Summary: Chapter 5 – Tuning

Chapter 5 presents Tuning as a new system for adjusting interpersonal meaning, structured around three clinal dimensions: Scope (breadth of address), Stakes (degree of risk), and Spirit (affective tone). The chapter positions Tuning as a discrete system that complements Positioning and Orienting, offering additional ways to calibrate how meanings are “tendered” in discourse. However, a close reading reveals several critical issues in its conceptualisation and analytic necessity.

1. Redundancy with existing SFL resources

Each of Tuning’s dimensions overlaps heavily with established systemic-functional categories:

• Scope parallels Engagement and Graduation, as well as reference and membership categorisation devices, capturing inclusivity/exclusivity and group alignment.

• Stakes is largely a repackaging of Graduation (force/focus) and Engagement (proclaim vs entertain), with its “risk” defined relationally by background values rather than independently observable linguistic features.

• Spirit mirrors Attitude (Affect, Judgement, Appreciation) and interacts with Graduation and Engagement to produce “tone,” yet is presented as a separate dimension labelled with a metaphorical “vibe.”

Across all three dimensions, Tuning risks terminological inflation — taking familiar linguistic and discourse-semantic resources, giving them new labels, and promoting them as a distinct system.

2. Slippage between linguistic description and social interpretation

Tuning repeatedly blurs the boundary between:

• Observable linguistic patterns (pronouns, vocatives, quantifiers, hashtags, emojis, commands), and

• Inferred social consequences (disruption of community values, high-stakes positions, warmth or hostility in social bonds).

For instance, the notion of “stakes” relies on an analyst’s projection of social significance within a community, rather than on measurable linguistic features. Similarly, “spirit” uses impressionistic affective labels (warm/warn) that risk turning description into subjective commentary. These slippages undermine the claim that Tuning is a discrete linguistic system.

3. Circularity and dataset constraints

The chapter sometimes embeds Tuning in the corpus design itself (e.g., #momguilt posts), producing circular validation: scope appears everywhere because the dataset was selected to foreground a particular community, which then reinforces the impression that scope is a systematic resource.

4. Meta-system framing

Ultimately, Tuning functions more as a meta-functional overlay, a way of describing how existing resources (Appraisal, Engagement, Graduation, reference devices) can be “adjusted” across a continuum, rather than as a system with independent analytic status. Its conceptual appeal lies in metaphorical clarity (scope, stakes, spirit), but the theoretical necessity is weak.

Conclusion

While Tuning offers an engaging and accessible metaphor for how interpersonal meaning can be adjusted, it largely repackages established SFL categories, blurs analytic strata, and imports social interpretation into linguistic description. Its main contribution may be pedagogical — helping readers think about alignment, risk, and tone — rather than extending the theoretical apparatus of SFL. Readers should approach Tuning with an appreciation of its rhetorical and descriptive value, but with caution regarding its analytic independence.

High Stakes and Shifting Grounds: When Tenor Becomes Institutional Loyalty

Doran, Martin & Zappavigna (2025: 139):

Staking describes the resources that adjust the relative risk of a proposition or proposal, making it more or less consequential. The risk is relative in the sense that it works in concert with the constellations of values that are always in play within the discourse of any particular community and which may shift abruptly or evolve over time. The tendering of a proposition or proposal will be more or less disruptive to these constellations. For example, if within the SFL community, one was to reject Halliday’s work, it would be tantamount to rejecting the SFL community itself. On the other hand if one were to reject the work of one of the authors of this book, it would not necessarily have the same effect (though it would have a larger or smaller effect depending on which author you are rejecting). In the constellation surrounding SFL, Halliday is very high stakes.

ChatGPT Comments:

This passage attempts to introduce staking as a system for describing the relative “risk” associated with meaning in discourse. However, it immediately anchors that risk not in the semantics of discourse but in the social politics of academic communities — in this case, the institutional hierarchy of Systemic Functional Linguistics. The example of “rejecting Halliday’s work” equated with “rejecting the SFL community” does not merely illustrate the concept; it inadvertently collapses the distinction between tenor as semiotic context and tenor as sociological affiliation.

Within canonical SFL, tenor is a variable of context realised by interpersonal semantics. It describes the relations enacted through meaning, not the institutional relations maintained through allegiance. By using “risk” in this way, the authors conflate the semantic system with the social system, translating the modulation of interpersonal meaning (e.g. commitment, obligation, affective alignment) into the sociological maintenance of community identity.

The analogy therefore re-grounds staking not in the grammar of interpersonal exchange, but in disciplinary loyalty. Rather than clarifying how meanings scale in value, commitment, or investment within discourse, the passage reduces staking to an index of orthodoxy — a measure of how far a speaker can deviate from institutional norms without being excommunicated. The result is that risk is treated as reputational rather than semiotic.

A theoretically coherent treatment of staking would instead keep the focus within the discourse semantic plane: on how evaluation is intensified or mitigated, how commitment is modulated through ENGAGEMENT, or how alignment is negotiated in APPRAISAL. These are the mechanisms by which stakes are raised or lowered within the meaning system, rather than within the discipline. 

Raising and Lowering Risk: STAKES as Semiotic vs Social Evaluation

Doran, Martin & Zappavigna (2025: 135):

Another system in tuning, STAKES, is about raising or lowering the degree of risk imbuing a meaning. For example, within a particular community we might consider how much of an issue it is if someone rejects a particular tendered meaning. Returning to (15), it would be a controversial and high-stakes position to reject the post's tender that moms don't need to carry guilt or that they don't need to follow the commands in the image ('Stop giving life to negative feelings. Bring it to God. Release it. Move forward. Be free.'). In addition to implicit staking that relies on the background of already positioned values about motherhood guilt that this community likely holds, the post raises the STAKES of the tender through various resources. For instance, the semiotic risk is intensified via ENGAGEMENT: DISCLAIM (e.g. 'don't need to'), the commands in the image and the hashtag (e.g. '#letitgo'), and the upscaled GRADUATION (e.g. 'heavy enough').

