Seriously Misrepresenting Behavioural And Verbal Processes

Martin, Matthiessen & Painter (2010: 124, 124n):

Note that there are a number of processes representing verbal behaviour – talk, chatter, gossip, speak, lie – that are behavioural rather than verbal¹⁸; they cannot project. (See Table 4.9 above for examples.) For example, we cannot say they gossiped that their neighbours had had a wild party. (Note that while behavioural process can, in a written narrative, project by quoting: ‘I enjoyed it thoroughly’, he lied, a true verbal process can project both direct and indirect speech in all contexts.)

¹⁸ IFG3: 251 Table 5(24) lists these as verbs functioning in behavioural clauses; however they also appear on p. 255 as examples of verbs functioning in verbal clauses of ‘activity’. The inability of such clauses to project is an argument for preferring the behavioural interpretation.

Blogger Comments:

[1] This is misleading, because it misrepresents behavioural processes. In such cases, the lexical choice adds a behavioural feature to a verbal Process. Halliday (1994: 139):

…while ‘behavioural’ clauses do not ‘project’ indirect speech or thought, they often appear in fictional narrative introducing direct speech, as a means of attaching a behavioural feature to the verbal process of ‘saying’…

However, in the case of lied, what is added to the verbal Process is the quality 'untruthful':

[2] To be clear, this is a bare assertion unsupported by argument. Just as not all mental processes project, not all verbal processes easily project reported speech, Halliday (1994: 141) observes:

[3] This is misleading, since the only verb common to both lists is talk. The 'behavioural' list is chatter, grumble, talk, gossip, argue, murmur, mouth whereas the 'verbal activity' list is praise, insult, abuse, slander, flatter, blame, criticise, chide, speak, talk.

But, more importantly, the fundamental misunderstanding here is that verbs can be assigned to process types, without regard to the clauses in which they function.

[4] To be clear, even if a difficulty in reporting were equivalent to an inability to project, this is demonstrably still a very inadequate reason for interpreting targeting verbal clauses as behavioural clauses, because it creates unnecessary theoretical inconsistencies.  For example, a behavioural clause is middle voice, with the Behaver as Medium, whereas a targeted verbal clause is effective voice with Target as Medium, so that in a behavioural interpretation, the Behaver of such clauses must function as Agent, not Medium.

Moreover, a behavioural clause construes its first participant, Behaver, as conscious, whereas a targeted verbal clause, like other verbal clauses, only construes its first participant, Sayer, as a symbol source: the newspapers blamed the Trump supporters. And following on this, a behavioural clause construes its Medium participant, Behaver, as conscious, whereas, in a targeted verbal clause, the Medium participant, Target, need not even be animate: the report of the bureau of meteorology blamed La Niña.

To be clear, Halliday & Matthiessen (2014: 354) provide eight criteria that distinguish verbal processes. Singling out just one of these, the ability to project, and giving it priority, results in unnecessary theoretical inconsistencies which add nothing to the explanatory power of the theory.