Deriving A System From Structural Realisations

Martin (2013: 43-4):

Blogger Comments:

[1] Again, this misrepresents SFL methodology. Again Martin constructs a system on the basis of how it is realised structurally. This is giving priority to the view 'from below', instead of the view 'from above'. Again, he explicitly distinguishes mood types on the basis of their structural realisations instead of in terms of the speech functions they congruently realise.

[2] To be clear, to specify which interpersonal function the interpersonal WH- element conflates with, it is not necessary to incorporate choices beyond the interpersonal MOOD system. For example, Halliday & Matthiessen (2014: 162):

Distinguishing Mood By Structural Realisation Instead Of The Meaning It Realises

Martin (2013: 43):

Blogger Comments:

Again, this misrepresents SFL methodology. Again Martin constructs a system on the basis of how it is realised structurally. This is giving priority to the view 'from below', instead of the view 'from above'. Here he also explicitly distinguishes mood types on the basis of their structural realisations instead of in terms of the speech functions they congruently realise.

And again, the inclusion of the verbal group feature 'non-finite' in this clause system confuses ranks, and misrepresents imperative clauses, which can include Finite and Subject elements, as non-finite.

Misrepresenting The Finiteness Of Imperative Clauses

Martin (2013: 42-3):

Blogger Comments:

Again, this misrepresents SFL methodology. Here Martin constructs a system on the basis of how it is realised structurally. This is giving priority to the view 'from below', instead of the view 'from above'.

A further theoretical inconsistency is the inclusion of the verbal group feature 'non-finite' in this clause system. As well as confusing ranks, imperative clauses are finite clauses and can include Finite and Subject elements. Halliday & Matthiessen (2014: 165):

In the analysis, the unmarked positive has no Mood element, it consists of Residue (Predicator): the verb form (e.g. look) is Predicator only, with no Finite in it. The other forms have a Mood element; this consists of Subject only (you), Finite only (do, don’t), or Finite followed by Subject (don’t you). Any of these can be followed by a Mood tag: won’t you?, will you? – showing that the clause is finite, even though the verb is non-finite (the imperative of be is be, as in Be quiet!, not the finite form are.

Devising A System 'From Below' Instead Of 'From Above'

Martin (2013: 41, 42):

Blogger Comments:

To be clear, this misrepresents SFL methodology. Here Martin constructs a system on the basis of how it is realised structurally. This is giving priority to the view 'from below'. In SFL Theory, priority is given to the view 'from above', which, in the case of the MOOD system, is the interpersonal meanings being realised: the clause as exchange. The meanings realised by the grammatical system of MOOD are those of the semantic system of SPEECH FUNCTION: giving vs demanding information vs goods–&–services.

It is from this mistaken inversion of SFL methodology that the misunderstanding of "axial relations" in this monograph derives. 

Confusing Function With Form

 Martin (2013: 40):

Blogger Comments:

To be clear, in SFL Theory, features of systems, such as the MOOD system, are not classes. It is form, not function, that is categorised into classes; e.g. word classes, group classes.

Misunderstanding Redundancy And Structure

Martin (2013: 40):

Technically speaking, as noted above, a sequence of classes redounding with one another is referred to as a syntagm. Redounding functions are referred to as structures. … By 'redound' we mean that the classes of functions are redundant with respect to one another — in other words they are mutually expectant (given one class or function the co-appearance of another is not random). This is simply a more technical way to say that the classes form a syntagm or that the functions form a structure and so make a meaning beyond the sum of their parts.

Blogger Comments:

[1] This misunderstands the notion of redundancy. To be clear, 'redundancy' refers to the skewing of probabilities away from equiprobability. Halliday (2003 [1987]: 122):

In an ideal system, one having two states that are equiprobable, there is no redundancy. Once we depart from equiprobability, redundancy sets in. In all open systems the probabilities are skewed, so that the system carries redundancy.

In SFL Theory, 'redundancy' is used to refer to realisation relations between strata. Halliday (2002 [1992]: 356):

Consider a minimal semiotic system, such as a protolanguage – a system that is made up of simple signs. This is based on the principle of redundancy. When we say that contents p, q, r are “realised” respectively by expressions a, b, c, what this means is that there is a redundancy relation between them: given meaning p, we can predict sound or gesture a, and given sound or gesture a we can predict meaning p. This relationship is symmetrical; “redounds with” is equivalent both to “realises” and to “is realised by”.

with these probabilities variably skewed along the cline of instantiation. Halliday (2002 [1992]: 359):

The [language] system is permeable because each instance redounds with the context of situation, and so perturbs the system in interaction with the environment.