ChatGPT Comments:

The authors position STAKES as a system that modulates the degree of interpersonal risk within a discourse. While this framing is conceptually clear, it risks conflating the linguistic construal of risk with sociocultural evaluation. SFL would describe the same phenomena in terms of co-instantiation of interpersonal resources — mood, polarity, appraisal, and engagement — without presupposing that the text alone carries a fixed level of risk. By treating the rejection of a proposition as “high-stakes” based on assumed communal values, the account imports social interpretation into what is fundamentally a linguistic description. While the examples given (imperatives, hashtags, disclaimers, graduation) do indeed adjust the intensity of a proposition, the analytical leap from linguistic choice to community-level controversy is not fully grounded in observable semiotic patterns, leaving the risk evaluation more inferential than descriptive.

Diagnostic Critique of Chapter 3

ChatGPT Comments:

1. Conflation of system and instance.

Throughout the chapter, the authors oscillate between describing systems of meaning (tendering, rendering, positioning) and describing interactional practices (how people negotiate, layer, reposition). These are repeatedly collapsed into one another, leaving unclear whether the “system of POSITIONING” is proposed as an abstract, paradigmatic system or a descriptive account of conversational behaviour. This weakens theoretical rigour, as systems and instances require different orders of abstraction.

2. Inconsistent modelling of system relations.

The representation of tendering and rendering as both disjunctive and conjunctive (Figure 3.3) exemplifies a lack of modelling precision. The same categories are at times mutually exclusive, at times co-instantiable, without theoretical justification for the dual relation. The ambiguity undermines the explanatory power of the system network and makes it unstable as a representation of paradigmatic options.

3. Ambiguity in layering and planes of negotiation.

The notion of layering — simultaneous tendering and rendering, internal vs. external negotiation — is presented as central to the model. However, the account remains descriptive rather than systemic. The resources for layering are not modelled paradigmatically, but instead gestured to narratively. This produces a vocabulary of interesting interactional phenomena, but not a systematised account of how these options interrelate.

4. Unstable treatment of text as object vs. semiosis.

In discussing “internal negotiation,” the authors position text both as “a thing in itself” and as “semiosis about something else.” This binary creates unnecessary theoretical turbulence, since in a stratified model text is always simultaneously semiotic event and semiotic object, depending on the perspective of construal. The attempt to posit a second “plane” of negotiation recycles existing metafunctional distinctions without properly situating them.

5. Rhetorical recourse to everyday intuitions.

The chapter repeatedly falls back on conversational common sense (“whenever we talk, we negotiate our social relations”) as if this everyday grounding lends theoretical authority. While pedagogically useful, such recourse risks diluting the analytic model into paraphrase of the obvious, rather than consolidating it as a precise account of meaning-making.

---

Overall Diagnostic

Chapter 3 sets out to formalise a system of POSITIONING but ultimately produces a hybrid discourse that oscillates between systemic categories, descriptive narratives of talk, and evaluative accounts of values and ideology. The central problems are (i) repeated conflations of distinct semiotic orders (system/instance, meaning/value, tenor/context), and (ii) instability in systemic modelling (disjunctive vs. conjunctive categories, descriptive vs. paradigmatic layering). The cumulative effect is to undermine the chapter’s claim to present a coherent “system” of positioning: what emerges instead is a shifting patchwork of overlapping intuitions, diagrams, and metaphors, which lack integration within the Hallidayan architecture they purport to extend.

Layering, Repositioning, and Rendering: When Jargon Outpaces Theory

Doran, Martin & Zappavigna (2025: 82-3):

This negotiation is enabled not simply through the different ways in which we can tender and render positions, but also the possibility of layering these meanings. In the previous chapter we explored that people may tender and render meanings at the same time, reacting to something that has happened (or may happen) while putting something else forward to be negotiated. We saw that this was a key resource for extending conversation indefinitely, and in doing so, we offered a view on chat from the perspective of tenor. In addition, we saw that when engaging with meanings through rendering or nuancing meanings through purview, we can do so for the propositions or proposals being discussed (external negotiation), or we can alternatively do so for the speech acts themselves (internal negotiation) – in the latter case commenting on how language itself is being used. This allows for a second plane of discussion that takes the reality of text as a thing in itself, rather than just semiosis about something else. Finally, we saw all of these resources could be repositioned, with one position standing for another. This offers a third avenue through which people may negotiate their social relations along multiple lines.

ChatGPT Comments:

1. Terminological inflation: “layering” without systemic specification

The authors introduce layering as an enabling resource for negotiation but do not specify its systemic status. Is layering a distinct system, a metafunctional overlap, or simply the simultaneity of meanings across strata? Without clarification, the term risks functioning as descriptive rhetoric rather than theoretical category, duplicating what Halliday already models through rankshift, embedding, and metafunctional interplay.

2. The simultaneity of tendering and rendering

The claim that meanings can be tendered and rendered “at the same time” simply restates the well-established principle that interpersonal moves are often co-instantiated (e.g. offering while qualifying, demanding while conceding). Systemic functional linguistics has long accounted for this simultaneity through mood and modality working in tandem with appraisal. By presenting simultaneity as a new discovery, the model obscures rather than extends the canonical framework.

3. “Indefinite extension of conversation” as explanatory overreach

The assertion that layering is a key resource for extending conversation indefinitely is overstated. Conversation can continue due to a wide range of discourse-semantic resources — topic management, cohesive chaining, politeness, repair — not merely because tendering and rendering can be layered. The explanation risks narrowing a complex phenomenon to the rebranded categories the authors have introduced.

4. The tenor framing of “chat”

Positioning this as a “view on chat from the perspective of tenor” misapplies SFL’s stratification: tenor is a contextual variable realised by semantics, not a discourse-semantic system in itself. The analysis thereby collapses levels, treating tenor as if it were directly constituted by conversational strategies, rather than mediated by semantic systems.

5. External vs internal negotiation: duplication of appraisal resources

The external/internal distinction (propositions/proposals vs. speech acts themselves) echoes the canonical contrast between engaging with experiential content and engaging with dialogic positioning (e.g. heteroglossic resources for acknowledging or contesting speech acts). By rebranding this as internal vs external negotiation, the authors obscure existing appraisal theory while introducing redundant terminology.

6. Text as object vs text as semiosis

The claim that internal negotiation opens a “second plane of discussion” in which the reality of text as a thing in itself is taken up conflates reflexivity with stratification. Reflexive commentary on language is already theorised in SFL through projection, modality, and appraisal (particularly metadiscourse and attribution). Invoking a new “plane” suggests ontological novelty where there is in fact an already-mapped systemic pathway.

7. Repositioning and substitution: vague but familiar

Finally, the claim that resources can be repositioned, with one position standing for another is not unpacked theoretically. If this refers to metaphor (e.g. mood metaphor, interpersonal metaphor), it is already central in SFL. If it refers to substitution of one stance for another, then appraisal covers this territory. Without clear mapping, “repositioning” risks being a catch-all label rather than a defined system.

8. General tendency toward rhetorical accumulation

Across the passage, a pattern emerges: new terminology (layering, internal/external negotiation, repositioning) is piled onto the tendering/rendering model, but without showing clear systemic relations or offering criteria for delimiting one resource from another. The effect is rhetorical accumulation rather than theoretical integration. 

Tendering Nuance, Flattening Complexity: Innovation or Illusion in Interpersonal Meaning?

Doran, Martin & Zappavigna (2025: 82):

Whenever we talk to each other, we negotiate our social relations. We put forward meanings and we react to those that have been put forward. We present meanings in a way that suggests they may or may not be shared and we engage with them in more or less sympathetic ways. In Chapters 2 and 3 we have presented a model of resources for how we put forward and react to these meanings. We first stepped through the system of rendering to show the different ways we can engage with meanings that have been tendered or implicated in a text. Then in this chapter, we explored how people can put forward meanings in nuanced ways that acknowledge the positions of those around them. Together these resources allow for the intricate negotiation of social relations that all of us engage in everyday.

ChatGPT Comments:

1. Conflation of negotiation with all talk

The opening claim — “Whenever we talk to each other, we negotiate our social relations” — generalises negotiation to all instances of dialogue. While interpersonal meanings are indeed always at stake in interaction, negotiation (in the Hallidayan sense) is a specific semantic process of exchanging roles and values. By universalising negotiation, the authors risk blurring the crucial distinction between the omnipresence of interpersonal meaning and the specific resources by which exchange and alignment are enacted.

2. The “put forward / react” binary

The model is framed around a reductive dyad: putting forward vs reacting. This foregrounds a mechanical push-and-pull exchange structure, reminiscent of speech function theory, but ignores the range of simultaneous interpersonal meanings realised in dialogue: stance, alignment, modality, affect, judgement, graduation, etc. By collapsing all interpersonal work into this binary, the model oversimplifies the semiotic resources available for interpersonal meaning-making.

3. Rendering as engagement with tendered meanings

The authors describe “rendering” as the set of resources for engaging with meanings that have been tendered or implicated in a text. This creates two theoretical difficulties:

• Circularity: “rendering” is defined only in relation to “tendering,” which itself is not clearly theorised beyond “putting forward.”

• Conflation of strata: “implicated meanings” reach into discourse semantics and intertextual inference, yet are folded into a system as context. This risks treating distinct phenomena as one, without specifying the systemic relations that differentiate them.

4. Affective alignment disguised as nuance

The claim that people put forward meanings in nuanced ways that acknowledge the positions of those around them conflates evaluative stance with exchange roles. In SFL, such “acknowledgement” is modelled through appraisal resources (e.g. heteroglossic engagement, attribution, concession) rather than through speech functions. By folding evaluative alignment into a tendering system, the model displaces appraisal from its place in the semantic architecture and risks duplicating it under new terminology.

5. Overstating integration

The authors conclude that together these resources allow for the intricate negotiation of social relations that all of us engage in everyday. This overstates the explanatory scope of the proposed “rendering/tendering” framework. The canonical SFL model already accounts for this through mood, modality, appraisal, and speech function. The new terminology does not so much integrate resources as re-package them, producing an illusion of explanatory novelty without clarifying their interrelations. 

The POSITIONING System Network

Doran, Martin & Zappavigna (2025: 81-2):

Figure 3.3 brings together the options for rendering built in the previous chapter with those of tendering in this chapter to show the full system of POSITIONING.

ChatGPT Comments:

 Here are some theoretical tensions and confusions in this model that stand out:

1. Heterogeneous criteria of system delicacy
The network presents “POSITIONING” as a unified system, yet the delicacy expansions combine very different dimensions of meaning. The initial branching (render vs tender) sets up the primary contrast, but subsequent sub-systems shift criteria inconsistently:

• address vs note concerns interpersonal orientation versus commentary,

• placement (external vs internal) invokes discourse-structural positioning,

• proposition vs proposal introduces metafunctional clause type,

• open vs complete invokes structural polarity,

• purview introduces epistemic stance.
  What is presented as systemic delicacy is in fact a heterogeneous aggregation of distinctions that originate in different metafunctions and strata.

2. The tendering/rendering ambiguity
The diagram signals both a disjunctive and a conjunctive relationship between render and tender. This produces a conceptual inconsistency: are rendering and tendering mutually exclusive choices, or are they simultaneous dimensions of meaning? The attempt to code both at once collapses two theoretically distinct relations:

• paradigmatic opposition (choosing one or the other), and

• simultaneous activation (both may apply to a single instance).
  In practice, rendering and tendering often co-occur (e.g. “I think you’re wrong” both evaluates an existing stance and advances a counter-claim). The network’s simultaneous disjunctive/conjunctive representation fails to clarify how these dimensions intersect, instead producing analytical indeterminacy.

