Misreading Modality

Doran, Martin & Zappavigna (2025: 47-8):

Looking at the I think now, as (Halliday & Matthiessen, 2014, pp. 686-687) notes, the I think here is unlikely to be providing a full proposition. We can see this by the fact that if we were to tag the clause, we would more likely tag in relation to the you’re upset, rather than the I think; We would more likely say: I think you’re upset because the TV wasn’t working, aren’t you; as opposed to I think you’re upset because the TV wasn’t working, aren’t I. Similarly, if Kristy was to support what her mother was saying, she would likely support the ‘you’re upset…’, rather than the ‘I think’ with Yeah, I am, rather than Yes, you do. This is because the I think is not presenting a proposition about thinking, but rather being used metaphorically to mean something like ‘probably’. In Halliday’s terms, it is an interpersonal metaphor realising modality. In this sense, the I think is in fact realising support for the idea that Kristy is upset because the TV isn’t working. As we will discuss in Chapters 4 and 5, the I think also makes explicit that what is being tendered is a personal opinion and it lowers the stakes of the position.

The I don’t think in the first line functions similarly to the second line, but realises a rejection. We can see this again by the fact that it is unlikely that we will reject the ‘thinking’ (Yes you do), but also the fact that the negation can be easily transferred to the you’re upset without significantly changing the meaning: I think you’re not really upset about me going (Halliday & Matthiessen, 2014, p. 689). So Kristy’s mother has tendered two propositions in opposition to each other in order to reject one and support the other – i.e., that Kristy is not really upset about her going but that she is upset because the TV wasn’t working.

ChatGPT Comments:

This discussion attempts to leverage Halliday’s account of interpersonal metaphor to justify the claim that I think in I think you’re upset because the TV wasn’t working constitutes a form of support in the authors’ tenor system of positioning. The claim appears to be that I think realises support for the proposition that follows, rather than functioning as a proposition in its own right.

[1] There is a serious category error here: I think is rightly analysed (in Halliday & Matthiessen 2014) as an interpersonal grammatical metaphor for modality, not as a move in a system of tenor-positioning such as “support” or “reject.” The authors ignore the distinction between interpersonal grammar and contextual tenor, collapsing them into one another in order to preserve the illusion of theoretical innovation.

[2] The suggestion that I think somehow “realises support” misconstrues both the nature of modality and the analytic function of interpersonal metaphor. I think modifies the speaker’s commitment to the proposition; it does not itself “support” that proposition in a second-order sense.

[3] Worse still, the authors treat grammatical diagnostics such as tag questions and preferred responses (e.g., aren’t you vs. aren’t I) as if they confirm the categories of their own invented tenor system. But these diagnostics are evidence for clause structure and mood, not for tenor roles or speaker positioning. This is a classic case of retrofitting grammatical facts to support a speculative model.

[4] Finally, the apparent opposition between I think you’re not really upset about me going and I think you’re upset because the TV wasn’t working is a red herring. The former encodes a single modalised proposition; it is not a rejection of a second speaker's stance. The claim that Kristy’s mother “tenders two propositions in order to support one and reject the other” is interpretive projection masquerading as analysis.