LINGUIST List 14.1916

Sun Jul 13 2003

Sum: VVCC Superheavy Syllables

Editor for this issue: Karen Milligan <>


  1. Andrew Horne, Summary: VVCC superheavy syllable

Message 1: Summary: VVCC superheavy syllable

Date: Sun, 13 Jul 2003 14:55:21 +0000
From: Andrew Horne <>
Subject: Summary: VVCC superheavy syllable

Two weeks ago (Linguist 14.1796) I posted a question requesting
information on langauges which have underlying VVCC superheavy
syllables in monomorphemic forms.

The responses were extremely helpful and I would like to thank the
following persons who responded to my query:

Edward Vajda (EV)
S�ren Wichmann (SW)
Charles E. Cairns (CEC)
John Phillips (JP)
Yuni Kim (YK)
Michael Johnstone (MJ)
Willem Visser (WV)
Katalin Balogne Berces (KBB)


I am looking for languages with morphologically simplex superheavy
syllables of the format VVCC e.g. English bold, find etc. This
structure appears to be extremely rare cross-linguistically and many
apparent cases turn out to involve either morphological boundaries (as
in dine-d) or alternating epenthetic forms i.e. (arguably) an
underlying VVCVC structure e,g. Somali maalim day, maalm-o days. I
would be grateful for information on any languages which have
(underlying) VVCC superheavies and for which, ideally, data is readily
available. I am particularly interested in any correspondence between
the occurrence of such forms and phonetically identical
morphologically complex forms e.g. English bold v. bowl-ed; find
v. fine-d etc.

EV: I am a specialist of the Ket language isolate spoken in Central
SiberiaKet phonology does resemble English in having such
codas as you describe. While there are no morphological word-initial
or even morpheme initial consonant clusters allowed in Ket, there are
many words (and monomorphemic words) that do end with two

there are quite a number of monomorphemic Ket
monosyllables that have a rhyme containing a superheavy coda followed
by geminate or a half-long vowel (Ket is tonal, and the three vowel
types: regular, half-long, geminate are actually non-melodic
concomitants to the tone -- I wrote all this up in a short monograph
''Ket prosodic phonology'' published by Lincom Europa. Anyway, here is
an example: biilt 'with rising falling tone on the geminate /i/. This
word means 'sand martin' (at type of riverbank-nesting bird in Central
Siberia). There is no way of showing that the final /t/ is or ever was
a separate morpheme.

My full-length Lincom Europa grammar of Ket is is the
process of being written. I would be happy to send you pdf files of
the phonology section if this would help answer questions (though
unfortunately I myself didn't pose any special questions regarding
coda structure).

SW: concerning your inquiry on Linguist, in the Mixean subgroup of
Mixe-Zoquean languages you will find languages that have the
''superheavy'' syllables. In my book The Relationship among the
Mixe-Zoquean Languages of Mexico (University of Utah Press, 1995)
there are synchronic as well as diachronic descriptions (if it is not
in a library near you there are $10 offers on the internet). The
majority of diachronic phonological developments in the family relate
to quantity--either of the syllable nucleus or the final
consonants. In some Mixean dialects length is reduced before clusters
of consonants, but one also finds extra-heavy syllables emerging or
reemerging as a result of disyllables loosing their second vowel. The
book has tons of data, so you should be able to find something useful

CEC: In the case of English, I am convinced that in fact in words like
those you cite that the segments that make the syllable appear
''superheavy'' are in fact not part of a syllable at all. Either they
are directly dependent on the Prosodic Word, or they are ''extra
metrical'' in a Hallean sense. My reasons are that, word internally,
-VV or -VC seem to be the maximal size of the rhyme (see Borowsky's
work from the late 1980s); the only counterexamples are a few words
like ''council,'' where the segment that makes the syllable appear
longer than -VX is always a nasal homorganic to the following stop.

JP: You'll find plenty of these in Welsh, e.g. 

Aifft (Egypt) 
enghraifft (an example)
aillt (a serf) 
cainc (a tune) 
braint (priviledge) 
sawdl (heel) 
llawdr (a pair of trousers) 
dieithr (strange) 
meistr (master) 
meirw (dead) 
meirch (horses) 
beirdd (poets) 

(The last 3 are plurals formed by vowel alternation, like English foot-feet.) 

It should be easy enough to find or make morphologically complex forms
with the same consonant clusters, e.g. blawdlyd (floury) from blawd
(flour) + -llyd (-y); llawdrwm (heavy-handed) from llaw (hand) + trwm
(heavy). Some of the examples above came to mind immediately, others I
found flicking through a rhyming dictionary, Roy Stephens: Yr
Odliadur, Gwasg Gomer 1978.

