Blocking, as traditionally defined, refers to a situation where the mere presence of a form or pattern in language preempts the application of another form or pattern. Recent studies on blocking, especially in derivational morphology, have provided many examples which seriously cast doubt on the assumed discreteness of the notion. Using data from sizeable corpora, this paper seeks to address this issue in inflectional morphology. Within that domain, it is illustrated that – throughout time – there may have been an alternation between competing morphological patterns. In particular, pluralization by suffixation and transfixation, may have alternated in assuming the role of a blocker. On the basis of frequency data of inflectional classes in New Persian, it is shown that whenever there is an increase in the usage of either of these two strategies, there is also a decrease in application of the other. This alternation follows a neat and systematic pattern strongly suggesting that transfixation and suffixation have relative blocking effects on each other. It can therefore be concluded that there is essentially no need for one form, or a number of forms, to be eliminated in favor of a more productive one. In fact, competing forms can co-exist, decline and flourish alternatively, and thrive for centuries.
