What makes French a Latin-Germanic mixed language

<< That is not an answer. I can tell you for English--it's about 10 Celtic words (not counting recent Irish borrowings), that's it. >>

-- Great, then I know all Gaulish words passed to English:

- mutton
- buck, butcher
- mare, marshall
- car, to carry
- ton, (metric) tonne, tun
- town
- battle, to batter, combat, to beat
- cavalier, chivalry
- gable, javelin
- lance, lancer, to launch
- embassy, ambassador
- league (distance)
- yew etc...

- mutton <Yes>
- buck, butcher <"Buck" <No>, "Butcher" <Maybe>>
- mare, marshall <"Mare"<No>, "Marshal<No>>
- car, to carry <Yes>
- ton, (metric) tonne, tun <No>
- town <No>
- battle, to batter, combat, to beat <"battle, better, combat" <Yes>, "beat" <No>>
- cavalier, chivalry <Yes, Yes>
- gable, javelin <"gable"<No>, "Javelin"<Yes>>
- lance, lancer, to launch <Yes, Yes, Yes>
- embassy, ambassador <Yes, Yes>
- league (distance) <Yes>
- yew etc... <No>

Many of the words cited, in their French equivalents, are Celtic ("butcher" is a good byspell) but the English words 1). didn't come from French ("buck" < Anglo-Saxon bocca) and are native germanic words. The English word "yew" is also native English, not coming through French by directly from AS. "Mare", "town", "beat" likewise...

But I see your point regarding the Gaulish. Yes, many entered French even through Vulgar Latin.

Others, like "pièce", "bijou" were borrowed later from Celtic (Gaulish; Breton repectively)

<< regarding the Gaulish. Yes, many entered French even through Vulgar Latin. >>

-- ... or the other way round?
And did the alleged "Vulgar Latin" ever exist?

This might be just me but I have a feeling that:

- so-called Romance was already widely present as a lingua franca in Gallia long before the first Roman soldier crossed the Alps,
- the process of substitution (or derivation) from Gaulish to proto-Romance took a thousand years or probably much more, (*)
- related forms of Proto-Romance extended simultaneaously into Hispania and even Italy, ultimately replacing Latin itself.

As for "town", "to beat", "mare" and others, orthodox linguists have a hard time finding satisfactory etymologies in Old Germanic and/or Old Norse (similarly, fishy or downright nonsensical Latinate origins are often offered for French words that obviously always were here). For example "town" is clearly a cognate of NL "tuin" (garden), DE "Zaun" (fence, enclosure), probably originating from Gaulish "Dun-" (enclosed town) which spread into Continental Western Germanic (but NOT into Northern Germanic).
BTW the semantical kinship between "city" and "garden" is a classic, cf. 'garden' and 'yard' related to Slavic 'gorod' and 'grad'.

(*) An awesome read on the topic of inherent stability of languages:
http://www.applied-epistemology.com/phpbb2/viewtopic.php?t=25&postdays=0&postorder=asc&start=0

Often a tad facetious but loads of food for thought.

"tun" is not Gaulish word-it means fenced area in Swedish ,homestead in Old Norse.
"marshal" is not Gaulish word too-Old High German marah "horse" (related to English "mare"),schalh-"servant" (stable keeper)
http://en.wikipedia.org/wiki/Marshal

<< Old High German marah "horse" >>
-- Derivates of "marah" are less present in Germanic languages than in the Celtic ones (compared to the more successful cognates of "horse", "häst", "Pferd", "paard" etc.). "Marc'h" is still the dominant word for "horse" in Breton.

<< "tun" is not Gaulish word-it means fenced area in Swedish ,homestead in Old Norse >>
-- There are indeed toponyms in "-tun-" in Sweden (especially in Uppland: Sigtuna, Vallentuna, Sollentuna and so forth) with the meaning "trädgård", "enclosed yard" (exactly like NL "tuin"), but I do not think it ever was as general in Scandinavia as "dun-" and cognates in Celtic languages, specifically with the meaning "town".

A large part of Gaulish/Celtic vocabulary is strikingly similar to either Germanic or Latin words (which was maybe the reason why it was so easily absorbed and/or replaced by both) :
- "viros": like Lat. "vir" (man)
- "tarvos" (bull): like Latin "taurus"
- "novios" (new)
- "jovincos" (young): like Lat. "juvenus"
- "strimon" (river): like "Strom", "stream"
- "briva" (bridge)
- "briga" (hill): like "Berg"
- "vectos" (fight): like "Gefecht"
- "hall-" (salt): like in Old German
- "isarn-" (iron)
- "oxos" (ox)
- and thousand others

This is always the big question with ancient Celtic languages: who borrowed from whom?

Parisien<<<- so-called Romance was already widely present as a lingua franca in Gallia long before the first Roman soldier crossed the Alps, >>>


It is true that in Greek empire everywhere pure greek was spoken. Why do we assume that in Roman empire everywhere forms of vulgar Latin (=Romance) were spoken? Any facts? Or just guessing...?

