7.1164, Sum: TH-substitution

The Linguist List linguist at tam2000.tamu.edu
Sat Aug 17 21:02:45 UTC 1996

LINGUIST List:  Vol-7-1164. Sat Aug 17 1996. ISSN: 1068-4875. Lines:  193
Subject: 7.1164, Sum: TH-substitution
Moderators: Anthony Rodrigues Aristar: Texas A&M U. <aristar at tam2000.tamu.edu>
            Helen Dry: Eastern Michigan U. <hdry at emunix.emich.edu> (On Leave)
            T. Daniel Seely: Eastern Michigan U. <dseely at emunix.emich.edu>
Associate Editor:  Ljuba Veselinova <lveselin at emunix.emich.edu>
Assistant Editors: Ron Reck <rreck at emunix.emich.edu>
                   Ann Dizdar <dizdar at tam2000.tamu.edu>
                   Annemarie Valdez <avaldez at emunix.emich.edu>
Software development: John H. Remmers <remmers at emunix.emich.edu>
Editor for this issue: lveselin at emunix.emich.edu (Ljuba Veselinova)
Date:  Sat, 17 Aug 1996 14:07:38 +0800
From:  cpeust at gwdg.de
Subject:  TH-substitution
Date:  Sat, 17 Aug 1996 14:07:38 +0800
From:  cpeust at gwdg.de
Subject:  TH-substitution
This is a revised version of my summary on th-substitution which
I posted a while ago. I have written this update since
1) my first version was a bit uncomfortable to read and
2) I received some additional material.
I had posted the following question on Linguist List:
Dear list members,
Most languages appear to lack a dental spirant as the english th is (I
mean basically the voiceless variant here), so when speakers of such
languages try to imitate such a spirant they are likely to replace it
by something else. But by what? German natives generally tend to use
/s/ instead when speaking English improperly, while it can be observed
that in Russian, at least at an earlier period, /f/ was used, so
Athenes, mythos etc. from Greek were taken over as Afina, mif etc. In
Arabic, the classical th-sound developed into /t/ in the modern
dialects but today the same sound is regularly substituted as /s/ in
secondary loans from Classical Arabic. Ancient Egyptian obviously
replaced the sound by /t/ (e.g. in the Persian name Mithras).  I
wonder whether the choice of t, s or f respectively can tell us
something about the internal structure of the phonemic system of the
receiving language.  I would be glad if you 1) could give me more
facts about th-substitution in various languages 2) know of a
treatment of this question in the literature.  I will post a summary
on the list if I receive enough answers.  Thank you,
I received a lot of answers and am very glad about the interest Thanks
a lot to all who took time to write! My thanks go to:
Theriault Alain       theriaal at ERE.UMontreal.CA
Jorge Baptista        jbaptis at mozart.si.ualg.pt
Karen S. Chung       karchung at ccms.ntu.edu.tw
Ivan A Derzhanski    iad at banmatpc.math.acad.bg
Malin Ericson        malin at ling.su.se
JOSEPH F. FOSTER     Joseph.Foster at UC.Edu
Osamu Fujimura       fujimura.1 at osu.edu
Sandra Golstein      sandra at tovna.co.il
Mark H.-M. Hansheng  mhansell at carleton.edu
James Kirchner       JPKIRCHNER at aol.com
Roger Lass            ROGER at beattie.uct.ac.za
Stephen G. Lambacher  steeve at u-aizu.ac.jp
Wenchao Li            wcli at vax.ox.ac.uk
Daniel Loehr        loehrd at gusun.georgetown.edu
Waruno Mahdi        waruno at fritz-haber-institut.mpg.de
Larry Mitchell      j-mitchell at tamu.edu
Bente H. Moxness    benmox at alfa.itea.unit.no
France Mugler       MUGLER_F at usp.ac.fj
Caoimhin P. ODonnaile  caoimhin at smo.uhi.ac.uk
MARC PICARD         PICARD at vax2.concordia.ca
Ingo Plag           plag at Mailer.Uni-Marburg.DE
Charles Rowe        rowe at email.unc.edu
GAVIN O SHEA        GOSHEA at acadamh.ucd.ie
Laura Catharine Smith       lcsmith at acs.ucalgary.ca
Yasuhiko Sukegawa          suke at insc.tohoku.ac.jp
Allison Mary Teasdale       amteasda at acs.ucalgary.ca
Gary H. Toops:       TOOPS at TWSUVM.UC.TWSU.EDU
John Verhaar         101457.3114 at CompuServe.COM
Guido Vanden Wyngaerd
Guido=Vanden=Wyngaerd%OWP%UFSAL at ufsal3.kubrussel.ac.be
The results are as follows, given in short form:
Afrikaans: f and v
Bulgarian: t and d
Chinese from Hong Kong: f and n
Chinese from Malaysia/Singapore: t and d
Chinese from Taiwan: s and l
Czech: t and d, more rarely f and v
Dutch: t and d
Fijian: Fijian has only a voiced dental spirant, written 'c', which
    is also used to substitute the English voiceless dental spirant
French: in France s and z, in Canada t and d
     (this interesting fact seems very certain since a lot of
	people told me so)
German: s and z; in Austria s and d which is easy to understand since
             the Austrian vernacular has no /z/.
Hebrew: s and d
Hindi: th (aspirated plosive) (no information on the voiced variant)
Hungarian: s (written sz) (no information on the voiced variant)
Icelandic: no substitution since th-sounds exist in Icelandic
Japanese: s and z, also ts and dz are possible which seem to gain
              phonematic character in the current language
Norwegian: t and d
Old Church Slavonic: f (no information on the voiced variant).
      Through this language some cases like Feodor for Theodoros
      entered Russian but in modern Russian the normal substitutions
      are t and d
Polish: t and d
Portuguese: s and d
Russian: t and d
Spanish: t and d, although a voiceless th-sound (c, z) exists at
	least in many ialects!
Swedish: s and d respectively. The assymetry is due to the fact
                  Swedish has no /z/.
Turkish: t and d
Yiddish: t (no information for voiced th)
Several people also told me of special reflexes of th in English
f and v respectively in Cockney
also in Southern or Black American English, but subject to
restrictions t and d respectively in New York and parts of Ireland,
these sounds seem not to be identical to etymological t and d
I also heard that British children who not yet master the (voiceless)
th-sound substitute it by f.
Now some additional information I got:
Laura Catharine Smith is preparing a paper on this topic which will
appear in the Calgary Working Papers in Linguistics' next volume to
appear in January of 1997.
Allison Mary Teasdale is also working on this subject; she recently
posted a query on linguist list where her approach is described.
Not long ago, Wenchao Li asked about historical developments of
th-spirants and posted a summary on Linguist List, titled "dental
spirant summary", on Fri, 26 Jul 1996 17:03:19 +0100
It seems very remarkable that - closely related languages can differ
markedly in their practice of substitution, cf. French of France (s-z)
and Canada (t-d), German (s-z), Dutch/Yiddish (t-d) and Afrikaans
(f-v) or the Chinese data.  - The substitutions are sometimes
assymetrical, cf. e.g. Swedish and Hebrew, which seems very
interesting but not always easy to explain.
Let me add additional observations from Ancient Egyptian: There is
quite a lot of Semitic loan words in Egyptian around 1200 or 1000
B.C. Here the voiceless dental spirant of Semitic is rendered
regularly as /s/, the voiced by /t-/ or /d-/ in about equal
distribution. /t-/ and /d-/ are palatal stops the exact difference
between them being a matter of dispute (in my opinion aspirate -
non-aspirate).  The substitution of the voiced th seems very
interesting; I would not exclude that it was an affricate in the
source language. Let me add that Egyptian had no voiced /z/ and not
even a voiced /d/ at this time.  Although some Semitic languages
confuse the th-sounds with /s/ and /z/ respectively, this was probably
not true of the source languages of the loan words since Semitic s or
z is usually rendered differently: by /sh/ (palatal fricative) and
/d-/ (never /t-/!) respectively. It is well known that Semitic /z/ was
originally an affricate.  (I have used material from James E. Hoch,
Semitic Words in Egyptian Texts of the New Kingdom, Princeton 1994)
>>From Ptolemaic times, however, we know that Persian names with
/th/-fricative were substituted with /t/ (= aspirate t). This is also
consistently true of Greek theta of which it is unknown for how long
it remained a plosive and when it turned into a fricative as in modern
Greek.  So Egyptian changed its habits of th-substitution in less than
1000 years. (Well, I cannot tell whether the Persian names might have
been intermediated to the scribes by some Semitic or other native
speakers; we have to be very careful in such matters if extinct
languages are concerned).
Carsten Peust
Seminar of Egyptology and Coptology
cpeust at gwdu20.gwdg.de or cpeust at gwdg.de
LINGUIST List: Vol-7-1164.

More information about the Linguist mailing list