YK: I saw your posting on the Linguist List and thought I would write
to you about a dialect of Finland-Swedish I'm working on, which has
superheavy syllables in both monomorphemic and multimorphemic
words. The dialect is East Nyland Swedish (ENS), spoken along the
southern coast of Finland east of Helsinki. There are two kinds of
VVCC syllables. The first type is long vowel + consonant cluster.
/la:mb/ 'lamb' /ha:nd/ 'hand' /bry:st/ 'breast' /ru:nst-krank/
(compound word; adjective referring to hip deformity)

Standard Swedish has this syllable type also, in words like /mo:ln/
'cloud', although it's rarer than in ENS (the example words here have
short vowels in Standard Swedish). This type seems to behave as
regular-heavy, rather than superheavy (as far as I know nongeminate
coda consonants don't contribute to syllable weight in Swedish). Also,
many dialects that have these VVCCs don't allow ''real'' superheavy
syllables, which can be defined as long vowel + geminate consonant.

ENS has long vowel + geminate syllables (also diphthong +
geminate). In monomorphemes: /tve:tt/ 'to wash', inf. -- alternates
with /tve:tta/ /ly:ss/ 'to listen' /tr�:ttog-�r/ 'tired'
/sveittog-�r/ 'sweaty' (not sure how the morphology works on
these last two)

..Two possible sources on Scandinavian are Arnason (1980)
'Quantity in phonological theory' and Riad (1992) 'Structures in
Germanic prosody'. For Finland-Swedish dialects (ENS is not the only
one with superheavy syllables), Kiparsky has a paper on his webpage.

MJ: I haven't sat and thought in depth about this, but Hungarian seems
to have plenty of VVCC syllables with long vowels (phonologically, I'd
argue that Hungarian doesn't have diphthongs). For example,

e'szt /e:st/ 'Estonian' cf. ve'szt /ve:s+t/ 'disaster'+ACC

ke'nt /ke:nt/ 'as' cf. ke'nt /ke:n+t/ 'sulphur'+ACC

ime'nt /ime:nt/ 'just now' (quite a few adverbs end with -e'nt or
-int. I'm not sure of the origin, but may well be a combination of
suffixes, historically.)

sa'nc /Sa:nts/ 'earthwork' (loanword, cf. German Schanze) 

aja'nl /aja:nl/ 'recommend' (the only word I know of ending with /nl/) 

e'rv /e:rv/ 'argument' 

o'zd /o:zd/ 'name of a town' 

pe'nz /pe:nz/ 'money' 

ta'rs /ta:rS/ 'companion' 

se'rv /Se:rv/ 'hernia' (but se'r- 'injure' exists as a bound morpheme) 

keno~cs /ken�:tS/ 'ointment' (though /tS/ counts as a single phoneme) 

fogo'dzkodik /fog+o:dz+kod+ik/ 'cling onto'

WV: I think (West) Frisian, my mother tongue, might be of interest to
you. Frisian is a West germanic language, spoken by about 450,000 of
the 600,000 inhabitants of Frysl�n/Friesland, a province in the
North West of the Netherlands. In section 2.2.5 (The shape of the
rhyme) of the second chapter of my doctoral dissertation, I listed all
words I was able to find with underlying fourpositional rhymes,
i.e. not only VVCC but also VCCC. Most of these end in a coronal
segment, as expected, but here are also some ending in a dorsal one. I
give you the full reference: Willem Visser, The Syllable in Frisian
(Hill Dissertations; 30), The Hague, 1997, ISBN: 90-5569-030-9

KBB: The Hungarian vowel inventory consists of the following vowels:
/a, e, i, o, �, u, �/, and each has a long
counterpart. However, with a few exceptions (e.g. _t[o:s]t_ 'toast'),
tautomorphemic superheavy syllables can only contain /a:/ or /e:/, as
in _f[e:]rc_ 'tack', _m[a:]rt_ 'dip'. There are apparent
counterexamples with [i:, u:, �:, �:], but (i) there's
few of them, (ii) these vowels are usually shortened. What is
interesting is the fact that these 2 ''misbehaving'' vowels, /a:, e:/,
are the ones whose short and long counterparts considerably differ
phonetically. Short /e/ is open-mid whereas long /e:/ is close-mid,
short /a/ is back whereas long /a:/ is front/central. For more
information, I can refer you to Peter Siptar and Miklos Torkenczy, The
Phonology of Hungarian (in the Phonology of the World's Languages

Thanks again to everyone.

Andrew Horne
Graduate student
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue