Lingua Graeca
- Thread starter sprog
- Start date
I've only just begun teaching myself Attic Greek (the prestige dialect of Athens and the one most commonly denoted by the more general term "ancient" or "classical" Greek) a few days ago, and I was actually struck by how many parallels with Latin I'm already finding, not just with the grammatical structures (which wouldn't have surprised me as much) but in the specifics of morphology and lexicon. I knew going into it that Latin and Greek were distantly related, but I suspect they may have split off from each other a bit later in Indo-European history than I'd thought.
Here are some fascinating tidbits!
αγω (agō) - lead, bring (possible semantic shift in Latin to "do, act, peform")
αναμενω (anamenō) - wait, stay (possible metathesis/clipping to form "maneō")
απo (apo) - from ("from" might have been used metaphorically to mean "according to," hence "apud")
εγω (egō) - I
εκ = (ec/ek) - out of
εν (en) - in, on
λεγω (legō) - speak, say (possible semantic shift in Latin to "read")
'oτι (hoti) - that [subord. conj.] (possibly related to Latin "ut")
πατηρ (patēr) - father
φευγω (pheugō) - flee
φερω (pherō) - carry, bring
θεoς (theos) - god
Morphological observations:
The letter 'ω' ('ō') serves as the first person singular marker in more than one verb tenses.
In at least two of the verb tenses thus far learned, 'ς' ('s') serves as the second person singular marker.
The letter 'ν' ('n') occurs in at least two of the three Greek declensions as the thematic/terminal consonant of the singular accusative. Being a voiced alveolar nasal, the 'n' sound is very close phonologically to the voiced bilabial nasal otherwise known as 'm.'
Similarly, the masculine second-declension nominative ending is usually "-oς" ("-os"). If I recall correctly, the vowels 'o' and 'u' are phonologically almost as close as the consonants 'n' and 'm.'
These two correspondences ('n'/'m' and 'o'/'u') converge and therefore become even more obvious in the singular masculine accusative and singular neuter nominative/accusative second-declension forms, most of which end in "-oν" ("-on"). Reverse both changes, and we arrive at our familiar "-um."
The neuter pleural nominative/accusative forms in the second declension end in '-α' ('-a').
As mentioned briefly in the two observations above, the nominative and accusative forms of neuter nouns tend to be identical, as they are in Latin.
The second person plural verb forms I have so far learned tend to end in "-(ε)(σ)τε" ("-(e)(s)te"). On a related note, the present indicative form of the verb "be" in that same person and number is, lo and behold, "εστε," which transcribes directly as "este." In the same tense and mood of the same verb, there's also a near-direct phonetic equivalence in the third person singular "εστι(ν)" ("esti(n)").
A few definitions for those unfamiliar with linguistic jargon:
metathesis = rearrangement of component sounds into a different order (the Spanish verb meaning "forget" is "olvidar," a likely metathesis of Vulgar Latin "oblitar(e)," from the 4th prin. part of "oblivisci," accompanied by the lenition, or softening, of intervocalic 't' into 'd' and 'b' into 'v' in certain positions)
clipping = the loss of certain portions of a word (I believe this is how we got "bra" from "brassiere")
phonology = the science of sounds in language
morphology = essentially the science of how roots and affixes are used to form words and inflected forms thereof
semantic shift - a shift in usage of one word that eventually gives it a meaning that is distinct yet thematically related to its cognate in another language (a good example is English "tide," a cognate of German "Zeit," meaning "time")
Here are some fascinating tidbits!
αγω (agō) - lead, bring (possible semantic shift in Latin to "do, act, peform")
αναμενω (anamenō) - wait, stay (possible metathesis/clipping to form "maneō")
απo (apo) - from ("from" might have been used metaphorically to mean "according to," hence "apud")
εγω (egō) - I
εκ = (ec/ek) - out of
εν (en) - in, on
λεγω (legō) - speak, say (possible semantic shift in Latin to "read")
'oτι (hoti) - that [subord. conj.] (possibly related to Latin "ut")
πατηρ (patēr) - father
φευγω (pheugō) - flee
φερω (pherō) - carry, bring
θεoς (theos) - god
Morphological observations:
The letter 'ω' ('ō') serves as the first person singular marker in more than one verb tenses.
In at least two of the verb tenses thus far learned, 'ς' ('s') serves as the second person singular marker.
The letter 'ν' ('n') occurs in at least two of the three Greek declensions as the thematic/terminal consonant of the singular accusative. Being a voiced alveolar nasal, the 'n' sound is very close phonologically to the voiced bilabial nasal otherwise known as 'm.'
Similarly, the masculine second-declension nominative ending is usually "-oς" ("-os"). If I recall correctly, the vowels 'o' and 'u' are phonologically almost as close as the consonants 'n' and 'm.'
These two correspondences ('n'/'m' and 'o'/'u') converge and therefore become even more obvious in the singular masculine accusative and singular neuter nominative/accusative second-declension forms, most of which end in "-oν" ("-on"). Reverse both changes, and we arrive at our familiar "-um."