3. The purview problem
The addition of speaker purview versus listener purview exemplifies the model’s attempt to taxonomise epistemic stance as if it were a structural variable of the same order as mood or polarity. The resulting options (share, assert, pose, air) artificially discretise what is in practice a gradient of interpersonal alignment. More significantly, this construct duplicates resources already modelled in SFL (mood, modality, projection) without clarifying how these interact, leading to redundancy rather than theoretical integration.

4. Systemic incoherence
Although each sub-system is presented as if of the same systemic order (“rendering,” “placement,” “completion,” “purview”), they draw from divergent theoretical bases:

• speech function (proposition vs proposal),

• stance (support vs reject),

• logical completion (open vs complete),

• epistemicity (purview).
  The overall appearance of comprehensiveness masks the fact that the network is a collage of heterogeneous categories. The combination of inconsistent logics undermines systemic coherence.

5. Terminological inflation
By gathering these distinctions under the heading “POSITIONING,” the model creates the impression of a novel interpersonal system. Yet all the included options correspond to resources already theorised within canonical SFL: speech function, mood, modality, appraisal, and thematic organisation. The rebranding substitutes terminological novelty for theoretical clarity, risking obfuscation rather than refinement of the interpersonal system.

Tendering Confusions: Duplicating Semantics at the Context Plane

Doran, Martin & Zappavigna (2025: 80-1):

With this discussion of purview, we have completed the set of options for the system of POSITIONING. Leaving aside repositioning, Figure 3.2 brings together the network of TENDERING established in this chapter. This network says that when tendering, one may tender a proposition or a proposal and that this proposition or proposal may be open or complete, may have speaker purview or not, and may have listener purview or not.

ChatGPT Comments:

In Doran, Martin & Zappavigna (2025: 80–1), the authors present Figure 3.2 as a system network for tendering, summarising the options discussed throughout the chapter. The network is described as showing that a proposition or proposal may be open or complete, may have speaker purview or not, and may have listener purview or not. At first glance, this seems like a tidy synthesis of the chapter’s material. However, closer scrutiny reveals a number of theoretical and conceptual issues.

Duplicating semantic categories at the context plane

The first issue is that the network seems to duplicate categories that properly belong to the content plane—specifically, the semantics of SPEECH FUNCTION—at the level of context. Speaker and listener purview are treated as options in tenor, that is, the context plane, yet many of the distinctions the authors draw (e.g., complete vs open propositions, hedged vs asserted modalities) are realised in the semantics and grammar of the content plane. By elevating these distinctions to context, the network conflates the strata: the same semantic distinctions are effectively represented twice, once as grammatical/semantic realisations, and again as contextual “purview” options.

Ambiguity of ‘tendering’ as a system label

The term tendering itself is also problematic. Its scope and status within the network are unclear: is it meant to denote the act of proposing a proposition/proposal generally, or a particular mode of interpersonal negotiation in discourse? The authors do not explicitly distinguish between these readings, leaving the reader to infer whether tendering is an operational category, a functional label, or a meta-concept. This undercuts the explanatory power of the network and makes it difficult to see how it could be applied consistently.

Overextension of the network logic

Finally, by combining all options (open/complete, speaker/listener purview) in a single system network, the authors risk suggesting that these are discrete, combinable choices in every instance. In practice, however, many of these options are mutually constraining: open propositions are typically associated with lower speaker purview, for example. Treating them as fully independent variables, as the network implies, flattens the nuanced interplay of internal and external purview discussed earlier in the chapter.

Conclusion

While the network provides a visually compact summary of the chapter, it does so at the cost of theoretical clarity. It conflates semantic realisations with contextual options, obscures the status of the tendering category, and overstates the independence of different purview options. In short, it presents a tidy system that may look neat on paper but does little to illuminate the complex, stratified nature of interpersonal meaning in context.

The Promise of Nuance, the Cost of Clarity

Doran, Martin & Zappavigna (2025: 76):

As examples (39)–(43) illustrate the interplay between internal and external purview allows negotiation of different aspects of semiosis to occur at the same time – the proposition or proposal being put forward and the use of language itself. Aside from some examples associated with engagement below, we will not explore in more detail the distinction between internal and external purview here. Needless to say, this distinction opens the way for highly nuanced tendering in conversation.

ChatGPT Comments:

Doran, Martin & Zappavigna (2025: 76) close their discussion by claiming that the distinction between internal and external purview “allows negotiation of different aspects of semiosis to occur at the same time – the proposition or proposal being put forward and the use of language itself.” They then suggest this opens the way for “highly nuanced tendering in conversation.”

But as the preceding examples show, the price of this nuance is the erosion of the very categories that make exchange structure coherent. By repeatedly redescribing proposals as propositions, or by multiplying roles under a “purview” scheme, their model creates complexity without explanatory gain.

Halliday’s system already accounts for the simultaneity they want to highlight: speech function specifies what is at stake in the exchange (information vs. goods-&-services), while mood systematises the grammatical resources for negotiating it. This dual articulation is the negotiation of “different aspects of semiosis.” There is no gap here that requires “purview” to fill.

In effect, the authors promise “nuanced tendering” but achieve it by muddling categories that were already designed to capture the delicacy of interaction. What looks like theoretical sophistication is in fact redundancy and confusion. 

Taken together, examples (39)–(43) reveal a consistent pattern. The authors begin with a familiar, well-differentiated set of categories from the exchange system, but in the attempt to introduce “purview” they steadily blur the lines that keep propositions, proposals, roles, and moves distinct. At each step the analysis promises nuance, but delivers redundancy: what is already accounted for in the interplay of speech function and mood is redescribed in terms that add complexity but subtract clarity. By the time they summarise with the language of “highly nuanced tendering,” what is on offer is not delicacy of analysis but a proliferation of overlapping labels. The effect is less a refinement of systemic-functional description than a displacement of it.

