Misrepresenting SFL Theory On Paradigmatic Relations

Fawcett (2010: 43n):

If you are not a systemic functional linguist, you may be asking at this point: "Why do systemic functional linguists give priority to paradigmatic relations between meanings rather than forms?" It is a good question, and it may be helpful to say briefly what my answer is. Ultimately, it is because generating a text involves making choices, and it is clearly the contrasts between alternative meanings between which we choose — rather than the contrasts between the forms. For example, if two outputs from the grammar display a contrast in form, as between that student and those students, the importance of the contrast is that the two forms express a contrast in meaning which the Performer wishes to communicate to the Addressee. In other words, the difference between 'singular' and 'plural' is ultimately a difference of meaning rather than form. (But there is, of course, no meaning without form.)

Blogger Comments:

This is misleading.  SFL theory gives priority to view from above, from meaning (semantics), in modelling the grammar (wording), but it nevertheless distinguishes between paradigmatic relations at the level of meaning (semantics) and paradigmatic relations at the level of wording (lexicogrammar).  Halliday & Matthiessen (2014: 49) explain:

Being a ‘functional grammar’ means that priority is given to the view ‘from above’; that is, grammar is seen as a resource for making meaning – it is a semanticky kind of grammar. But the focus of attention is still on the grammar itself.  Giving priority to the view ‘from above’ means that the organising principle adopted is that of system: the grammar is seen as a network of interrelated meaningful choices. In other words, the dominant axis is the paradigmatic one: the fundamental components of the grammar are sets of mutually defining contrastive features (for an early statement, see Halliday, 1966a). Explaining something consists not in stating how it is structured but in showing how it is related to other things: its pattern of systemic relationships, or agnateness …

On the other hand, SFL theory includes form in the paradigmatic relations at the level of wording (lexicogrammar) in the guise of the rank scale — clause, group/phrase, word, morpheme — each of which provides the entry condition for systems of paradigmatic relations between functions.

Misrepresenting Halliday On Form [2]

Fawcett (2010: 42-3):

The second key point is that, while it is incontestable that there are relations of contrast at the level of form, and while Halliday's concept of 'system' in "Categories" was, like that of Firth, a system of contrasts at the level of form, in a modern SF grammar the system networks model choices between meanings. And it is these that are seen as the generative base of the grammar. The result is that the purely formal contrasts in a language play no role in how the grammar operates in the generation of a sentence. … Thus choice between meanings is the key concept in a systemic functional grammar. However, the focus of this book is on the level of form, so I shall have very little more to say about the system networks.

Blogger Comments:

[1] This is misleading.  In Categories, Halliday (2006 [1961]: 39) uses 'form' in a different sense to that used by Fawcett:

The form is the organisation of the substance into meaningful events …

[2] This is misleading.  In "a modern Systemic Functional grammar", system networks model choices on all linguistic strata: meaning (semantics), wording (lexicogrammar) and sounding (phonology).  The lexicogrammatical networks model functional wording choices at each of the levels of form on the rank scale.  In the absence of grammatical metaphor, those functional choices at the level of wording agree (are congruent) with functional choices at the level of meaning.

[3] In "a modern Systemic Functional grammar", since form realises function, contrasts in form can realise significant contrasts in function, most notably in instances of grammatical metaphor, where what would congruently be realised by a clause is instead realised incongruently as a nominal group.  A major shortcoming of the Cardiff Grammar is its inability to systematically account for grammatical metaphor.

[4] To be clear, the focus of this book on Systemic Functional grammar is on neither system nor function.

What Fawcett Understands By 'Paradigmatic Relations'

Fawcett (2010: 42):

Paradigmatic relations are relations of contrast. There are two key points that must be made about them. Firstly, paradigmatic relations are unlike syntagmatic relations in that they exist only in the potential and never in an instance.  From the viewpoint of the text analyst, they express a contrast between (1) the meaning (and so the form) that was chosen for use in the text and (2) the one or more meanings (and so forms) that might have been chosen (but were not). In other words, paradigmatic relations exist only in the language that is used to produce a text-sentence — and not in the sentence itself.

Blogger Comments:

This continues Fawcett's confusion of the realisation relation between the paradigmatic axis (system) and the syntagmatic axis (structure) with the instantiation relation between potential and instance.  As previously explained, in Fawcett's model, schematised as Figure 4, paradigmatic system is equated with (meaning level) potential, and syntagmatic structure is equated with (form level) instance.

Fawcett On The Interplay Of Realisation And Instantiation In A Systemic Functional Grammar

Fawcett (2010: 41-2):

Let us return to Figure 4. Its significance is that it brings together, in a single diagram, two key pairs of concepts that correspond, broadly speaking, to two pairs of Saussurean concepts: meaning and form, and potential and instance. In a systemic functional grammar, meaning and form are related by the general relationship of realisation but, as we have seen, this relationship does not operate directly. Instead, it operates via the concept of instantiation. Instantiation occurs first at the level of 'meaning, when a traversal of the system network generates a selection expression of features, i.e. what Halliday has called an 'act of meaning' (Halliday 1993:4505). Then the realisation rules that specify the 'form potential' come into play and act upon the selection expression to realise it, and the final output from the grammar is the generation of a second 'instance', i.e., one unit that adds a layer of structure to the 'tree' representation of a text-sentence that is being built. 

Taken together, these concepts model the basic components of a systemic functional grammar, so that Figure 4 represents, at a fairly high level of abstraction, the main components of the model of language within which the alternative current theories of syntax in SFL can be set.

Blogger Comments:

This continues the discussion of Figure 4:

[1] This is misleading.  In a systemic functional grammar, stratal realisation doesn't "operate" and it doesn't do so "via the concept of instantiation"; this is Fawcett's misunderstanding only.  In a systemic functional grammar, realisation is an identifying relation between two levels of symbolic abstraction.  The notion of realisation "operating" derives from Fawcett's misunderstanding of the dimensions of SFL theory as interacting components, and the orientation of Fawcett's model to text generation by computer, rather than to human language itself.

Also in a systemic functional grammar, stratal realisation and instantiation are distinct dimensions, and, in terms of the theoretical architecture, form a matrix like the following:

[2] This is misleading.  This claim has not been supported by reasoned argument, merely asserted — and done so on the basis of theoretical misunderstandings.

[3] The terms 'first' and 'then' are misleading.  In a systemic functional grammar, there is no sequencing relation between two levels of symbolic abstraction.  The identifying relation is intensive, not circumstantial (temporal).  This again derives from Fawcett's misunderstanding of the dimensions of SFL theory as interacting components, and the orientation of Fawcett's model to text generation by computer, rather than to human language itself.

[4] This is misleading.  In a systemic functional grammar, realisation statements are located at the same level of symbolic abstraction as the network of features to be realised, not at a lower level.  Problematically, in Fawcett's model, a system network is realised by realisation statements.

[5] This is misleading.  In a systemic functional grammar, realisation statements apply to potential, not instances; the process of instantiation includes the activation of realisation statements.  Problematically, in Fawcett's model, a structure is an instance of realisation statements.  Problematically, in Fawcett's model, realisation statements specify an instance, rather than a realisation.  Problematically, in Fawcett's model, the realisation relation between paradigmatic axis (system) and the syntagmatic axis (structure) is confused with the instantiation relation between potential and instance.

[6] Given the theoretical (and logical) inconsistencies outlined above, it is very misleading to claim that Figure 4 genuinely represents a systemic functional grammatical model of language.