Misrepresenting Halliday On Nominal Group Structure In A Footnote

 Fawcett (2010: 213-4, 213n-4n):

The main characteristic of an element is that it is defined functionally, rather than positionally. This should surely be a founding principle of a functional approach to syntax — and yet the tradition of using positional labels still lingers on in many functional grammars.¹⁷

¹⁷ For example, terms such as 'pre-deictic' and 'pre-numerative' (as found in Halliday 1994:195-6) are simply positional labels. Rather similarly, the terms 'premodifier' and 'postmodifier' (Halliday 1994:194-5), signal a positional meaning more strongly than they signal a functional meaning. This is because, at Halliday's primary level of delicacy in the analysis of a nominal group, the term 'modifier' means little more than 'anything other than the head'. So these terms give virtually no information about the element's function. 

However, I must admit to retaining one traditional 'positional' label in the Cardiff Grammar, i.e., 'preposition'. As said in Section 10.2.6, I would have preferred an explicitly functional label, but terms such as 'relator' are not specific enough, and we retain the traditional term "preposition" both because there is a lack of a clear alternative and because it is so strongly established. We define it as functioning to express a 'minor relationship with a thing'. Note that in the description of English one item that actually occurs 'postpositionally' is included as a 'preposition' — i.e ago, as in five years ago.

Blogger Comments:

[1] This is misleading (and hypocritical) because, contrary to the implication, Fawcett's own approach to functional syntax includes elements that are defined positionally. As can be seen from the previous post, Appendix B (p304), lists:

• 'Starter' and 'Ender' as elements of the clause,
• 'finisher' as an element of the quality group,
• 'quantity finisher' as an element of the quantity group,
• 'starter' and 'ender' as elements found in all groups, and
• 'Opening Quotation mark' and 'Closing Quotation mark' as elements of a text.

[2] This is misleading, because it is untrue. The terms 'pre-Deictic' and 'pre-Numerative' are not "simply positional labels", since they identify the functions 'Deictic' and 'Numerative' respectively. However, the function 'pre-Deictic' — unlike 'post-Deictic' — is not an element in Halliday's model, and the 'pre-Numerative' was later termed 'extended Numerative' (Halliday & Matthiessen 2004: 333).

[3] This is misleading, because it is untrue. The term 'pre-Modifier' and 'post-Modifier' do not signal positional meaning more strongly than functional meaning, since they identify the function 'Modifier' and, through prefixes, additionally provide the location of the modification relative to the Head.

[4] This misrepresents Halliday. To be clear, in SFL Theory, delicacy is a dimension of paradigmatic system, not syntagmatic structure.

[5] To be clear, here Fawcett misleads through his ignorance of the notion of univariate structure: 'an iteration of the same functional relationship' (Halliday & Matthiessen 2014: 390, 451).

[6] As can be seen from the four preceding points, this is the direct opposite of what is actually true.

[7] To be clear, 'preposition', as a class of word, is a unit of form, so renaming it with a function label would be theoretically inconsistent. In SFL Theory, a preposition functions as a minor Process/Predicator of a prepositional phrase.

[8] To be clear, in SFL Theory, the experiential function of a preposition, minor Process, relates to the function of a nominal group, Range, in a prepositional phrase, whereas Thing is an experiential function of a word in a nominal group.

[9] To be clear, the word ago is an adverb, not a preposition, and five years ago is a nominal group, not a prepositional phrase:

Element Of Structure: Cardiff Grammar Vs SFL Theory

Fawcett (2010: 213-4):

The concept of element of structure (or element for short) is the second of the three fundamental categories in syntax in the present framework (the others being class of unit and item). The term "element (of structure)" and the term "structure" are used here in broadly the same sense as in "Categories" — but with two important provisos that I shall mention below.

Elements of structure are the immediate components of classes of units. …

Appendix B summarises the main elements that occur in each of the five major units of English (and also in the genitive cluster — a class of 'cluster' that may itself contain a nominal group).

Blogger Comments:

[1] To be clear, this is significantly different from SFL Theory, where elements of structure are not the immediate components of classes of units. In SFL Theory, the immediate components (constituents) of a unit are units of the rank below. For example, the immediate constituents of a clause are groups/phrases.

Elements, on the other hand, are the functions served by units in the structure of a unit of the rank above. For example, the element 'Senser' is a function served by a nominal group in the structure of a clause.

[2] Appendix B provides the following elements of units. It can be seen that Fawcett models quotation marks (graphology) — i.e. written mode only — as elements of text structure (semantics). 

Fawcett's Claim That Clusters Do Not Function Directly As Clause Elements

Fawcett (2010: 213):

The human proper name cluster (hpnclr) always and only fills the head of a nominal group. Three of its most frequent elements are title (t), one or more forenames (fl, f2, etc), and the family name (fn). Other classes of cluster also name objects, e.g., the address, date, and clock time clusters. See Fawcett (in press) for more details of these.