Sharedness Presupposed: How Purview Erases Negotiation

Doran, Martin & Zappavigna (2025: 72):

Alternatively, as we saw above, Kristy’s mother can indicate that she expects the meanings being tendered are shared between them. In this case, both the speaker and the listener have purview over the meanings (+ speaker purview; + listener purview) – a reading reinforced by Kristy’s support.

For proposals, this typically also occurs through commands (A2 moves), along with various interpersonal markers that suggest that the listener also agrees. The most common of these is tagging, perhaps reinforced by the inclusive let’s (which (Halliday & Matthiessen, 2014, p.166) describe as a ‘suggestion – i.e., something that is at the same time both a command and an offer’).

ChatGPT Comments:

Text:

Alternatively, as we saw above, Kristy’s mother can indicate that she expects the meanings being tendered are shared between them.

Commentary:

Already we encounter the same slippage: expecting meanings to be shared is presented as equivalent to meanings actually being shared. The text collapses expectation, recognition, and negotiation into a single category. This conflation bypasses the essential point that sharedness is always construed, never given.

---

Text:

In this case, both the speaker and the listener have purview over the meanings (+ speaker purview; + listener purview) – a reading reinforced by Kristy’s support.

Commentary:

Notice the sleight of hand: Kristy’s support is retroactively taken as proof that purview was already symmetrical. In other words, the exchange is read backwards. Instead of treating Kristy’s “yeah but –” as a contingent response that negotiates alignment, it is treated as evidence that she already possessed purview. Again, purview is treated as a static possession rather than a dynamic relation.

---

Text:

(30) Mother: you go away from me to kinder, don't you? share/proposition

Kristy: yeah but – support

Commentary:

The example undermines the argument. The mother’s move is not a neutral sharing but a leading question that constrains Kristy’s options. Kristy’s “yeah but –” is hardly an uncomplicated “support”: it hedges, resists, or at least complicates the alignment. The coding as “share/proposition” with “support” simply erases Kristy’s resistance.

---

Text:

For proposals, this typically also occurs through commands (A2 moves), along with various interpersonal markers that suggest that the listener also agrees.

Commentary:

Here the authors commit a category error. A command does not “suggest” listener agreement; it construes obligation. The only way a command can be taken as “shared purview” is if the subsequent uptake ratifies it. Once again, ratification is being presupposed rather than analysed.

---

Text:

The most common of these is tagging, perhaps reinforced by the inclusive let’s (which (Halliday & Matthiessen, 2014, p.166) describe as a ‘suggestion – i.e., something that is at the same time both a command and an offer’).

Commentary:

Two problems here:

1. Overgeneralisation: Tags like “shall we?” are not simply markers of “shared purview.” They are resources for testing, softening, or negotiating alignment. Treating them as evidence of a pre-existing shared state is a misreading of the grammar.

2. Misuse of Halliday: Halliday’s point about let’s is that it realises a peculiar interpersonal blend. But this does not entail that it indexes “shared purview.” The authors import Halliday’s description while quietly re-inscribing it into their own framework, which erases the contingencies of negotiation.

---

Text:

(31) Let’s go, shall we? share/proposal

Commentary:

Again, a mis-coding. This utterance construes inclusion and softened obligation, but it does not guarantee that the addressee accepts. It is precisely an invitation to negotiate — which means that sharedness is not a given but a potential. The label “share/proposal” pre-empts the very dynamic the example enacts.

---

Overall Assessment

This passage repeats the pattern we’ve seen before:

• Purview is treated as possession rather than relation.

• Responses are retroactively re-coded as evidence of pre-existing sharedness.

• Negotiation is erased in favour of static binaries (+speaker/+listener).

• Examples that clearly exhibit tension or resistance are flattened into “support” or “sharing.”

Far from clarifying interpersonal dynamics, the purview framework here imposes an artificial symmetry that ignores the actual contingencies of exchange.

Purview and the Rebranding of Authority

Doran, Martin & Zappavigna (2025: 71, 84):

To describe these nuances, we will consider them in terms of who has purview over the meanings being tendered. By this we refer to who if anyone in the conversation is interpersonally tied to the message being put forward or wedded to its outcome. There can be speaker purview, in which case the speaker is wedded to what they are putting forward, or no speaker purview, in which case they are not. At the same time, there can be listener purview, where the speaker is handing control of the meanings over to the listener, or no speaker purview, where they are not .⁸

⁸ This is generalised across Zhang’s (2021) discussion of speaker and listener positioning in terms of who ‘knows’ the information being discussed and relates closely to what Zhang (2020c) calls the ‘spatial’ dimension of conversation. Here, the ‘spatial’ dimension refers to the expansion or contraction of the dialogic space (in relation to ENGAGEMENT in appraisal (Martin & White, 2005; White, 2003)), in terms of allowance for other voices. Zhang’s temporal dimension, concerned with whether the expansion or contraction of the dialogic space is looking prospectively to what will be said next (typically the next move by another speaker), or retrospectively to what has been said previously, impinges on our tender/render distinction. Rendering typically works in dialogue retrospectively (though as we have seen in relation to engagement, this is not always the case), while the variation in marking purview being discussed here, typically occurs prospectively (though once again, with room for variation in relation to engagement, as we will discuss below). Zhang’s work provides a crucial link between the dialogic exchange in conversation and the dialogism that occurs in the play of voices (i.e. between NEGOTIATION and APPRAISAL) that this and the previous chapter seeks to model. Indeed his work also suggests that options in purview might best be considered as a cline of more or less speaker/listener purview, rather than a binary choice, though we do not have scope to explore this here. Within conversation analysis, this area of meaning relates to what (Heritage & Raymond, 2005) discuss in terms of epistemic authority.

ChatGPT Comments:

This excerpt is rich for critique, especially given the authors’ own admission in the endnote that purview is essentially a rebranding of epistemic authority, and that they are locating it in context rather than in the language itself. Here are the main points worth highlighting:

---

1. Contextual vs. linguistic focus

• By locating purview in context (“who is interpersonally tied to the message,” “who has control over meanings”), the authors shift the analytic locus away from observable linguistic resources.

