On 11/05/2022 2:03 a.m., Bebercito wrote:
> Le lundi 9 mai 2022 à 22:17:33 UTC+2, Peter T. Daniels a écrit :
>> On Monday, May 9, 2022 at 2:57:20 PM UTC-4, Bebercito wrote:
>>> Le lundi 9 mai 2022 à 19:32:07 UTC+2, Peter T. Daniels a écrit :
>>>> On Monday, May 9, 2022 at 10:22:46 AM UTC-4, Bebercito wrote:
>>
>>>>> Two French words, "charretière" and "jarretière" phonetically differ
>>>>> only by their first consonants being respectively the unvoiced
>>>>> and voiced forms of the same consonant, i.e [ʃ] and [ʒ].
>>>>> Now, the English counterparts of those two words, "carter" and
>>>>> "garter", have the same trait with different first consonants, i.e [k]
>>>>> and [g], where the former is the unvoiced form of the latter.
>>>>> Isn't that weird?
>>>> Not in the slightest. It's called regular sound change.
>>>
>>> But what were the odds that French and English both had
>>> words that sounded alike for two notions as different as
>>> charretière-carter and jarretière-garter, and that the only
>>> phonetic difference in both cases was the (de)voicing
>>> of their initial consonants? Next to zilch, if you ask me.
>> The odds are enormous, since it's highly likely that English borrowed
>> both words from French, as it did thousands of words beginning in
>> 1066 (or maybe even before) and continuing to this day.
>
> No, because that the _two_ words would be borrowed from
> French in itself was highly unlikely (10% of all English words
> are, which makes a probability of 1% for the two words)
This can't be right. If you really want to talk about a "probability" of
a French word being borrowed into English, you would have to divide the
number of words which have actually been borrowed by the total number of
words available. (Both figures likely to be fuzzy in the extreme.) Then
you could square that to get a "probability" of two given words both
being borrowed. But this assumes equi-probability of borrowing for all
words, which is certainly not the case.
and
> that they would then be subjected to the same sound change with
> different consonants than in French was even more so.
In fact it is the French words that have changed. And there is nothing
improbable in the changes of k- > ch- and g- > j- in the same
environment. It's typical regular sound change, which doesn't apply to
words here and there at random.
Now the relation between the French and English words is a little more
complicated than has been recognized here so far.
Old French has both gartier and jartier. Standard French continues the
latter, but English has borrowed from a dialect which has the former.
(The root is apparently Celtic *gar for a part of the leg.)
Charretier* could be directly related to carter, if English had borrowed
from a dialect with ca- corresponding to cha-; and Norman French would
be such a dialect (compare castle/chateau, catch/chase). However, carter
seems to appear earlier in English than chartier does in French, and OED
sees it as formed from cart + er in English (13th century). And "cart"
itself seems to be immediately from Old Norse kart-r. Quite possibly all
these words ultimately go back to Latin carrus or its Celtic antecedent,
but that's where I'll stop.
*You gave us charretière, but that would be a female cart-driver, no?
Oh by the way -- OED online is so full of stuff -- English did actually
borrow both charet(te) for the vehicle and chareter for the driver, from
Standard French, but neither survived into modern times.
If "the odds
> are enormous", there should be dozens of other such French-English
> pairs of words. Can you name just one?
Why should there be dozens? And why should we have to search for them?
How long a list of French ch-/j- pairs do you have?
Summary: Yes, it's a coincidence, but not one I find of any linguistic
interest.