Rival morphological processes 1: The productivity of verb-deriving processes Conclusion

It was shown that both the OED-based and the corpus-based productivity measures have turned out to be useful analytical tools, contrary to earlier claims in the literature. In general, the nature and results of the OED measure are similar to those of global productivity P* but also incorporate an element of probability by counting the number of neologisms for a given period. In the text-based measure P, this aspect is ex pressed by dividing the number of new types by the number of all tokens of the relevant category. In the light of these considerations, the OED measure combines aspects of Baayen's P* and Ρ without being exactly equivalent to either one of them.

Although the quantitative results are useful and important, the methodological problems involved with both types of data-bases are enormous, and no single measure should be overinterpreted in its estimation of the productivity of a process. In fact, the above account can be seen as a case for the combination of both types of measures, since a careful comparison of different data bases can help to detect and balance the bias necessarily involved with each individual measure. It seems, however, that the text-based measures are much better suited for affixation processes which are both semantically more clearly defined and phonologically more transparent. Especially with the verbal suffix -ate, the methodological problems abound because of its high degree of semantic, phonological and morphological opacity.

The quantitative assessment of the productivity of the verb-deriving processes can be summarized as follows. The most productive one is conversion, followed by -ize, the most productive overt suffix by far, and the suffixes -ify and -ate. All other processes examined are practically dead. In comparison to -ize, the suffixes -ify and -ate seem to be more severely restricted, in that -ate has a high global productivity, but the probability of encountering new forms is rather low. Conversely, -ify has a higher chance of occurring in new formations, but the sheer number of these instances is rather low.

Where do these differences among the suffixes come from? As already pointed out, the productivity measures are only the first step in our analysis. They state the problem, but do not solve it. We will see that the observable differences in productivity between the verb-deriving processes are largely a reflection of the structural restrictions imposed on these processes.