wurde" ersetzt worden
@@davidaxelos4678das stimmt nicht. ward ist einfach das Präteritum Singular als Stamm von werden, im Gegensatz zum Plural im Präteritum wurdun. ähnlcih wie früher, starb, sturbun, aus dem Plural Stamm wurde der Konjunktiv II mit Umlaut abgeleitet, also würde für werden aus wurdun und stürbe von sterben aus sturbun.
Something's unclear to me. You say that PIE has different lexical verbs for different aspects, but then you demonstrate the imperfective and stative aspects of the same verb... ?
"Different lexical verbs" basically just means that the root forms are different words. I myself am not really sure if this applies to PIE, given how different the conjugation charts are between the aspects, but if you're familiar with Slavic languages, they have a system where perfective verbs look like their imperfective equivalents, but in an inconsistent fashion. That randomness is the extra detail I don't know about regarding PIE.
Well, this is all quite a lot to take in for someone trying to own this knowledge. There is so much important information presented so quickly both in written and spoken format (usually at the same time) that it's VERY difficult to follow. The speaker, obviously with a deep knowledge of the subject, speaks so quickly that I (and perhaps many others?) have a hard time absorbing important concepts before something new is I feel like I'm getting whiplash. My plea for future videos is "Please please slow down." Yes, I know what the pause button is for, but even using it (over and over and over) it breaks what would otherwise be a clear formulation of thought into chaos. Regardless, thanks for your very interesting submissions.
4:53 In Ancient Greek ει represented a long i [i:], not a long e [ɛ:], long e was η. Ει being long i is contrasted by ου being long u [u:] where e and o are counterparts and i and u are counterparts. Long e and o also having their own, unique characters to describe their long forms with η and ω.
It represented a long [e:] in Attic Greek in the 5th century BC. The change to a long [i:] was later. (though ου was a long [u:] at the time, it came from an [o:] sound, most likely while a single υ was [u], short or long, which eventually became pronounced as [y]). This is actually well-documented, though the change from [e:] to [i:] did show up early on in Koine (looking at the Koine Greek Phonology article on Wikipedia, it looks like it was pronounced *both* ways for a while in learned pronunciation from the 4th century BC to the 2nd century AD. Here's a quote, which has a citation next to it: "The ει pseudo-diphthong was confused with ι in manuscripts, except before a vowel, where it was confused with η, so it probably retained its ancient value there.").
@@akl2k7 ει as [i:] in Attic dialect would be just one dialect. Specifically as well in -500, in the Boetian dialect ει was also [i:]. Generally when we are talking about blanket Ancient Greek phonology it's referring to Koine. But also, even further back in history during Archaic Greek some dialects also had ει was also [i:].
"Some PIE-cists owe me an explanation about this A here" *_*Okie dokie!*_*_ (Disclaimer: just a linguist who's followed this stuff for a long time, not actively in IE research atm)_ First of all, many just straight up propose "-teh₂m", rather than "-tām", which I assume you can see the reasoning for yourself; Fortson even proposed that the M is not necessarily part of the ending, citing Sanskrit's -/s/ in such forms. The version with long A derives ultimately from Sihler's [incredibly thorough] analysis, which frequently seems to use laryngealless forms in larger tables as shorthand for forms that suddenly _have_ a laryngeal when mentioned in the text; the dual simply has the misfortune of being skipped over in his analysis. I will go one step further, however: the reflexes that Sihler lists don't all necessarily "agree", for lack of a better term (OCS has 2 forms, Avestan indicates a shortened vowel); if we take into account Greek's propensity towards inserting schwas before 0-grade laryngeals, then 3 out of 4 reflexes can be explained with "-th₂m", the same number as can be explained with "-tām(-teh₂m)". To be clear, I don't actually advocate for that analysis (I like "-teh₂m" best); I only want to make the point that -tām" is a single unexpounded-upon entry in a single table in a book that plays fast-and-loose with notation in tables, an entry which doesn't even seem to fully explain its own reflexes"Some PIE-cists owe me an explanation about this A here" *_*Okie dokie!*_*_ (Disclaimer: just a linguist who's followed this stuff for a long time, not actively in IE research atm)_ First of all, many just straight up propose "-teh₂m", rather than "-tām", which I assume you can see the reasoning for yourself; Fortson even proposed that the M is not necessarily part of the ending, citing Sanskrit's -/s/ in such forms. The version with long A derives ultimately from Sihler's [incredibly thorough] analysis, which frequently seems to use laryngealless forms in larger tables as shorthand for forms that suddenly _have_ a laryngeal when mentioned in the text; the dual simply has the misfortune of being skipped over in his analysis. I will go one step further, however: the reflexes that Sihler lists don't all necessarily "agree", for lack of a better term (OCS has 2 forms, Avestan indicates a shortened vowel); if we take into account Greek's propensity towards inserting schwas before 0-grade laryngeals, then 3 out of 4 reflexes can be explained with "-th₂m", the same number as can be explained with "-tām(-teh₂m)". To be clear, I don't actually advocate for that analysis (I like "-teh₂m" best); I only want to make the point that -tām" is a single unexpounded-upon entry in a single table in a book that plays fast-and-loose with notation in tables, an entry which doesn't even seem to fully explain its own reflexes
Great video! just a small nitpick. english definitely doesn't have /ø/ in foot, it's more like [ɘ~ə~ɤ] However I have heard a guy speaking a dialect from the UK that had /ø/ for the stone-vowel, I don't know what area he was from, but he was a minecraft youtuber iirc. so he went "to /ɡø/ mine /støn/"