Halliday (2003 [1997]: 260):

If we consider register variation first: viewing from the "instance" end, we can recognise a text type as a collection of similar instances. But when we shift perspective and see it as systemic variation, each of these text types appears as a register, a kind of subsystem which redounds with the properties of the context in terms of field, tenor and mode.

Martin, however, misunderstands the skewed probabilities of redundancy as just any mutual expectancy, and applies this to the sequencing of items on the syntagmatic axis.

[2] As demonstrated above, redundancy is not a more technical way of saying that syntagms and structures "make a meaning beyond the sum of their parts". The crucial point that Martin omits is that, just as a system is the relationships among its features, a structure is the relationships among its functions. Halliday & Matthiessen (2014: 451):

Note that, although it is the functions that are labelled, the structure actually consists of the relationships among them.

Confusing Mood Adjuncts With Comment Adjuncts

Martin (2013: 37, 39):

And in other contexts verbs are used to classify (and are thus not gradable):

running shoes

industrialised nations

… Imperative clauses, giving commands, are typically indicated in English by the absence of a Finite and a Subject function.

In addition Halliday recognises Modal Adjuncts which add an interpersonal comment of some kind to the clause (a modalisation of probability below) …

Blogger Comments:

[1] To be clear, industrialised is gradable: there are less industrialised nations and more industrialised nations.

[2] To be clear, 'giving' is a poor choice of words, pedagogically, since commands involve demanding goods-&-services, as opposed to giving them.

[3] This is misleading, because, within the category of modal Adjunct, SFL Theory distinguishes between mood Adjuncts and comment Adjuncts. Halliday & Matthiessen (2014: 184):

We can recognise two types of modal Adjuncts, (i) mood Adjuncts and (ii) comment Adjuncts. (i) Mood Adjuncts serve within the Mood element, and are closely associated with the meaning of the Finite element – the limiting case being modality, which (as we have seen) can also be realised by the operator serving as Finite. (ii) Comment Adjuncts serve outside the Mood + Residue structure of the clause. They are not part of the proposition realised by Mood + Residue, but are instead comments on it (propositional) or on the act of exchanging it (speech-functional).

Misrepresenting Single Forms As Function Structures

Martin (2013: 28, 31):

… we have been dealing with pronoun systems realisation by single unit structures (by single words) above …

In the next chapter we'll move on to consider system networks realised by structures consisting of more than one element …

Blogger Comments:

This is misleading because it is not true. Martin has not even been dealing with functional elements of structure, let alone structures. He has only been dealing with forms, pronouns, and devising networks by giving priority to form over function in doing so — the opposite approach to that taken in SFL Theory. To be clear, a structure is the relationships among functions. Halliday & Matthiessen (2014: 451):

Note that, although it is the functions that are labelled, the structure actually consists of the relationships among them.

Comparing Tagalog And English Person & Number Systems

Martin (2013: 24-5):

Our system network formulation of the revised paradigm is presented below, with PERSON and NUMBER as simultaneous systems, but with a different set of oppositions than those we distinguished when working on the English systems. From this we can see that although English and Tagalog have comparable regions of meaning as far as their pronouns grammar is concerned, no pronoun in one of the two languages can ever be the exact equivalent of a pronoun in the other — because the valeur, to use Saussure's term, is different. The systems are comparable but not equivalent.

Blogger Comments:

[1] To be clear, Martin's revised paradigm is just Schachter & Otanes' paradigm with different labels:

Martin has replaced Schachter & Otanes' 'non-plural' and 'plural' with Schachter & Otanes' gloss of their own labels, and replaced their formal categories of person with Halliday's functional categories, as previously explained.

[2] This is misleading. Because Martin used formal categories for English and functional categories for Tagalog, the comparability of the meanings is obscured. For example, if English were presented in these functional terms, the only difference between the two would be the absence of the "speaker/addressee" feature in English.

Note that, from an SFL perspective, English does make a distinction in meaning between 'only' and '+others' for 'addressee', it is just that the same form you realises both feature combinations. The distinction in meaning is demonstrated, for example, by the dialectal use of youse and y'all as realisations for the '+others' meaning.

Misunderstanding Halliday On The Ineffability Of Grammatical Categories

Martin (2013: 24):

Blogger Comments:

This is misleading, because On The Ineffability Of Grammatical Categories (Halliday 1984) is not at all about a balancing act of 'respecting both terminological traditions and the language to hand'. Instead, the paper explains that grammatical categories can only mean themselves. That is, while it is possible to provide an encoding definition of a grammatical category (how it is expressed), it is not possible to provide a decoding definition of a grammatical category (what it means). Halliday (2002 [1984]: 292-3, 303, 307, 309):

Once having reified these abstract categories by naming them, the Greek grammarians went on to ask what the names meant. What ‘is’ a noun? they wondered. At first, this was a question of: how do I recognise a noun when I see it? how do I know whether something is a noun or not? But before long the questions came to be asked in the other direction: what ‘is’ a noun, in the sense of what function does it serve? At this second stage, instead of treating ‘noun’ as the Value and then supplying a Token for it, the definition now treats ‘noun’ as the Token and seeks to supply a Value for it. Instead of ‘a noun is a word that inflects for number and case’, we have ‘a noun is the name of a person, creature or thing’. This is a decoding definition, one which embodies a notion of ‘what the category means’. … But how does one define by decoding? how do we say what a grammatical category means? …

The meaning of a typical grammatical category thus has no counterpart in our conscious representation of things. There can be no exact paraphrase of Subject or Actor or Theme – because there is no language-independent clustering of phenomena in our experience to which they correspond. If there was, we should not need the linguistic category to create one. If language was a purely passive partner, ‘expressing’ a ‘reality’ that was already there, its categories would be eminently glossable. But it is not. Language is an active participant in the semogenic process. Language creates reality – and therefore its categories of content cannot be defined, since we could define them only by relating them to some pre-existing model of experience, and there is no model of experience until the linguistic categories are there to model it. The only meaning of Subject is the meaning that has evolved along with the category itself. …

To understand these categories, it is no use asking what they mean. The question is not ‘what is the meaning of this or that function or feature in the grammar?’; but rather ‘what is encoded in this language, or in this register (functional variety) of the language?’ This reverses the perspective derived from the history of linguistics, in which a language is a system of forms, with meanings attached to make sense of them. Instead, a language is treated as a system of meanings, with forms attached to express them. Not grammatical paradigms with their interpretation, but semantic paradigms with their realisations. …

What I have been trying to show with this illustration is that while, with a category like Subject, it is impossible to answer the question ‘what does it mean?’, this does not signify that it has no meaning. The problem of ineffability is common to all grammatical categories; there are various reasons why some may seem less problematic than others, but it is an illusion to think that any can be exhaustively defined. And this, as I remarked above, is not because of the shortcomings of natural language for serving as a metalanguage, real though such shortcomings are. Rather the converse: it is the very richness of natural language, its power of distilling the entire collective experience of the culture into a single manageable, and learnable, code that puts its categories beyond the reach of our conscious attempts at exegesis.

Misrepresenting Others' Work As Martin's Invention

Martin (2013: 24-5):

We've adjusted the labelling for the paradigm below, to better capture the proportionalities involved. Instead of using traditional PERSON and NUMBER labels we've turned to discourse semantics, and chosen more transparent terms to capture what is going on. One problem with this for some linguists is that our labels now sound more contextual than grammatical, and formal grammarians might object to them on such grounds; and they are perhaps less familiar to linguists than the traditional ones, and so outsiders might find them harder to understand. …

Blogger Comments:

[1] This is very misleading indeed, because discourse semantics is Martin's model, and so claiming to be drawing on discourse semantics here is claiming that this labelling is Martin's work, whereas the truth is that it is not. On the one hand, the PERSON categories are from Halliday's lexicogrammar, not Martin's discourse semantics. For example, Halliday & Matthiessen (2004: 40, 41):

… there is a system of PERSON, based (in English, as in most other languages) on the opposition of ‘you-&-me’ versus ‘everyone (and perhaps everything) else’, and then on that of ‘you’ as opposed to ‘me’ (see Figure 2-2). …

On the other hand, the NUMBER categories are derived from Martin's source, Schachter & Otanes (1972: 88), not Martin's discourse semantics:

The plural pronouns have, as a group, the meanings of the corresponding non-plural pronouns with the additional meaning 'and others' (or 'and another'): thus the first-person plural is literally 'I and others', the dual plural 'you (singular) and I and others', etc.

[2] To be clear, the claim here is that a problem with more transparent terms is that they might be harder to understand.

[3] Trivially, in SFL Theory, 'context' is the culture modelled as a semiotic system, and formal grammarians do not use 'context' in this sense, so the labels that Martin has falsely claimed to be his own would not look more 'contextual' to them, and so would not be their grounds for objecting to them.

Misrepresenting A Formal Approach To Tagalog Pronoun Classification

Martin (2013: 23-4):

Tagalog makes a distinction between what is generally referred to as inclusive and exclusive 'we' — a distinction in other words between reference to the speaker and one or more others, including the addressee, and the speaker and one or more others, excluding the addressee. … Schachter & Otanes (1972) label the paradigm they propose as follows:

Their labelling raises two questions. The first issue is the choice of the label non-plural instead of singular; this is presumably because the pronoun kata, which refers to the speaker and one addressee, involves more than one person (so is not singular). The second concern is the label dual; this is a rather odd choice of term, since dual is usually used as part of a NUMBER system ([singular], [dual], [trial], [paucal], [plural] etc.) not a PERSON system; and in any case, the 'plural' inclusive pronoun tayo refers to more than two people (the speaker, addressee, and one or more others). But we can see the problem Schachter & Otanes have here: the categories they inherited from the Greco-Roman tradition don't fit — because Tagalog has four persons not three.

Blogger Comments:

[1] To be clear, there is no need for Martin to presume this about the label 'non-plural' because Schachter & Otanes (1972: 88) are quite explicit about it:

The first group is called non-plural rather than singular because it included dual pronouns, meaning literally 'you (singular) and I'.

[2] To be clear, this misleads by ignoring the explanation given by the authors. Schachter & Otanes (1972: 88) explain the reasoning behind their number proportionalities as follows:

The plural pronouns have, as a group, the meanings of the corresponding non-plural pronouns with the additional meaning 'and others' (or 'and another'): thus the first-person plural is literally 'I and others', the dual plural 'you (singular) and I and others', etc.

Martin's purpose in ignoring the explanation given by the authors will be seen in the following post.

[3] This is misleading. An alternative Formal approach to classifying personal pronouns does "fit":

[4] This is misleading because it is not true. Tagalog has three persons, not four, since, as Martin himself acknowledges, 'dual' is a number category, not a person category.

Misrepesenting The 'General Disposition' Of Grammarians

Martin (2013: 23):

Blogger Comments:

To be clear, the treatment of you as an outsider and it as an exception reflects the disposition of Martin to misrepresent a systemic approach to formal grammar as a systemic approach to functional grammar. In the next section, on Tagalog pronouns, where he switches to functional features, he will falsely claim that this will be taking a discourse semantic approach, and so Martinian, rather than a grammatical approach, and so Hallidayan.

Classifying Forms (Personal Pronouns) In Terms Of Formal Categories (Case)

Martin (2013: 22-3):

As far as NUMBER and GENDER are concerned it is a regular part of the system; but as far as CASE is concerned it is an exception. The usual strategy for showing exceptions of this kind in SFL is to use a marking convention that blocks specific combinations of features. We can read the 'I' superscript on the feature [neuter] along these lines and connect it to the 'T' superscript on [nominative] (read as 'if [neuter] then [nominative]', and so not [accusative]). Note that this means arbitrarily describing it as [nominative], and lacking an accusative form. …  

CASE is introduced as a third system, simultaneous with NUMBER and GENDER below. The if/then marking convention has been deployed in the network, but realisations have been informally provided as a list, rather than being tied to specific features (except for you).

 

Blogger Comments:

[1] To be clear, this is again classifying form (a pronoun) in terms of formal categories (case) instead of systemising functions.

[2] Note that this means misrepresenting the data, since it is the accusative form as well as the nominative form. That is, it is a misuse of the conditional marking.

[3] To be clear, these are not functions, but forms specified by formal categories. Martin's motivation for modelling forms instead of functions will become clear in the review of his Tagalog PERSON systems.

Giving Priority To Form Over Function

Martin (2013: 22):

As a final step we can consider adding CASE to the system, and so distinguish between pronouns which reference the nub of the argument (grammatically the Subject of a clause) and those which don't (grammatically the Complement of a clause or prepositional phrase, or the 'subject' of a non-finite clause) — cf. She noticed him texting me for them.

I, she, he, it, you, we, they

me, him, her, us them

Fortunately you remains the odd one out, since it doesn't distinguish either NUMBER or CASE. That effectively leaves us with the problem of it, which is [singular] and [neuter], as we saw above. As far as NUMBER and GENDER are concerned it is a regular part of the system; but as far as CASE is concerned it is an exception.

 Blogger Comments:

[1] To be clear, 'case' is a formal category, not a functional one, so it is misleading to present it as a system in a work purporting to be 'a next step' into Systemic Functional Grammar.

[2] To be clear, distinguishing between pronouns is distinguishing between forms, not functions, so it is misleading to present them as features in a system in a work purporting to be 'a next step' into Systemic Functional Grammar.

[3] This is misleading. Pronouns don't reference the Subject; the nominal group in which a pronoun figures realises the Subject. Such a pronoun references an entity elsewhere — either (endophorically) in the text, or (exophorically) in the environment of a text.

[4] This is misleading because it uses the opposite methodology to that of Systemic Functional Grammar. That is, it gives priority to the view 'from below', form, instead of the view 'from above', function.