B. Connell
One of the more common means of typologizing tone languages is
according to their characteristics with respect to pitch realization. So, early
on, Pike (1948) distinguished between register and contour tone languages; it
is register type systems that are typically found in Africa. Later, Welmers
(e.g. 1973), differentiating among register tone languages, identified discrete
and terracing types of systems. Some languages are known to have one or more of
a number of possible types of downtrends: declination, downdrift, downstep
(automatic or non automatic), and final lowering. The first of these four terms
is relatively new in the study of African tone systems, originating in work on
Dutch intonation; downdrift and downstep are now common in studies of tone and
intonation, and originated in work on languages on West Africa; final lowing,
too is now commonly identified in work on non-tone languages, but Welmers
(1973) again identifies this as common feature of pitch realization in African
tone languages.
The Problem: There is a problem, though, in attempting to understand or typologize African tone systems in terms of these characteristics. The terminology arose largely through impressionistic work, and consequently use of these terms as found in the literature, especially downstep and downdrift is rather inconsistent: one author’s downstep may be another’s downdrift, and vice versa. Stewart (1983) went some way to addressing the problem in distinguishing between automatic vs non-automatic downstep, but still, automatic downstep is frequently taken as synonymous with downdrift. The recent introduction of the concept of declination into studies of African tone systems brought in an additional element of terminological complexity; declination is may be taken to be the same as downdrift, though it is not, it seems, confused with automatic downstep. So the questions arise: are each of these terms to be identified with distinct phenomena, or are one or more of them in fact redundant? And, more important, is this merely a problem at the level of terminology; or does it indicate a lack of clarity at a theoretical level. If the answer to the first question is ‘yes’, we must be aware that incorrectly labelling a particular phenomenon may lead to the formulation of inappropriate research questions, and thence to dubious theoretical conclusions.
The solution: The aim of this paper is to examine the set of terminology used to discuss downtrends in African languages, with a view to answering the questions posed above. Instrumental evidence drawn from both published and unpublished work on a number of different languages is adduced. It is shown that each of non-automatic downstep, automatic downstep, downdrift, and declination are separate phenomena. The confusion, or overlap, in their usage stems partly from the simple fact that until recently an insufficiently large range of languages to demonstrate the different phenomena had been examined; but also perhaps to some extent it can be attributed to an uncritical acceptance of the status quo. Separating the terminology is also related to more theoretical questions: certain of these terms refer to what may be considered ‘local’ phenomena, while others are seen as ‘global’ phenomena. Seeing them in this light, again, allows the differences among them to be clarified.
Hombert, J-M. (1974) Universals of Downdrift: their phonetic
basis and significance for a theory of tone.
Studies in African Linguistics, Suppl. 5 169–183.
Laniran, Y. (1992) Intonation in a tone language: the
phonetic implementation of tone in Yoruba. Ph.D., Cornell,
Pike, K.L. (1948) Tone Languages . Ann Arbor: University of Michigan Press.
Stewart, J.M. (1983) Downstep and Floating Low Tones in
Adioukrou. Journal of African
Languages and Linguistics, 5 57-78.
Welmers W. (1973) African Language Structures . Berkeley: University of California Press.
Also papers by Ahoua & Leben, Clements, Clements & Laniran, Leben, Stewart, Urua, etc.