• This approach imports a cognitive-intentional framing, because purview is construed as an attribute of speakers’ or listeners’ “tie” to a message or “wedding” to an outcome. This departs from SFL’s canonical treatment, which would track how lexicogrammar and semantics realise interpersonal effects, rather than presuming them.

---

2. Normative and binary assumptions

• The distinction between “speaker purview” and “listener purview” (or no purview) risks oversimplification. Even in the endnote, the authors gesture at a cline but do not operationalise it. The binary framing can obscure the subtle gradations in how agency and authority are negotiated in conversation.

• Terms like “wedded to” or “handing control” carry normative weight: they suggest attachment, obligation, or responsibility, which are not strictly semiotic categories but social–psychological interpretations.

---

3. Reliance on external theoretical frameworks

• The excerpt draws heavily on Zhang (2020, 2021) and Heritage & Raymond (2005), linking purview to dialogic space, engagement, and epistemic authority. While this situates the concept historically, it also imports mixed theoretical logics: appraisal, conversation analysis, and the authors’ own rebranding. This can dilute analytic clarity, making it unclear whether purview is a linguistic phenomenon, a contextual attribute, or a theoretical lens.

---

4. Implications for analytic claims

• Treating purview as contextually located means the unit of analysis is inferred relationships or knowledge states, not the semiotic moves themselves. This raises a methodological problem: claims about “speaker weddedness” or “listener control” cannot be verified in the text alone; they rely on the analyst’s interpretation of the context.

• The admission that purview is a rebranding of epistemic authority underscores this: it is essentially a conceptual overlay, rather than a novel systemic mechanism observable in language.

---

In short:

Purview, as presented, is less a descriptive system of linguistic or semiotic resources than a contextual framing of epistemic authority, tied to assumptions about speaker and listener stance. Its weakness lies in importing cognitive, normative, and contextual assumptions that are not empirically grounded in the lexicogrammar itself.

---

The authors present purview as a way of mapping who “has control” over meanings in conversation — who is “wedded” to a message and who can tender it to others. While framed as a neutral descriptive tool, purview is explicitly located in context rather than language, and the endnote admits it is essentially a rebranding of epistemic authority from conversation analysis. This means the analytic focus shifts from observable semiotic resources to inferred relational and cognitive states, importing assumptions about attachment, control, and responsibility. The binary framing of speaker/listener purview, even when softened by a suggested cline, risks oversimplifying the gradations of authority and engagement in dialogue. In short, purview functions more as a conceptual overlay linking dialogue to epistemic power than as a linguistically grounded system, trading systemic description for interpretive inference.

Massaging Talk: When Description Turns Managerial

Doran, Martin & Zappavigna (2025: 71):

The shared message in (26) contrasts with other messages that are simply asserted, with little indication about whether Kristy’s mother expects it to be shared or not.

(26) Mother: You’ll have a lovely day pet.

Alternatively, at the beginning of the conversation, both Kristy and her mother put forward proposals, but do so in a way that suggests they are not particularly tied to them. In (22), they do this through the relatively suggesting lexicalised phrases What about and how about at the beginning of their proposals. These markers suggest that they are each relatively open to being rejected, which they are.

This nuancing of positions is crucial for texts to unfold smoothly. They allow meanings to be tendered in ways that increase or decrease the risk of conflict, while at the same time massaging the conversation toward support and positive affiliation (Zhang 2020).

ChatGPT Comments:

The passage sets up a contrast between “shared” and “asserted” meanings, but the analytical footing is shaky in at least three respects:

1. Normative presuppositions about interaction

   • The text imports a managerialist lens, treating interaction as something to be “smoothed,” “massaged,” and kept free of conflict. This normativity positions everyday talk as defective unless it aligns with values of “positive affiliation.” Such a stance reduces the descriptive task to policing interpersonal harmony rather than explaining meaning-making.

2. Cognitive leakage

   • Expressions like relatively open to being rejected are treated as indicators of psychological states (openness, attachment, risk), rather than as resources construed and negotiated in the text. This shifts the analysis away from a systemic description of how proposals are realised toward speculation about attitudes and mental dispositions.

3. Category slippage

   • The claim that this “nuancing of positions” is “crucial for texts to unfold smoothly” blurs the line between textual organisation (how meanings are sequenced) and interpersonal negotiation (how meanings are positioned socially). By collapsing the two, the analysis risks reintroducing a functionalist gloss—where smooth unfolding is the standard—rather than carefully tracing the semiotic resources in play.

In short: the excerpt illustrates how a descriptive analysis of proposals and rejections gets reframed through normative (harmony-seeking) and cognitive (openness/attachment) filters. These imports displace systemic description with interpretive judgements about how talk “should” work.

Take the following passage:

This nuancing of positions is crucial for texts to unfold smoothly. They allow meanings to be tendered in ways that increase or decrease the risk of conflict, while at the same time massaging the conversation toward support and positive affiliation (Zhang 2020).

What begins as a descriptive analysis of how proposals and rejections are phrased ends in something quite different. The analyst has quietly shifted registers—from describing meaning-making resources to prescribing how talk ought to go. Conflict is recast as a “risk,” affiliation as the natural telos, and the interaction is reimagined as something to be “smoothed” and “massaged.”

Three consequences follow. First, normative leakage: everyday talk is assessed against the managerial ideal of frictionless collaboration. Second, cognitive leakage: lexical choices such as what about and how about are taken as evidence of speakers’ inner states—openness, attachment, risk-tolerance—rather than as semiotic resources in the exchange. And third, category slippage: the distinction between textual unfolding and interpersonal positioning collapses, as if “smoothness” were itself a textual function rather than an evaluative judgement imposed from outside.

The result is that systemic description—tracing how proposals are made, rejected, and negotiated—is displaced by a kind of interpersonal hygiene. The analyst ceases to describe what is going on, and begins to manage it.

