: German compounds

[Robert Whiting]

On Thu, 15 Apr 1999 CONNOLLY at LATTE.MEMPHIS.EDU wrote:

>Peter wrote:

>>Bob mentioned the German "Handschuh" as a light-hearted example
>>of compounding from language poverty.

>It's not, actually.  My Kluge, _Etymologisches Woerterbuch der
>deutschen Sprache_, 21st edition (1975), say the following (my
>translation of German original):

>     Germanic *_andasko:haz_ 'countershoe' [Kluge glosses as
>     _Gegenschuh_, the meaning of which is at least very
>     obscure], also found in the Old English given name
>     _Andsce:oh_, has been reinterpreted into Old High German
>     _hantscuoh_ [hand-shoe]; the place name _Handschuhsheim_ is
>     also related.

>Place names in -_heim_ often have a personal name as the first
>component. Kluge adds that there was at least one Germanic word
>word for glove, the form of which must have been approximately
>*_wand_-, referring to something knitted ("wound"), with a
>possible second one *_skinthaz_ 'hide' borrowed into Finnish to
>yield modern _kinnas_ (genitive _kintaan_) 'glove'.  If Kluge is
>right, _Handschuh_ represents not compounding for lack of a
>proper word, but rather folk etymology.

I find this all a bit difficult to accept uncritically.  First,
Buck, DSS, doesn't have any problem with a compound of 'hand'
and 'shoe'.  Although this is now 50 years old, here is what he
has to say (p. 435, 6.58 glove):

     OHG hantscuoh, NHG handschuh (OE handscio: only a proper
     name; see now Bosworth-Toller, Suppl. s.v.), Du. handschoen,
     MLG hantsche (> late ON hanzki, Dan., Sw, handske), cpd. of
     words for 'hand' and 'shoe'.  Falk-Torp 380.

Of course the problem is that we don't have the word in Gothic so
the developments become speculative.  But if we consider that OHG
hantscuoh is a folk etymology then either all the other forms are
loans (or at least reconvergence) from OHG (not impossible,
especially as Buck sees the Scandinavian word coming from MLG) or
all the other branches used the same folk etymology (unlikely).
Accepting 'hand' + 'shoe' just seems more likely all around.

Second, I am curious about the discrepancy between the two forms
(Andsce:oh in Kluge and handscio: in Buck (and presumably in
Toller, although I can't check it), since Buck's seems to confirm
*hand- rather than *anda-.

Third, both 'hand' and 'shoe' are PIE words so there is no
difficulty getting them into a much earlier stage of Germanic
than OHG.  *_andasko:haz should mean 'other shoe' so it is less
specific than "handshoe."  And if 'shoe' is correctly
etymologized as coming from the PIE root 'to cover' (ultimately
the same root that produced English 'sky') then the original
meaning is "covering" and the compound "handcovering" sounds less
risible than "handshoe" and less unlikely than "other covering."
"Shoe"/"handshoe" would then just be unmarked/marked
counterparts; "shoe" being the more common (all God's chillun got
shoes), "handshoe" gets the marking (rather than the other way
around "shoe"/"footshoe").

Fourth, words for 'glove/mitten' in IE languages often have the
word for 'hand' in them [subconscious-level double entendre,
discovered on proofreading], e.g., Gk. xeiris, Lat. manica, Russ.
rukavic'a, etc., so there is nothing a priori implausible about
"handshoe" as a coining for 'glove', especially if "shoe" still
had the meaning 'covering'.

All in all, I find less difficulty with an original 'hand' +
'shoe' than with a folk-etymologized word for 'glove' that did
not already contain the word for 'hand'.  I also think that the
word was likely to have been older than OHG.  I would even go so
far as to suggest that there was originally a pair, "handshoe"
and *"footshoe," meaning "hand covering" and "foot covering"
respectively, and that *"footshoe," through frequent use, was
shortened to "shoe" while "handshoe" remained.  The change from
*"footshoe" to "shoe" would have happened already in
Proto-Germanic since all the Germanic languages (including
Gothic) have only the "shoe" word.

To me, Kluge's explanation sounds like an attempt to get rid of
a form, Handschuh, that has often been a source of amusement.
But in the final analysis I think that the form is a quite
logical development that has just happened to leave a humorous
relic.  It is fortunate that the words for 'hat/cap' developed
from a different IE root for 'cover, protect' else German might
have ended up with "headshoe"* which would have been a perennial
gag writer's delight.

I will agree, however, that German "Handschuh" does not represent
compounding for lack of a proper word; but, rather than folk
etymology, it was simply a logical development.  Please remember
that the purpose of the original posting was to make fun of the
idea that Proto-Germanic was linguistically "poor" or "primitive,"
not to prove that it was actually the case.

The other Germanic words for 'glove/mitten' are interesting as
well (ON vo,ttr ~ EFris. want/wante, "mitten"; ON glo:fi ~ OE glo:f,
"glove"), but this is long enough already.  I will only say that
want/wante is more likely to refer to a 'winding' than to
something knitted, and was apparently strips of cloth wrapped
around the hand to protect it from sword-blows (cf. Buck. loc.
cit.).  This word went from Frankish (wanth) to OFrench as
want/guant/gant and then from MFrench into English as gantelet
(diminutive) to become modern English gauntlet.

<snip>

>And really finally, _Broetchen_ is *not* a compound, since
>-_chen_ is merely a suffix.

Don't be too quick to dismiss "mere suffixes" from compounds.
The process by which separate words become suffixes or case
endings through compounding or cliticization is called
grammaticalization and is quite common.  I don't know about
-chen, but many forms that look like "mere suffixes" are
originally separate words.  In German drittel, viertel, etc.,
the -tel looks like a suffix, but it is actually a form of
Teil, thus "third part," "fourth part," etc.  English -hood
(~ German -heit/-keit) was once a separate word (ha:d) meaning
"state, rank, position" but now is only a noun-formative suffix.
The list of examples is very long.

<snip>

Bob Whiting
whiting at cc.helsinki.fi