Occasionally, when a cluster fills the head of a nominal group and there is no other element in it, the cluster may at first appear to function as a direct element of the clause. Examples of the two classes of cluster described here can be seen in the underlined portions of This is my father's or She admired Sir Terence Conran. But this is not evidence that they are functioning directly as elements of the clause, any more than is the occurrence of the items his in This is his and him in She admired him. These are all simply cases of what is termed "singularly branching". Compare examples such as He's not the Terence Conran who I knew twenty years ago.

Blogger Comments:

[1] To be clear, Fawcett (in press) is still unpublished, 21 years after the first edition of this book.

[2] This is misleading, because it is the opposite of what is true. Whether these wordings are termed nominal groups (SFL Theory) or clusters within nominal groups (Cardiff Grammar), they serve functions at clause rank. In SFL Theory, their experiential and interpersonal functions can be identified as follows:

The absurdity of Fawcett's claim can be demonstrated by removing the cluster (nominal group) serving as the final participant from each of these clauses.

[3] This is misleading, because the term 'singularly branching' — merely the notion of a unit (nominal group) composed of a single element (head) — says nothing about whether or not these units function "directly" as clause elements. Fawcett (pp245-6):

Componence is normally represented diagrammatically by lines going down the page, either vertically or diagonally. The lines start from the unit and branch out downwards to the elements below. Sometimes only one element of a unit is used, and in such cases of what is sometimes termed singulary branching there is simply a vertical line from the unit down to the element.

Problems With Fawcett's Argument For The Genitive Cluster

 Fawcett (2010: 212):

It is the fact that 

(1) nominal groups such as the dog undeniably occur within the unit of that we are here calling the "genitive cluster" and 

(2) the genitive element 's functions as a 'relator' to the whole nominal group (and not just dog) 

that has led to the introduction to the grammar of the present unit.

See Appendix B and, for a slightly fuller picture of this unit, Fawcett (in press).

 

Blogger Comments:

[1] To be clear, neither of these justifications for the genitive cluster withstand close scrutiny.

On the first point, the fact that nominal groups "undeniably occur" within the genitive cluster is no justification for the genitive cluster because nominal groups "undeniably occur" in a range of units, including the clause, prepositional phrase and the nominal group.

On the second point, the fact that the genitive element 's functions as a 'relator' to the whole nominal group is evidence that it functions in the nominal group rather than the cluster. (For it to function in the cluster, it must relate just to the dog).

[2] To be clear, Appendix B (pp304, 307) provides the following:

Note that the structure of the genitive cluster includes both a conjunction and an "ender" (e), as well as an element ('own') named for the word that realises it.

[3] To be clear, Fawcett (in press) is still unpublished, 21 years after the first edition of this work.

Mistaking A Structure Marker For A Functional Element

  Fawcett (2010: 212):

Note however that, despite the small size of the item expounding the genitive element, it serves a function that is equivalent to a preposition, through its expression of the relationship of 'generalised possession' (i.e., 'part-whole' relationships, etc, as well as 'ownership'). Compare the dog's back legs and the back legs of the dog.

Blogger Comments:

[1] To be clear, in SFL Theory, the preposition of in the nominal group the back legs of the dog is not a functional element but a structure marker. Halliday & Matthiessen (2014: 341, 425):

Note that the structure the cooking of the rice, where the Medium follows of, is not an exception; of is functioning here, as it typically does, not as preposition but as structure marker – cf. genitive ‘s in the rice’s cooking. …

The exception is prepositional phrases with of, which normally occur only as Postmodifier; the reason is that they are not typical prepositional phrases, because in most of its contexts of use of is functioning not as minor Process/Predicator but rather as a structure marker in the nominal group (cf. to as a structure marker in the verbal group).

[2] To be clear, the part-whole relation obtains within the nominal group, not within the cluster, thereby invalidating the notion of the cluster as the unit that construes (generalised) possession.

[3] To be clear, in SFL Theory, these are both nominal groups, one with a nominal group — the dog's — embedded as Deictic, and the other with a prepositional phrase — of the dog — embedded as Qualifier:

Problems With The Structural Elements Of A Genitive Cluster

 Fawcett (2010: 212):

The two most frequent elements of a genitive cluster are the possessor (po) and the genitive element (g). The possessor is typically filled by a nominal group (and occasionally by two or more co-ordinated nominal groups), and the possessor [genitive element] is always and only filled by the morpheme 's (or, after certain word endings, just an apostrophe, as in the Jones' dog).

Thus this cluster has the unusual characteristic that one of its two main elements is typically filled by a group and the other is always expounded by a morpheme.

 

Blogger Comments:

[1] To be clear, this instance of Fawcett's genitive cluster is structured as follows:

One problem here is that an apostrophe is graphological form, not grammatical form, and does not apply to language in spoken mode. A second problem is that 'genitive' is not a function. A third problem is that the possessor is not related to a 'possessed' within this structure, but to an element (Head/Thing) of another structure (nominal group). This is a problem because, in SFL Theory, a structure is the relation between elements. Halliday & Matthiessen (2014: 451):

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

In contrast, in SFL Theory, unknown to Fawcett, this genitive cluster is a rank-shifted nominal group serving as the Deictic of another nominal group:

It is clear that one reason why Fawcett proposed the notion of a genitive cluster is because he could not fathom how to analyse such forms using SFL Theory.

[2] To be clear, this "unusual characteristic" might be more accurately described as a theoretical inconsistency.

The Genitive Cluster

Fawcett (2010: 212, 212n):

The first of the two classes of cluster in English is the genitive cluster (genclr). This always fills either (1) the deictic determiner (dd) of a nominal group, (2) the head (h) or, less frequently, (3) a modifier (m), as in the underlined portion of a girl's bike, where the meaning is 'a bike that is suitable for a girl'), or (infrequently) (4) the type of deictic that occurs in a quality group (qld), as in my sister's most precious doll.¹⁶

¹⁶ It is not the case, of course, that all deictic determiners and heads are filled by genitive clusters; there are choices in the system network that lead to either the direct exponence of an element by an item (e.g., the) or to re-entry to the network, in order to generate a cluster (e.g., this university's). See Section 11.6.2 of Chapter 6 for 'variation in depth of exponence'.

Blogger Comments:

[1] To be clear, in SFL Theory, this genitive cluster is simply a word (girl's) serving as the Classifier of a nominal group:

[2] To be clear, in SFL Theory, this genitive cluster is an embedded nominal group (my sister's) serving as the Deictic of another nominal group:

Clearly, from the perspective of a functional grammar, the genitive cluster is a redundant class of form that only arises from Fawcett giving priority to form over function.

[3] To be clear, Fawcett does not provide the system network.

Fawcett's Reason Why Clusters Occur Within The Nominal Group

Fawcett (2010: 211-2):

Clusters are a special set of units whose function is to carry complex meanings associated with two elements of the nominal group in particular: the deictic determiner and the head. It seems that the semantics of referring to things is so complex that we regularly need to introduce units within the nominal group, and embedded groups of all classes therefore occur quite frequently within the nominal group. … But certain classes of unit that occur within the nominal group can only occur within this unit (with one rare exception). We term these units clusters.

The reason why clusters occur within the nominal group is that the types of meaning that they express are inherently 'enrichments' of certain types of meaning that are inherently associated with things. Thus they do not realise meanings that can also be elements of situations, as groups do, so that they can never function as direct elements of the clause. They are therefore effectively sub-units of the nominal group. In this way, then, they are quite different from the four classes of group, all of which can fill various elements in various units.

As a consequence of the fact that they function within the nominal group, clusters can never function as referring expressions, as clauses and groups can. In other words, they can never be the answer to a question such as Who's this?, Where do you live? etc.

 

Blogger Comments:

[1] To be clear, the nominal group is the grammatical resource for constructing taxonomies of things. Halliday & Matthiessen (1999: 265, 518):

Participants are realised by nominal groups, which allow more or less indefinite expansion (through the univariate structure of modification). This expansion is the grammar’s way of constructing taxonomies of things: grouping them into classes, assigning properties to them, quantifying them and then uniquely identifying any individual thing, or any number, set or class of things in relation to the “here–&–now” of the speech event. The expansion involves open sets of things and qualities, realised by lexical items; but it can also capture a circumstance, realised by a prepositional phrase, or an entire figure, realised by a clause, and put it to use as a quality in describing or identifying such a thing or set of things, e.g. this unique 20-piece handpainted china dinner service with optional accessories never before offered for sale at such a bargain price. …

The other resource for constructing taxonomies of things is the expansion of the nominal group, and here the picture is very different from that with verbs. Nouns are expanded lexically as well as grammatically, so that, while entities (like processes) are located deictically relative to the ‘here–&–now’, they are also (unlike processes) extensively classified and described. … Thus the grammar has the potential for construing a complex arrangement of classes and subclasses for any entity which participates in a process; … common nouns are almost indefinitely expandable, and it is this resource which organises our universe into it elaborate taxonomies of things.

[2] This claim is falsified by Fawcett's own examples, such as This is my father's* and She admired Sir Terence Conran where the genitive cluster my father's and the proper name cluster Sir Terence Conran each function as a "direct element" (Complement participant) in the clause:

[3] To be clear, this is true. Fawcett's clusters are constituents ("sub-units") of nominal groups, which makes them quite different from groups.

[4] This claim is falsified by Fawcett's own example, She admired Sir Terence Conran where the proper name cluster Sir Terence Conran does function as what Fawcett terms a 'referring expression', since it "refers" to a specific person.

To be clear, by 'referring expression', Fawcett means ideational denotation, not reference in the SFL sense of textual cohesion.  From the perspective of SFL Theory, Fawcett's denotation relation here obtains between language and perceptual categorisations of experience, whereas in SFL Theory, the denotation is the realisation relation between semantics and lexicogrammar.

∞

* Halliday & Matthiessen (2014: 295):

If the relationship is construed as the Attribute, then it takes the form of a possessive nominal group, e.g. Peter’s; the thing possessed is the Carrier and the possessor is the Attribute. These are not, in fact, syntagmatically distinct from ‘identifying’ clauses; the clause the piano is Peter’s could be either ‘attributive’. ‘the piano is a member of the class of Peter’s possessions’ or ‘identifying’, ‘the piano is identified as belonging to Peter’. (Note that the reversed form Peter’s is the piano can only be ‘identifying’.)

Fawcett's Concept Of 'Cluster'

Fawcett (2010: 211, 211n):

This unit of syntax is not found in any other grammatical framework. Yet the fact is that there are frequently occurring units in English syntax that have the characteristics of the classes of unit to be described below. Perhaps I may add, in support of the proposals put forward here, that most other grammarians simply do not say how they would handle, in a systematic manner, the syntax of the types of phenomena for which the various classes of cluster are used.¹⁵

¹⁵ In my experience, the need to recognise these units only becomes imperative (1) when one is working with a functional theory of syntax in which the criteria for recognising a unit in are made fully explicit (as they are here) and (2) when there is a serious commitment to the full analysis of large quantities of naturally occurring text. The need to attend to individual orthographic words is also increased when one tries to build a computer model that parses a text consisting of a string of words into its functional syntax.

Blogger Comments:

[1] This is misleading, because it misrepresents SFL Theory as not providing the means of analysing the wordings that Fawcett models as clusters. This claim is invalidated by providing the relevant analyses. For example, the genitive cluster girl's in the nominal group a girl's bike ('a bike that is suitable for a girl') is a word serving as Classifier, whereas the genitive cluster my sister's in the nominal group my sister's most precious doll is an embedded nominal group serving as Deictic.

[2] This is misleading, because it misrepresents personal anecdote as argument in support of a theoretical proposal, and implies that only Fawcett's approach is "fully explicit" and has a "serious commitment".

See also Why Fawcett's Unit 'Cluster' Is Unnecessary

Misunderstanding Interpersonal Metaphor As The Experientialisation Of Non-Experiential Meaning

 Fawcett (2010: 210-1):

While the nominalisation of events is both the most frequent type of incongruence and the type with the most subtle variations, there are many others, such as the expression of a quality as a thing, e.g., his great happiness.

Halliday discusses such phenomena under the general heading of "grammatical metaphor" (e.g., Halliday 1994:340f.). However, the scope of the term has become very broad, and I find it more helpful to think in terms of specific types of phenomena. Thus, nominalisation occurs within experiential meaning, so that it is a different matter from the experientialisation of non-experiential meaning — e.g., saying It's possible that he'll be there rather than He may be there. See Fawcett (in press) for a rather fuller picture of these relationships.

 

Blogger Comments:

[1] This misunderstands nominalisation. To be clear, in SFL Theory, nominalisation is the reconstrual of processes and qualities as things (Halliday & Matthiessen 1999: 314). That is, Fawcett's "other" type of incongruence is actually just another example of nominalisation.

[2] This is misleading. Halliday's grammatical metaphor is an incongruent relation between semantics and grammar, not between an extra-linguistic belief system and linguistic meaning (semantics).

[3] To be clear, the theoretical notion of grammatical metaphor is clearly defined and systematically described. For ideational metaphor, see Chapter 6 in Halliday & Matthiessen (1999: 227-296).

[4] To be clear, in SFL Theory, this is an instance of an interpersonal metaphor of modality, not "the experientialisation of non-experiential meaning". The non-metaphorical use of may is an instance of implicit subjective modality, whereas the metaphorical use of it's possible that is an instance of explicit objective modality. Halliday & Matthiessen (2014: 186):

[5] To be clear, Fawcett (in press) is still unpublished, 21 years after the first edition of this publication.

Why Fawcett Needs To Propose A 'Belief System' Above Language

Fawcett (2010: 210):

Figure 12 shows two of the main types of 'congruent' and 'incongruent' relationship, as they affect clauses and nominal groups. In the terminology of the Cardiff Grammar, the language expresses 'higher' beliefs (these being represented in logical form). Then, within language, forms realise meanings. Thus, if an event in the belief system is mapped onto a situation in the semantics (which will in turn be mapped onto a clause in the syntax at the level of form), then the relationship between the 'event' and the 'clause' is said to be congruent — a term introduced by Halliday (1970:149). However, an event can be incongruently realised as a nominal group, i.e., when 'nominalisation' takes place, as in his entering the room at that moment. And an object can be incongruently realised as a clause, as in what you gave me for my birthday.

Blogger Comments:

[1] To be clear, in Fawcett's model of grammatical metaphor, clauses and nominal groups are not affected by the distinction between congruent and incongruent relationships, because in either case, a clause realises a situation and a nominal group realises a thing.

[2] As previously explained, Fawcett's 'belief system' outside language is inconsistent with the epistemological assumptions of SFL Theory (Halliday & Matthiessen 1999: 3). In SFL Theory, beliefs are ideas, which are mental projections of linguistic meaning.

[3] As explained in the previous post, Fawcett nowhere discusses his model of logical form, and his location of it outside language is inconsistent with its usage in linguistics, philosophy and mathematics.

[4] To be clear, in SFL Theory, grammatical forms are interpreted in terms of the functions they serve, and their function is to realise meaning. In the absence of grammatical metaphor, the function of grammatical forms and semantics agree (are congruent). It is because Fawcett's model exports grammatical functions to his level of meaning that he is required to invent a yet higher level, outside language, in an attempt to account for grammatical metaphor.

[5] To be clear, the in/congruent relation obtains only between adjacent levels of symbolic abstraction (strata) and only between strata that are related non-arbitrarily ('naturally'), that is: between semantics and lexicogrammar.

[6] To be clear, the wording what you gave me for my birthday is not a clause, but a nominal group with an embedded clause as Qualifier:

Fawcett's Grammatical Metaphor As An Incongruent Relation Between Ideational Meaning And Non-Language

Fawcett (2010: 209-10):

I have referred in previous sections to correspondences between 'syntactic', 'semantic' and 'conceptual' units. Let us now consider the relationship between the syntactic units and their 'higher' equivalents. We shall focus upon the two main classes of syntactic unit: the clause and the nominal group.

How do these come to be generated? The answer is that they originate (typically) at the level of logical form, in the planner (which draws in turn on the belief system). They are typically represented as events and the objects that occur in events. These are then processed in the semantics, where an event is typically expressed as a situation and an object as a thing — and these in turn are then realised as a clause and a nominal group. However, there is not necessarily a one-to-one relationship between logical form, which is extralinguistic) and the intra-linguistic level of semantics. Figure 12 gives an overview of the two possibilities for incongruent correspondences between these. 

Blogger Comments:

[1] As this makes plain, Fawcett's theory of language is actually a model of text generation by computer.

[2] To be clear, Fawcett nowhere elaborates on the organisation of logical form, but it can be noted that his location of it outside language is inconsistent with its conception in linguistics, philosophy and mathematics. In linguistics, it refers to a syntactic structure that is interpreted semantically:

In generative grammar and related approaches, the logical Form (LF) of a linguistic expression is the variant of its syntactic structure which undergoes semantic interpretation.

whereas in philosophy and mathematics, it is the semantic interpretation of a syntactic expression:

In philosophy and mathematics, logical form of a syntactic expression is a precisely-specified semantic version of that expression in a formal system.

[3] To be clear, Fawcett's events and objects are ideational meanings relocated outside language. Interpersonal and textual meanings are absent, and so are not accounted for here at the level of belief system (or semantics).

[4] To be clear, Figure 12 misrepresents relations between levels, expression and realisation, as levels.

[5] To be clear, here Fawcett is interpreting grammatical metaphor — an incongruent relation between semantics and grammar — as an incongruent relation between ideational semantics and a higher level of meaning outside language. The possibility of interpersonal metaphor is not entertained.

Fawcett's Argument Against The Preposition And Conjunction Groups

Fawcett (2010: 209):

In the Cardiff Grammar we take the view that it is not worth setting up two new units for such relatively rare cases, and we therefore 'borrow' the quantity group for use in modelling the internal structures of prepositions, Linkers and Binders — but only when there is an internal structure. Notice that, if one gives weight to the criterion of the internal structure of such units (as we do here), it would be odd to treat immediately after in immediately after their visit to us as a 'prepositional group', and immediately after in immediately after they had visited us as a 'conjunction group'. By using the quantity group in both cases we avoid setting up two completely new and little-used units for each of the preposition and the Binder.

For a slightly fuller picture of the quantity group, see the examples in Appendix B, and for a full description of this unit in English see Fawcett (in press).

Blogger Comments:

[1] To be clear, preposition groups and conjunction groups are not "relatively rare cases". In SFL Theory, every minor Process/Predicator of a prepositional phrase is realised by a preposition group, and every conjunctive Adjunct is realised by a conjunction group, in line with the principle of exhaustiveness. Halliday & Matthiessen (2014: 84):

The general principle of exhaustiveness means that everything in the wording has some function at every rank (cf. Halliday, 1961, 1966c).

[2] To be clear, as previously explained, this misunderstands constituency. The internal structure of prepositions is realised by morphemes.

[3] This is true. However, giving priority to structure is giving priority to the view 'from below', which is the opposite of the SFL approach of construing units in terms of their functions ('from above'). From a functional perspective, it is worse than "odd" to treat the preposition group immediately after and the conjunction group immediately after the same, because each serves a different function (minor Process/Predicator vs conjunctive Adjunctive).

[4] Appendix B (p307) provides this "slightly fuller" picture of the quantity group:

It can be seen that Fawcett's quantity group, like his quality group, includes both clause-rank interpersonal Adjuncts (indeed) and punctuation marks ('e') from the stratum of graphology. That is, it is theoretically inconsistent in terms of both rank and stratification.

[5] To be clear, Fawcett (in press) is still unpublished, 21 years after the first edition of this book.

Misrepresenting Halliday On The 'Conjunction Group' And The 'Preposition Group'

Fawcett (2010: 208):

Finally, we should note that here we use the quantity group to analyse structures for which Halliday introduces the two units of the 'conjunction group' and the 'preposition group' (1994:211-2). The purpose of the first is to handle the relatively rare cases of internal structure within Binders ("subordinating conjunctions") such as almost as soon as (where as soon as is the "conjunction"), and the second is for the equally rare case of structure within a preposition. I know of no cases where a Linker ('co-ordinating conjunction') has internal structure (items such as and so, and and then being treated as single items that include a space, like the preposition in spite of). Logically, in view of the 'rank scale' principle, an IFG-style analysis should treat every case of a one-word conjunction or preposition as an element of one or other of these two classes of group — but there is no sign that Halliday would in fact do so.

Blogger Comments:

[1] See the earlier post Fawcett's Quantity Groups Reanalysed Using SFL Theory.

[2] To be clear, all classes of group are proposed in accordance with the theoretical principle of exhaustiveness. Halliday & Matthiessen (2014: 84):

The general principle of exhaustiveness means that everything in the wording has some function at every rank (cf. Halliday, 1961, 1966c). But not everything has a function in every dimension of structure; for example, some parts of the clause (e.g. interpersonal Adjuncts such as perhaps and textual Adjuncts such as however) play no role in the clause as representation.

[3] This is a serious misunderstanding of the rank scale. To be clear, the purpose of preposition group is not to handle structure within a preposition. (Structure within a preposition is realised by morphemes.) A preposition group is an expansion of a preposition, with a Modifier^Head structure, with a preposition serving as Head, as in right beneath. Halliday & Matthiessen (2014: 423):

Conjunctions also form word groups by modification, for example even if, just as, not until, if only. These can be represented in the same way, as β ^ α structures (or α ^ β in the case of if only). …

Prepositions … form groups by modification, in the same way as conjunctions; e.g. right behind, not without, way off as in right behind the door, not without some misgivings, all along the beach, way off the mark.

[4] This is misleading, because wordings such as and so and and then are indeed conjunction groups "with internal structure". Halliday & Matthiessen (2014: 478):

The combination of enhancement with parataxis yields what is also a kind of coordination but with a circumstantial feature incorporated into it; the most frequently occurring subtypes are those of time and cause. The circumstantial feature is typically expressed (a) by the conjunctions then, so, for, but, yet, still; (b) by a conjunction group with and: and then, and there, and thus, and so, and yet; …

[5] Clearly, the wording 'a space' is another instance that demonstrates that Fawcett's view on language is 'from below', in this instance from graphology. As previously explained, this is the direct opposite of a functional approach, where the view is 'from above': the functions being realised.

[6] This is almost true. In SFL Theory, every one-word conjunction or preposition realises an element of its respective group; see [7].

[7] This is misleading, because it misrepresents Halliday. In SFL Theory, a one-word conjunction or preposition serves as the unmodified Head of a conjunction or preposition group:

Misrepresenting Halliday On The Preposition Group

Fawcett (2010: 208-9):

Note that Halliday's "preposition group" is a unit that fills a preposition, and not, like our 'prepositional group', a unit that has a preposition as its pivotal element. Thus the meaning of its name runs counter to the pattern set by his use of the terms "nominal group", "verbal group" and "adverbial group". One of his few examples of a 'preposition group' is right behind (the door).

Blogger Comments:

[1] To be clear, this is an elementary error, and the reverse of what is true. A preposition "fills" (serves as the Head of) a preposition group.

[2] This is misleading. Fawcett's prepositional group is the counterpart of the prepositional phrase —not the preposition group — in SFL Theory. Moreover, in modelling the prepositional phrase as a group, Fawcett misconstrues it as the (univariate) expansion of a word (preposition) instead of the contraction of a (multivariate) clause, with preposition serving as minor Process/Predicator.

[3] This is misleading, because it is the opposite of what is true; see [1]. The Heads of nominal, verbal and preposition groups are nominals, verbs and prepositions, respectively. Halliday & Matthiessen (2014: 75):

[4] This is not misleading, because it is true.

Fawcett's Quantity Groups Reanalysed Using SFL Theory

 Fawcett (2010: 208):

Quantity groups in fact occur more frequently within nominal groups, as in She hasn't smoked very many cigarettes today, and it is not at all clear how this and the other types of quantity group illustrated in Appendix B would be handled in an IFG-style analysis.

Blogger Comments:

To be clear, from the perspective of SFL Theory, Fawcett's quantity group very many in the above example is a word complex serving as the Numerative of a nominal group:

Turning to Appendix B (p307), Fawcett's quantity group a great deal less than expected in a great deal less food than expected covers both the Numerative and Qualifier of a nominal group, with the Sub-Head of the Numerative premodified by an embedded nominal group:

On the other hand, Fawcett's quantity group immediately after in the Appendix B example immediately after the game is a preposition group serving as the Process/Predicator of a prepositional phrase:

whereas Fawcett's quantity group immediately after in the Appendix B example immediately after the he'd gone is a conjunction group serving as the conjunctive Adjunct of a clause:

That is, as the analyses demonstrate, in terms of SFL Theory, Fawcett's quantity group is distributed across both group classes (nominal, preposition, conjunction) and ranks (clause, phrase/group, word). On this basis, the quantity group might be said to represent a ragbag of miscellaneous functions. But importantly, proposing a quantity group results in failing to account for all the different functions that these similar forms serve.

Fawcett's Argument Against The Adverbial Group

Fawcett (2010: 208, 208n):

There is no equivalent in IFG — nor in any other grammar that I know of — for the concept of the 'quantity group'. Halliday would probably handle the type of meaning that occurs as a Degree Adjunct (e.g., the underlined portion of He loves her very much indeed) as an 'adverbial group', because of its potential for functioning as an Adjunct.¹⁴

¹⁴ In such cases, Halliday's stated criterion of the unit's ability to fill the relevant element in the unit above on the 'rank scale' would lead him to label this unit as an "adverbial group", but in cases such as the present one this criterion might be thought to be supported by the fact that in traditional grammar items such as much in such contexts are called "adverbs". However, the category of "adverb" in English is a 'ragbag', with items such as very also being traditionally classed as "adverbs", so these grounds are flimsy. It seems to respect the data more properly to introduce a unit that gives an appropriate weight to the concept of 'quantity' — as we do here.

Blogger Comments:

[1] To be clear, Halliday would analyse the underlined portion of He loves her very much indeed as two consecutive Adjuncts, the first circumstantial (Manner: degree), and the second modal (counter-expectancy: exceeding):

As Halliday & Matthiessen (2014: 158) point out:

modal and conjunctive Adjuncts occur finally only as Afterthought and can never carry the only tonic prominence in the clause.

In this instance, indeed is likely to carry the second tonic prominence of a compound tone group: 

// 13 ‸ He /loves her / very much in/deed //

[2] This is not misleading, because it is true.

[3] To be clear, the adverbial group is based on the word classes recognised by SFL Theory; Halliday & Matthiessen (2014: 75):

[4] To be clear, Fawcett's argument is:

1. the category of adverb in English is a ragbag,
2. therefore the grounds for an adverbial group are flimsy,
3. therefore the adverbial group does not properly respect the data,
4. therefore introducing a quantity group does properly respect the data,
5. because a quantity group gives an 'appropriate' weight to the concept of 'quantity'.

Clearly, this argument is a ragbag of non-sequiturs and bare assertions:

• 2 does not follow from 1, because the adverbial group is an expansion of the word class adverb, however varied the members of the class;
• 3 does not follow from 1 & 2, because it is a subjective judgement ('properly'), unsupported by evidence; 
• 4 does not follow from 1, 2 & 3, because it is a subjective judgement ('properly'), unsupported by evidence, that also ignores the possibility of alternative valid approaches;
• 5 does not follow from 1, 2, 3 & 4, because it is a subjective judgement ('appropriate'), unsupported by evidence, that also ignores the possibility of alternative valid approaches.