The neuter pleural nominative/accusative forms in the second declension end in '-α' ('-a').
As mentioned briefly in the two observations above, the nominative and accusative forms of neuter nouns tend to be identical, as they are in Latin.
The second person plural verb forms I have so far learned tend to end in "-(ε)(σ)τε" ("-(e)(s)te"). On a related note, the present indicative form of the verb "be" in that same person and number is, lo and behold, "εστε," which transcribes directly as "este." In the same tense and mood of the same verb, there's also a near-direct phonetic equivalence in the third person singular "εστι(ν)" ("esti(n)").
A few definitions for those unfamiliar with linguistic jargon:
metathesis = rearrangement of component sounds into a different order (the Spanish verb meaning "forget" is "olvidar," a likely metathesis of Vulgar Latin "oblitar(e)," from the 4th prin. part of "oblivisci," accompanied by the lenition, or softening, of intervocalic 't' into 'd' and 'b' into 'v' in certain positions)
clipping = the loss of certain portions of a word (I believe this is how we got "bra" from "brassiere")
phonology = the science of sounds in language
morphology = essentially the science of how roots and affixes are used to form words and inflected forms thereof
semantic shift - a shift in usage of one word that eventually gives it a meaning that is distinct yet thematically related to its cognate in another language (a good example is English "tide," a cognate of German "Zeit," meaning "time")
Have you consulted etymological dictionaries so as to confirm (or reject) your guesses? Anyway, similitude of pronunciation does not mean anything (think of Latin habēre and German haben, whose imperatives are almost identical (habē! habe!) yet they are not cognates). It’s only regular correspondence of sounds that matters. In order to make sure that given Latin and Greek words are cognates, one should prove that they originate from the same Indo‐European root (since it’s well known that Latin and Greek do not have any close relationship within the Indo‐European family).
They say that Sihler’s New comparative grammar of Greek and Latin is a decent book on the subject.
They say that Sihler’s New comparative grammar of Greek and Latin is a decent book on the subject.
Nope, those are all just possibilities and speculations at the moment. Correlations like these, coincidental or not, tend to fascinate me and get my mental gears turning. In fact, I even thought about how I might go about confirming or disproving some of my hypotheses, and looking for multiple instances of the same sound correspondence was what most readily popped into my head. It might make a fun project. I'll be going off to grad school to pursue a PhD in linguistics next fall, so it might even make a good thesis eventually!
Really? I'd be interested to know how that happened! It would surely be the linguistic equivalent of convergent evolution in biology (e.g. fish and whales, which have different ancestors but markedly similar forms).Quasus dixit:
I’m afraid it has already made a lot of theses.Gregorius dixit:
It’s neither convergention nor borrowing. German haben is the cognate of Latin capere: the correspondences Germanic h ~ other I.‐E. k is a particular case of Grimm’s law (cf. cornū — horn, caput — head < heafod). Here my linguistic knowledge gets exhausted and I can’t account for b ~ p.Gregorius dixit:
There are a lot of similarities between Greek and Latin, it is true. Both are fairly conservative Indo-European languages, so this isn't too surprising. The declensions in particular are very similar.Gregorius dixit:
The basic meaning of both verbs is "drive" in the sense of directing someone/thing by forcing them forward. Because of this there's a large degree of semantic overlap, though of course some of the more specialized meanings differ between the two languages.
The non-composite form of this verb is just μένω, so there's no clipping at all, actually. The preposition ἀνά "up, upon" is used in composition with many Greek verbs.
I'm not sure about the etymology of apud, but ἀπό is believed to be cognate with ab. Cf. the relationship of Latin super and sub with Greek ὑπέρ and ὑπό (initial 's' in Latin corresponds with initial rough breathing in Greek).
Yes, these are obviously all related. ἐξ is the form of εκ used before words that begin with a vowel.
The primary meaning for both verbs is "pick/gather", so again it's only in the extended meanings that Latin and Greek differ. In both languages the extended meaning of the verb has become the predominate meaning, but composite forms retain the older meaning, e.g. συλλέγω and colligo (3rd conj.) mean essentially the same thing: "gather together".
I'm not sure what the origin of ut/utī is, but I doubt it is related to ὅτι. ὅτι is a combination of the relative pronoun ὅ (neuter accusative singular) and the indefinite pronoun τι (neuter accusative singular), which together is used as an indefinite relative like Latin quidquid "what[so]ever". But as a subordinating conjunction it is in many respects functionally equivalent to Latin quod. In addition to the causal signification "because", ὅτι may also introduce indirect discourse in Greek, unlike quod in classical Latin. But in late Latin and into the Medieval period quod starts to be used for indirect discourse like ὅτι in Greek.
The nominative is πατήρ (πάτερ is vocative), but yes, they're obviously the same word originally.
Yes, fugio is from the same root.
Same root as fero, again. Interestingly both verbs are also suppletive, though Greek has a higher number of suppletive verbs than does Latin.
Believe it or not, despite the similarity θεός is not etymologically related to Latin deus. Interestingly, the name Ζεύς (gen. Διός) is, however, as is the adjective δῖος,-α,-ον (cf. divus,-a,-um).
Yes, these are both inherited from Proto-Indo-European.
The bilabial nasal 'm' was the original Indo-European masculine/feminine accusative singular marker for all nouns and adjectives, and this remains the case in Latin. At a relatively late (but still prehistoric) period of Proto-Greek it shifted to the alveolar nasal Nu in most positions.
For the third declension, on the other hand, the original 'm' was added directly to consonant stems as a syllabic nasal (like we see in Sanskrit) in Proto-Indo-European, which caused irregularities in some of the daughter languages. Latin just added an epenthetic short vowel 'e', but apparently at an early period in proto-Greek the syllabic 'm' became a short alpha, perhaps originally as a nasalized consonant. By historic times, however, all traces of the original 'm' were lost. Iota stem 3rd declension nouns do have Nu, however, e.g. πόλιν.
The 2nd declension nominative and accusative singular endings in Latin had short 'o' rather than short 'u' in the archaic period. Even in the Republican era -os and -om were retained when the base of the noun or adjective ended in 'u' or 'v', e.g. servŏs for later servus.
Both of these traits are again inherited from Proto-Indo-European, and I think it's a universal characteristic of Indo-European languages that retain the neuter gender for the nominative, accusative, and vocative cases to be identical when neuter, whatever the declension.
Yes, forms of the copula correspond very closely in many Indo-European languages.
"Semantic shift" is a well-established term in linguistics.
Interesting! I'll be on the lookout for that prefix now!Imber Ranae dixit:
That does make more sense. Thanks for pointing it out!
Again, interesting! I knew about the rare usage of "quod" for indirect discourse, but I wasn't sure when it appeared.
Duly noted. Thanks again! I'm curious: do we find the same similarity in other familial terms (e.g. do "mother" and "brother" translate to anything like "ματηρ" and "φρατηρ" respectively)?
Ah, yes. I remember reading something about this on Wikipedia. As I recall, "Iuppiter" is actually a slurring of "deus pater" (or rather, the archaic form thereof).
So they were once even closer. Fascinating!
Thanks again. I thought it might be, but I didn't want anyone quoting me thinking it authoritative when it really wasn't.
Mother is μήτηρ. There's a Greek word φράτηρ that is cognate and may have had the same meaning originally as frater, but the commonest term is ἀδελφός. Soror is ἀδελφή, filia is θυγάτηρ, filius is υἱός. I don't know their etymologies, but they don't look very similar.
I don't know anything about Irish, but I was struck by this phrase I came across:
Tír gan teanga, tír gan anam
It means 'a land without a language, a land without a soul'.
Tír looks similar to terra, teanga to tongue, anam to anima. I shouldn't be too surprised I guess, they're all cousins ultimately.
One a slight side note, the Greeks had names like Περσίᾱ and 'Pώμη. I'm not sure what the Persians called themselves, but the name for Rome at least is a clear transcription from the culture it describes. So where did the Romans get Graecia instead of something like Hellēnia? Does anyone have any insight?
Γραικοί is the name of a Western Greek tribe. Obviously they were the first Greeks who Romans came in contact with, so their name served to denote all the Greeks. It’s not uncommon that a neighbour people is called by the name of an adjacent tribe: e. g. Russian is krievs in Lettish: the latter word derives from krivich, a representative of an Old Russian tribe.Gregorius dixit:
Yes, it's μήτηρ in Attic and Ionic, but μάτηρ in Doric. The eta is due to a sound shift that occurs partially in the Attic dialect and more extensively in the Ionic dialect, whereby long alpha becomes eta in certain positions.Nooj dixit:
The shift does not occur in πατήρ because there the alpha is short, just as the a of Latin păter is short (cf. māter). The inherited difference in vowel quantity also accounts for the different vowel quality in the Germanic languages, as between English "mother" and "father" (OE mōdor/fæder; NHG Mutter/Vater)
It's also interesting that, whereas Greek does not use the descendants of the original Indo-European words for "brother" and "sister", Latin does not use the descendants of the original IE words for "son" and "daughter". Greek replaced the former with ἀδελφός and ἀδελφή, though φράτηρ survives with the new meaning "fellow clansman/one of the same clan". But θυγάτηρ is cognate with English "daughter", and υἱός is ultimately from the same IE root as English "son". Latin, on the other hand, replaced those with filius and filia, but maintained frater and soror which are cognates of English "brother" and "sister", respectively.
It appears you are right. Tír and terra both come from the IE root for dryness *ters-. Latin lingua (Old Latin dingua) is cognate with Irish teanga and English "tongue", from the IE root *dn̥ǵʰwéh₂s of the same meaning. Anam and anima (and likewise animus) are from the IE root *h₂en-, the basic meaning of which is air in motion, hence either breeze/wind (cf. Greek ἄνεμος "wind") or, by association with breath, the soul or spirit.Nooj dixit: