Wednesday, 6 October 2021

Matthews' Fifth Specific Problem For Halliday's Proposals On 'Rank'

 Fawcett (2010: 310):

Fifthly, Matthews points out that the unit that is here called the genitive cluster (e.g., my father's) is a serious embarrassment for the concept of the 'rank scale'. The reason (though this is not how Matthews presents it) is as follows. While any such expression is clearly a unit that consists of more than one word, the 'rank scale' principle denies it 'group' status because it cannot function as a direct element of the clause. In fact it always fills an element of a group (usually the deictic determiner in a nominal group), and this suggests that it must be 'lower' on the 'rank scale' than the group. However, while one of its main elements is expounded by a morpheme (almost always 's), its other principal element is always filled by one or more nominal groups (and occasionally by a clause, as in whoever ate it's mistake). The existence of the genitive cluster is therefore a serious problem for the concept of the 'rank scale'. (See Section 10.2.11 of Chapter 10 and Appendix B for the straightforward approach to this unit that is taken here.)


Blogger Comments:

[1] This is misleading, because what Fawcett terms a 'genitive cluster' is neither an embarrassment nor a problem for the concept of a rank scale. In SFL Theory, this is simply an embedded nominal group: one that has been shifted to word rank to serve as Deictic of a nominal group. For example:

[2] This is misleading, because it is not true. In terms of the rank scale, what Fawcett terms a 'genitive cluster' is indeed a nominal group, but one that has undergone rankshift to serve a function in nominal group structure.

[3] This is misleading, because it is not true. Such expressions also occur as ranking nominal groups that serve functions in clause structure. For example:

[4] This misleading, because it is not true. Even if it were true that such expressions are always rankshifted — in terms of the rank scale — the fact they are rankshifted demonstrates that they are of the same rank as, or higher than, the unit in which they function.

[5] To be clear, a nominal group like my father's consists of two words, my and father's, the first of which is realised by the morpheme my and second by two morphemes father and 's.

[6] To be clear, in this bizarre construction — which is found nowhere else on the internet — the clause whoever ate it is shifted to morpheme rank, as shown by its fellow constituent, the morpheme 's, and so serves as the Head element of a noun:


No comments:

Post a Comment