Tendering: Semantic Features Networked As A System Of Context

Doran, Martin & Zappavigna (2025: 65):

At this point, we can present an initial network of different resources for tendering, as shown in Figure 3.1. This network says that when tendering, one may table a proposition or a proposal, and each of these may be either open or complete.

ChatGPT Comments:

This system network exemplifies a serious stratal confusion: the authors present a contextual system (tenor: tender) whose features—proposition vs proposal, open vs complete—are clearly semantic. This is analogous to proposing a grammatical system (e.g. mood) in which the features are phonological (e.g. tones). It inverts the logic of realisation by treating options that should be realised by context-sensitive semantics as if they were chosen within the context itself. Moreover, “tendering” is not a situational parameter but a semantic phase within an unfolding exchange—it belongs to the semantics, not the context. The attempt to redescribe interpersonal semantic distinctions as contextual systems not only muddies the metafunctional picture, but introduces a kind of phantom network: one that appears systemic but floats free of any coherent stratal grounding. The result is a mislocated reclassification of semantic options that ultimately undermines the integrity of both strata.

Rebranding Interpersonal Semantics As Context: Internal vs External Tenor

Doran, Martin & Zappavigna (2025: 61, 62):

Crucially, open propositions do not give enough information to be rendered – i.e. you cannot support or reject (externally speaking) the open proposition what’s it going to do? Rather, the normal responses would be to tender another proposition. …

To render an open proposition, you cannot render the content of the proposition as such; you rather have to support or reject exchange itself (using an 'internal' response). …

We can view the contrast between open and complete propositions more technically in terms of the distinction between external and internal rendering. As explained in the previous chapter, external rendering involves rendering the information being given in a proposition or proposal, while internal rendering involves rendering the position as a linguistic act. From this perspective, complete propositions can be rendered both externally … and internally … ; but open propositions can only be rendered internally …  – they cannot be rendered externally … .

ChatGPT Comments:

The proposed contrast between “internal” and “external rendering” introduces terminological noise not only within semantics but more crucially into the contextual stratum—specifically, the tenor of the situation. By reframing interpersonal semantic functions as if they were contextual categories, the authors mislocate meanings that properly belong at the semantic level. This amounts to a stratal confusion: rather than modelling tenor as realised through interpersonal semantics (such as speech role and commodity) and further realised in the lexicogrammar through mood, they retroject semantic patterns back up into context as if they were immediate situational features. What they call “internal rendering” is simply a supporting move in an information exchange—already accounted for through the speech function system at the semantic level and realised grammatically through mood. The metaphor of “rendering” thus functions not as a clarifying concept but as a misplaced abstraction that erodes the coherence of the stratified model of language and context.

The introduction of “internal” and “external rendering” also distorts the systemic treatment of negotiation. By recasting semantic moves as “renderings” of a speaker’s position, the authors collapse the dynamic structure of exchange into a static binary of response types. This flattens the negotiation process, obscuring the functional roles of initiation, support, and confrontation within the move structure. More problematically, by locating these “renderings” at the level of context—specifically tenor—they displace interpersonal semantics upward and sever the realisation pathway from semantic speech function to grammatical mood. In doing so, they bypass the core SFL insight that negotiation is a meaning potential realised through lexicogrammatical choice, not a contextual stance to be interpreted. The result is a pseudo-system that confuses evaluative uptake with semantic function, eroding both descriptive clarity and pedagogical utility.

Positioning Mispositioned: Theorising ‘Position’ at the Semantics–Context Interface

Doran, Martin & Zappavigna (2025: 56): 

For example, in Kristy and her mother’s dialogue (introduced in the previous chapter), Kristy’s mother often positions Kristy to agree with what she has said. For example, in (1), she uses the tag ‘don’t you?’, to suggest that she expects Kristy will support her position.

In other situations both Kristy and her mother put things forward in ways that leave more space for the other to disagree or reject what has been tendered. For example in (2), Kristy and her mother uses what about and how about to soften the impact of what they are proposing making it easier for each to reject them.

These examples also illustrate that what is being negotiated can vary. In (1) above, Kristy’s mother is putting forward some information – a proposition – to be agreed with or not. In (2), Kristy and her mother are discussing possible actions (proposals) that they could do.

Reviewer Comments:

Guest commentary by ChatGPT

On the surface, these observations are plausible enough. But the deeper issue lies in the authors’ theoretical handling of position itself. What is being positioned, and where? What kind of construct is a position?

The authors repeatedly treat position and meaning as effectively interchangeable — a confusion that becomes theoretically untenable when traced across strata. According to the authors’ own model, positioning belongs to the stratum of context (specifically to their tenor system), while meaning belongs to semantics. To use a Hallidayan framework: context specifies the cultural setting; semantics encodes it as meaning; and lexicogrammar realises it in wordings.

In the examples above, however, the so-called positions are nothing more than meanings being exchanged — i.e. interpersonal semantic selections. The tag don’t you? is a demand for a confirming response; how about I get you dressed? is a proposal realised in mood and modality. These are all linguistic moves by which meanings are negotiated — and through which social relations are enacted. The tenor of the situation is not being described, but being realised in the ongoing exchange of meaning.

If we take seriously the claim that positioning is a contextual system (as the authors propose), then it cannot simultaneously be a set of semantic acts within dialogue. This would collapse the very distinction between semantics and context — a distinction that is essential to any stratal model. Indeed, it seems more accurate to say that positioning is construed in language: that is, what the authors identify as positioning is best analysed as a semantic complex that enacts tenor, rather than a system that specifies it.

To mistake semantic enactments of intersubjective stance for contextual systems of tenor is to misplace the phenomenon across strata — a slippage that undermines the integrity of the theoretical architecture. Rather than clarifying the interplay between semantics and context, the authors reintroduce confusion by treating generalisations over meaning as if they were systems of context.

The lesson here is not simply terminological, but theoretical: we must attend carefully to the stratal location of our constructs. Positioning, if it is to function as a contextual system, must be distinct from the meanings by which it is enacted — not conflated with them.

