LINGUIST List 8.1331

Sun Sep 21 1997

Sum: Tones and Morphemes in Dagbani

Editor for this issue: Martin Jacobsen <>


  1. Knut Olawsky, Sum: Tones and morphemes in Dagbani

Message 1: Sum: Tones and morphemes in Dagbani

Date: Thu, 18 Sep 97 12:28:02 +0200
From: Knut Olawsky <>
Subject: Sum: Tones and morphemes in Dagbani

Quite some time ago I posted the following query:

Dagbani (Gur, Niger-Congo) is a two-tone language of Northern Ghana.
Typical nouns have a monosyllabic or disyllabic stem plus a
(monosyllabic) nominal class suffix. It seems that each morpheme has
not more than one tone, at least on the surface, which spreads if
there is more than one TBU. Are there other languages which
illustrate the "1 tone per morpheme" principle? If you know of any,
please let me know.

- Thanks to Stephen Straight, Alexandre Kimenyi, Deborah Schmidt and
Malcolm Ross for their replies. Although the languages mentioned do
probably not exactly refer to the same phenomenon as in Dagbani, they
display interesting parallels regarding tonal structure.

In the meanwhile I found that the discovery of "one tone per morpheme"
in Dagbani is mainly a surface tendency, as nominal roots may be
toneless or contain a floating L tone as well. The surface tonal
structure of nouns, however is H or L, even with most polysyllabic
roots. A more detailed analysis is going to appear in the proceedings
of the 2nd World Congress of African Linguistics.

The following messages were received as replies to my query:
- --- Forwarded message begins here -----
From: H Stephen Straight <>
Date: Fri, 7 Feb 1997 15:22:16 -0500 (EST)
Subject: Yucatec Maya

Yucatec Maya roots are primarily monosyllabic CVC, with neutral,
rising-falling, low, or laryngealized V. All four of these tonal
types do appear in the first syllable of the few disyllabic roots I
can think of, but in every case the second syllable is neutral-toned.
I had never thought of the tone-only-on-initial-syllable
generalization that follows from these facts. Thank you for making me
recognize it, whatever you may find regarding the possible
universality of this phenomenon.

Date: Fri, 7 Feb 1997 18:38:45 -0700 (PDT)
From: Alexandre Kimenyi <>
Subject: Morpheme-Tone

Kinyarwanda has also has high and low tones. Like the language you
mention there is only one underlying high tone per morpheme. Other
phonetic high tones are obtained through tone rhythm rules (tone
spread, Meeussen rule, etc.). All this is described in my forthcoming
book, A Tonal Grammar of Kinyarwanda: an Autosegmental and Metrical
Analysis, Mouton de Gruyter (Trends in Linguistics).

From: Deborah Schmidt <>

I expect you will get many responses. My impression is that tonal
phenomena such as you describe are not terrifically unusual. I have
written two articles on Basaa, a Cameroonian Bantu language, which has
monosyllabic and disyllabic verb roots. All verb roots, whether
monosyllabic or disyllabic, are either toneless (and surface as low
toned throughout) or lexically high toned (and surface with the high
tone on the first syllable and a falling tone on the second, if there
is a second). The articles I have written do not deal with tone,
however, and there seem to be some interesting interactions that I
have not looked much into involving tone and epenthesis/deletion of
vowels. Have a look, especially at "Phantom Consonants in Basaa",
which is iin _Phonology_ 11.1. The other Basaa article is in the most
recent issue of_Phonology_.

From: Malcolm Ross <>

Yabem, an Austronesian language of Papua New Guinea, is a language in
which each "simple word" has a single tone. However, a simple word may
contain two morphemes (root and bound), so I am not sure how relevant
this is to your interests. See my brief account of Yabem phonology
(p708) in Tryon, D.T. (ed.), 1994. Comparative Austronesian
Dictionary. Berlin: Mouton de Gruyter.
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue