PLM 2005 ABSTRACTS VAULT
see http://elex.amu.edu.pl/ifa/plm
for further details
The paper aims to emphasise the view that the typical questions concerning the nature of morphonological as well as phonomorphological phenomena can be answered in a satisfactory way only if a restrictive model of phonology is first established, that is, one which makes it clear which processes are and which are not phonological. In this sense the most crucial question: ‘What is morphonology?’ can be answered: ‘it is what looks like phonology but is ruled out as such by phonological theory’. A strict model will also determine how morphology may intervene in the works of regular phonology.
The immediate question that should be asked at this point concerns the source of morphonological phenomena, e.g. alternations. Since they look like phonological, it should be perhaps assumed that they have a phonological basis. In the case of some morphonological phenomena, phonological basis should perhaps be understood as having phonological origin. It would, for example, mean that regular phonological alternations may cease to be phonological in the course of the development of a given linguistic system. However, instead of disappearing, they become preserved / petrified in some way and continue to constitute a significant pattern. Sometimes, like in the case of vowel – zero alternations in Polish we may be dealing with both regular phonology and morphonology co-operating.
Thus, next to purely phonological effects like final devoicing in Polish, or root-internal phonotactics, there are phenomena which are both phonological and morphophonological at the same time. Logically, there are also phenomena which are purely morphophonological, for example, the alternations [k - ts] ręka – ręce ‘hand, Nsg./loc.’.
At a descriptive level, the source of morphonological phenomena clearly lies in the logics of a given phonological model. For example, in a non-derivational model such as Government Phonology, which is sometimes referred to as the Principles and Parameters approach, the phonological representation is built according to universal principles of government and licensing and a battery of language specific parameters. It may be envisaged, that a possible source of morphonology may lie in the switch of a parameter that in its previous setting caused alternations which are no longer allowed after the setting has changed, and yet, the alternation is preserved by means of some kind of marking – a pattern without a cause.
Phonological representation in GP is ‘redundancy free and fully interpretable’ as a phonetic form without the need to derive any phonetic details. The question is if the interpretative aspect of this theory, which GP does not even intend to formalise, could also be the area where morphonological phenomena originate. An intuitive answer of a GP phonologist should probably be negative. However, it seems that the facts are different. Does it mean that interpretation of phonological structure should also be formalised?
Finally, there are phenomena with phonological basis which should perhaps be called phonomorphology, that is, phonologically conditioned morphology. This concerns situations where a morphological process respects the phonological structure of the base in selecting the shape of the affix, for example, –szy or –ejszy in the comparative allomorphy in Polish.