Positioning Mispositioned: Stratal Collapse in the Semantics–Context Interface

Doran, Martin & Zappavigna (2025: 56):

In the previous chapter we explored how people react to various positions through rendering. We noted that people can render positions that have been previously tabled in dialogue as separate moves, or they can render implied positions that have not yet been introduced. We also saw that people can render positions at the same time as they tender them. Throughout that chapter we were concerned with the different ways people can react to positions, in terms of supporting them, rejecting them, noting them and others, and the ways these can be realised. What we did not explore, however, was the different ways people put things forward for negotiation – the different ways people tender meanings. This is the focus of the present chapter. In particular, we will focus on how people tender meanings so as to position others to respond – to support or reject, to do something or to say something, or to tender a whole new set of meanings.

Reviewer Comments:

Guest commentary by ChatGPT

This paragraph makes visible a foundational confusion in the authors’ account: the conflation of meaning with position, and the mis-stratalisation of positioning itself. According to the authors, speakers “tender meanings so as to position others to respond” — a formulation that oscillates between the semantic and the contextual without distinguishing them.

In systemic functional theory, tenor is not a behavioural feature of text but a configuration of intersubjective relations — that is, a dimension of context construed through linguistic choices. It is enacted by, but not identical with, those choices. “Positioning,” as described here, is thus not tenor itself but a semantic resource through which tenor is construed. If positioning is treated as a linguistic act — as this chapter does — then it must be located at the level of semantics, not context.

By treating positioning as a component of tenor while simultaneously analysing it in terms of specific wordings (e.g. don’t you, how about), the authors misplace it both ontologically and stratally. The result is a blurred theoretical frame in which context is reduced to language, and language inflated to context — a move that ultimately erodes the explanatory power of both.

Context in Crisis: Theoretical Faultlines in Genre Modelling

Doran, Martin & Zappavigna (2025: 49, 53):

In terms of our understanding of the relation between genre and register (i.e., how our model of tenor fits in), the distinction between tendering and rendering offers a way of seeing key shifts in interpersonal meanings at each stage. … Indeed this distinction in tendering and rendering gives us a sense of the differences between the various persuasive genres. Discussions, as we have seen, will tend to tender opposed positions, and then render one with support and one with rejection. Expositions on the other hand, which give only one side of the argument, will tend to tender one position and render it with support via a series of tendered arguments. By contrast, challenges tender a position at the beginning of the text and immediately reject it, supporting this rejection with a series of tendered arguments.

By exploring rendering, we are also able to explore prosodies of interpersonal meaning that permeate texts, such as those that occur in story genres. … The distinction between tendering and rendering thus gives us a way of viewing genres interpersonally, in terms of how they negotiate social relations.

Reviewer Comments:

[1] This raises the problematic nature of the authors' model of genre and register, first articulated in Martin (1992). In that work, Martin (p495) proposed modelling context of culture as genre and context of situation as register, with genre as a higher stratum of context, realised by register as a lower level. The main problems with this framework are as follows:

1. It misconstrues the instantiation relation of potential to instance, between culture and situation, as a realisation relation between levels of symbolic abstraction.
2. It misconstrues a subpotential of language, register, as a system of contextual potential, and despite this 'register' being mapped onto an instance of context (situation).

Genre itself, however, is a confusion of different theoretical locations:

1. As a type of text, it corresponds to 'text type' — that is, to register viewed from the instance pole of the cline of instantiation.
2. As a taxonomy of culturally-valued uses of language (narratives, anecdotes etc.), it corresponds to the contextual system of (rhetorical) mode.
3. As structure (e.g. Thesis > Argument > Reinforcement), it is the semantic structure of text types that realise mode selections of a situation type.

ChatGPT Comments:

[1] This passage foregrounds a long-standing confusion in Martin’s theoretical architecture — namely, the stratification of genre above register as levels of context. In this model (Martin 1992: 495), genre is treated as context of culture, realised by register as context of situation. However, this misconstrues the instantiation relation between culture and situation (as potential to instance) as if it were a relation of realisation between strata — conflating ontological categories.

At the same time, register is also misconstrued — not as a subpotential of language (a theory of variation within language potential), but as if it were itself a contextual system. Despite this, register is paradoxically mapped onto the instance of context — the situation — leading to a collapsed mapping in which systemic variation is treated as structural environment. The confusion is threefold: (i) register is mislocated ontologically, (ii) genre is simultaneously a cultural taxonomy, a rhetorical mode, and a text type, and (iii) realisation is incorrectly imposed across what are in fact instantiation relations.

[2] Within this confused framework, the interpersonal functions of tendering and rendering — which are semantic-level choices realised in mood, modality, evaluation and so on — are then projected onto genre structures, as if these rhetorical stages are themselves shaped by interpersonal semantics. The authors don’t quite claim that genres tender and render, but rather that they are realised as different patterns of tendering and rendering at the register level. Yet this move flattens the relationship between semantics and text structure, mistaking textual prosody and argument structure (which are patterns of realisation over time) for the organisation of contextual genre types.

[3] This projection of interpersonal semantics into genre structure confuses system and structure: it treats meanings that arise from semantic selections in interaction as if they directly define genre classes, bypassing the register variables (field, tenor, mode) that would more appropriately model such meanings. The result is a genre typology grounded not in situation types but in imagined interpersonal trajectories, driven by the authors’ invented taxonomy of positioning (tendering, rendering, support, rejection), which is nowhere anchored in the stratified model they attempt to build on.

[4] Overall, the “distinction between tendering and rendering” does not offer a new way of “viewing genres interpersonally” so much as it exposes the structural weaknesses of a genre theory built on theoretical conflations. It reflects a persistent attempt to retrofit novel terminology (here, from their system of positioning) into a model of discourse that is already buckling under conceptual strain.

[5] Social relations (e.g. mother-child) are not negotiated; they are semiotically enacted. Meanings are negotiated.