Gaulish(Celtic) ,Latin (Romance),Germanic languages are of Indo-European origin.They have a lot of similar words.
For example , English "wolf" in Mazandarani (Iranian language) and Swedish (Germanic language) is "varg", English "brother","mother" in Pashto (Iranian)-"wror",in Swedish,Norwegian-"bror","mor"; English "daughter" in Balochi (Iranian) is "dohtir",in Swedish-dottir,in Dutch,Low German-"dochter", English "name" in Zazaki (Iranian) is "name", English "three" in Pashto is "dre",in German "drei",English "better" in Mazandarani is "better" and so on.
http://en.wikipedia.org/wiki/Iranian_languages
http://en.wikipedia.org/wiki/Mazandarani_language

"Mother" in Pashto (Iranian) is "mor",like in Swedish and Norwegian.

<<Derivates of "marah" are less present in Germanic languages than in the Celtic ones (compared to the more successful cognates of "horse", "häst", "Pferd", "paard" etc.). "Marc'h" is still the dominant word for "horse" in Breton.
>>

PARISIEN,

It appears you are citing Celtic & Germanic cognates, inherited on both sides from PIE, and then ascribing the Germanic cognate with a Celtic origin.

The English 'mare' is found in many IE languages:

Mare\ (m[^a]r), n. [OE. mere, AS. mere, myre, fem of AS. mearh horse, akin to D. merrie mare, G. m["a]hre, OHG. marah horse, meriha mare, Icel. marr horse, OCelt. marka (Pausan. 19, 19,4), Ir. marc, W. march. Cf. Marshal.] The female of the horse and other equine quadrupeds.--Webster's Revised Unabridged Dictionary, © 1996, 1998 MICRA, Inc.

Because it has a Celtic cognate does not mean the English word 'Mare' comes from Celtic. It comes from IE through Proto-Germanic into Old English as mere/mearh ("horse").

Same for 'town'

Town\, n. [OE. toun, tun, AS. tun inclosure, fence, village, town; akin to D. tuin a garden, G. zaun a hadge, fence, OHG. zun, Icel. tun an inclosure, homestead, house, Ir. & Gael. dun a fortress, W. din. Cf. Down, adv. & prep., Dune, tine to inclose.]

Although conjectured as a possible early borrowing from Celtic, this has never been confirmed. The shift of d- to t- certainly lends credence to the fact that it is either a cognate, or was borrowed so early as to not count as a borrowing (i.e. Celtic and Germanic were dialects of IE at the time, so in essence were still the same language)

'beat' is altogether different. It is defintely a confirmed germanic word:

Origin:
bef. 900; ME beten, OE bēatan; c. ON bauta, MLG bōten, OHG bōzzan; akin to MIr búalaim I hit, L fūstis a stick < *bheud-

O.E. beatan "inflict blows on, thrash" (class VII strong verb; past tense beot, pp. beaten), from P.Gmc. *bautan (cf. O.N. bauta, O.H.G. bozan), from PIE base *bhau- "to strike"

two etymologies, PIE *bheud- & *bhau-: either way, clearly through P.Germanic

<< It is true that in Greek empire everywhere pure greek was spoken. Why do we assume that in Roman empire everywhere forms of vulgar Latin (=Romance) were spoken? Any facts? Or just guessing...? >>

-- Facts:

1. 2000 years ago in the Middle East,
- social elites and scholars all knew Greek,
- most popular vernacular was Aramaic,
- nearly extinct languages like Hebrew were mostly used for liturgical scriptures and person names.

(I assume we had a similar situation in Gallia under the Roman Empire, with Latin instead of Greek, Romance instead of Aramaic, and Gaulish instead of Hebrew).

2. Pure Greek was spoken since present day Greek remains widely similar to Ancient Greek.

3. Vulgar Latin is a conjecture, while Romance languages are solid realities. They haven't evolved much since their first written apparitions.

<<<<Derivates of "marah" are less present in Germanic languages than in the Celtic ones (compared to the more successful cognates of "horse", "häst", "Pferd", "paard" etc.). "Marc'h" is still the dominant word for "horse" in Breton.
>>>>

One further interesting note about 'mare'. It's IE root *marko is seen only in Gmc & Celt in the West, however, it has cognates in other linguistic families in the east (cf. Chinese 'ma'; Korean 'mar' "horse") brought to the region by the horse cultures of western and central asia.

In this scheme of events, it seems more plausible that it be a Germanic > Celtic loan (if even a loan at all. I earnestly doubt it) as opposed to the other way around, being that the ancestors of today's germanic languages were closer to the region than those of surving Celtic languages (not counting ancient Celtic, which had no real contact with modern celtic after their initial split from them, hence no passing of etymons)

However, I believe that the word is cognate, and shows the shift of -k- to -h- in Germanic as expected.

What makes French a Latin Germanic mixture? Nothing. It's a Latin language

What makes French a Latin Germanic mixture? Frankisch. It's a German language

"What makes French a Latin Germanic mixture? Frankisch. It's a German language"

Where is the proof for this?

What makes French a Latin language? Latin. It's a Latin language