The Classification Of Continuatives

Martin & Rose (2007: 143):

So we can classify continuatives both by the type of logical relations, and the type of expectancy they realise, as in Table 4.7.

 

Blogger Comments:

[1] To be clear, firstly, in the foregoing discussion, Martin & Rose have classified a class of word, the continuative, a subtype of conjunction, and presented the classification as discourse semantic rather than grammatical.  This is inconsistent in terms of stratification.

Secondly, in doing so, Martin & Rose have taken a bottom-up perspective: beginning with forms and then classifying them in terms of function. This is the opposite perspective of SFL Theory, which takes a top-down perspective: first determining the distinctions in meaning and then identifying how such distinctions are realised.

Thirdly, in terms of SFL Theory, none of words claimed to be continuatives are actually continuatives. As previously demonstrated, they are a mixture of adverbs and conjunctions functioning as either mood Adjuncts (interpersonal metafunction) or conjunctive Adjuncts (textual metafunction). Interpreting the function of these items as logical is also theoretically inconsistent in terms of metafunction.

[2] To be clear, firstly, as previously demonstrated, in terms of SFL Theory, the items too, also and as well do mark a relation of addition, but they serve as conjunctive Adjuncts and mark a textually cohesive relation at the level of grammar, rather than continuatives marking a logically structural relation at the level of discourse semantics. The theoretical confusions here are both metafunctional and stratal.

Secondly, as previously demonstrated, in terms of SFL Theory, none of the items marks a relation of comparison. The item so (did he) serves as a conjunctive Adjunct and marks a textually cohesive relation of addition, whereas the items only, just and even function interpersonally as mood Adjuncts of intensity. The theoretical confusions here are in terms of expansion relation, metafunction and stratification.

Thirdly, as previously demonstrated, in terms of SFL Theory, the items already, finally, at last, still and again do mark temporal features, but not those ascribed to them by Martin & Rose. The items already and still function interpersonally as mood Adjuncts of temporality, whereas the items finally and at last function textually as conjunctive Adjuncts marking a cohesive relation of time. The instance provided of the item again, on the other hand, functioned experientially as a circumstance of temporal Extent. The theoretical confusions here, once more, are both metafunctional and stratal.

[3] To be clear, as previously demonstrated, in terms of SFL Theory, the only types of expectancy here are interpersonal, and apply to the mood Adjuncts of intensity and temporality (Halliday & Matthiessen 2014: 187-9):

• only and just: counter-expectancy: limiting;
• even: counter-expectancy: exceeding;

• already: 'by' the time at issue;
• still: 'since' the time at issue.

The theoretical confusions here, once again, are both metafunctional and stratal.