Sorites Paradox and Ahistorical Accretion

Martin E. HULD ylfenn at
Thu Mar 11 13:32:32 UTC 1999

----------------------------Original message----------------------------
Mark Hubey writes:
There are certain things that really get me confused ... I will demonstrate
via ... yoke.

Yes, you are really confused.  The PIE etymon *iugwom 'yoke' is not an
isolated reconstruction, but is an specific derivational case of the root
*ieugw- 'join'.  Because English employs zero-derivation (a talk, to talk, a
drive, to drive, a fax to fax ...) derivational processed are opaque to the
monolingual speaker, but IE possessed several devices for deriving nouns
from existing roots; *iugwom 'yoke' [Lat jugum, OE geoc, OCS igo, Gk. zugon,
Arm. louc, OIn. yugam] is specifically the zero-grade of the root with an
accented thematic vowel to make a neuter noun of instrument, 'the thing
which yokes'; a verb can also be derived from the root by adding an .ne.-
infix to the zero grade, thus * [Lat. jungunt, OIn yunjanti]
(Lat. had employed the thematic ending -onti, Gk. has further restructured
the cognate as zeugnuasi.  Thus, PIE *iugwom [in preference to other
reconstructions yugom vel sim.) is not without internal cognates within PIE.
 The term for the technological innovation of the yoke can be seen as built
by regular processes from the resources of the language just as the word
{microprocessor} has been.  I don't believe the latter is in any sense
restricted to 'scientists', but is the property of all users of English
morphemes as *iugwom was accessible by all users of PIE morphemes.

To get to your second point, "is it accidental that the word for
yoking/hitching is <cek> ...?"  What do you mean by accidental?  Are you
suggesting that God or Joseph Stalin or some other deity had some ulterior
purpose in mind or that the semantic content of the morpheme is somehow
responsible for the phonetic features of the form?  If so, that is
unscientific thinking.  Do you perhaps mean to ask if the phonetic
similarities between the Turkic morphemes and the IE morpheme *ieugw- are
indicative of a relationship?  The answer is there is no way to tell from
the limited data of one case.  Are the consonant variations c, ch, y typical
of Turkic cognate sets as the equation of Lat j-. OE ge- OInd. y- and Gk z-
are of IE?  Are the vowel variations e/U regular patterns as the IE
eu::ou::u series is?  Is the semantic discrepancy significant?  Is there
reason to believe that IE stem final -gw is connected with Turkic stem final
-k?  Is it an 'accident' -in your sense- that Alb. ju (IPA ju:]) and Lith.
jus (IPA [ju:s]) both signify the second person plural pronoun (Fr. vous,
Russ vy)?  The answer is yes; despite apparent phonetic and semantic
similarity, Alb. ju must be from PIE *uos/ues and cognate with R. vy.  gjesh
'boil' < PIE *ies- (Gk zeo:), gjesh 'gird' < PIE *ioHs- (Lith juos-iu), and
gjuaj 'hunt' < PIE *ieAgh-ni- (NHG jagen) show that the reflex of PIE
initial *i- in Alb. is gj-.

There are a series of tests for identifying loan words, one of which, as you
have guessed, is whether the word in question can be shown to be part of a
morphological family rather than an isolated structure.  Yes, we think of
these things all the time and try to avoid snap judgments based on single,
superficial coincidences or even a large collection of unrelated,
superficial coincidences, in the belief that, as with UFO sightings,
1,000,000 times 0 is still 0.
Name: Martin E. HULD
E-mail: Martin E. HULD <mhuld at>
Date: 03/11/1999
Time: 00:26:18

This message sent by NetManage's award winning
standards based e-mail client Z-Mail Pro
NetManage - Complete PC Connectivity Solutions

More information about the Histling mailing list