It might be added that adverbial groups are not restricted to the concept of 'quantity', but realise a wide range of Adjuncts: circumstantial, modal and conjunctive.

Fawcett's 'Quantity Group'

Fawcett (2010: 207-8, 208n):

The fourth and last class of group is the quantity group. This has as its pivotal element an expression of 'quantity'. It takes its name from the semantic unit of 'quantity' that it realises, and the "quantity" that it refers to may be a quantity of a 'thing', a 'situation', a 'quality' or, perhaps surprisingly, a 'quantity' (as in the underlined portion of very many more) — and so to equivalent conceptual units.

Blogger Comments:

[1] To be clear, as was also the case for the 'quality group', Fawcett provides no semantic network featuring his semantic unit 'quantity' in support of his claims.

[2] To be clear, Figure 12 (p210) presents Fawcett's 'thing' as the semantic counterpart of his nominal group, so the claim here is that meaning of the entire nominal group that is quantified, rather than an element within it. Similarly, Fawcett's 'situation' is the semantic counterpart of his clause, so the claim here is that meaning of the entire clause that is quantified, rather than an element within it. However, this latter case presents a further difficulty, since, for this to be coherent, there would have to be a (higher-ranked) grammatical unit of which a clause is a constituent, and which serves a function in that higher unit. Fawcett provides no clause functions within a higher unit.

[3] To be clear, in SFL Theory, the instance very many more is the Numerative element of a nominal group, with more as its Head, and very many as preModifier. That is, what Fawcett models as a 'quantity group', SFL models as the premodification of the Numerative of a nominal group:

[4] To be clear, Figure 12 (p210) presents Fawcett's 'object' as the 'belief system' counterpart of his semantic 'thing', and 'event' as the 'belief system' counterpart of his semantic 'situation' — with 'belief system' not a level of language, but a higher level system that is expressed as language. As previously mentioned, this is inconsistent with the epistemological assumptions on which SFL Theory is founded. Halliday & Matthiessen (1999: 2-3):

In modelling the meaning base we are building it 'upwards' from the grammar, instead of working 'downwards' from some interpretation of experience couched in conceptual terms, and seen as independent of language. We contend that the conception of 'knowledge' as something that exists independently of language, and may then be coded or made manifest in language, is illusory.

Misrepresenting Halliday (1994) On The Nominal Group

Fawcett (2010: 207):

As we saw in Section 10.2.5, Halliday treats the type of unit that has an adjective as its 'pivotal element' as a 'nominal group' — despite the fact that its internal structure is clearly similar to that of his 'adverbial group'. Thus Halliday describes the group very lucky as a nominal group that has the adjective lucky as its "head" — while also stating that such units are "sometimes referred to distinctively as adjectival groups" (1994:194). However, a few pages later in IFG he also recognises that examples such as more easily (and so presumably very cleverly too) are what he terms "adverbial groups", saying that they have a "modifier-head" structure. He recognises that this class of unit may have the "postmodification" of "comparison" (1994:210) — and yet he fails to mention that a group with an adjective at its "head" may also have this type of "postmodification" — and indeed that both types of "quality group" may also have other elements (a finisher and one or two scopes) also, as Appendix B shows. This area of IFG appears, frankly, to be internally inconsistent and to require considerably more work. Perhaps the next edition of IFG will draw on the major contribution within SFL to this area by Tucker (1997 and 1998).

See Appendix B, Fawcett (in press) and especially Tucker (1998) for a fuller picture of the quality group.

Blogger Comments:

[1] This is not misleading, because it is true.

[2] This is very misleading indeed, and clearly, deliberately so, because Fawcett has previously acknowledged that Halliday does not classify units on the basis of their internal structure, but on the function they serve in the structure of the unit above on the rank scale.

[3] To be clear, this is merely an acknowledgement of the alternative view from which Fawcett's model derives.

[4] This is misleading, because it is the opposite of what is true. Halliday (1994) explains nominal group postmodification (pp192-3), and then Epithet as Head of the nominal group on the following page (p194). And for those unable to join the dots, he provides a specific example (p210):

[5] This is misleading. To be clear, here Fawcett is claiming that Halliday is inconsistent in his own modelling because he fails to mention Fawcett's model, the quality group, which, as previously demonstrated, is inconsistent with Halliday's model.

[6] This is misleading, because it is the opposite of what is true. As demonstrated above, Fawcett's conclusion is made on the basis of the three misleading claims identified above in [2], [4] and [5].

[7] To be clear, the latest edition of IFG, Halliday & Matthiessen (2014: 376) acknowledges Tucker's work, with clues on how to re-interpret it in a manner that is consistent with SFL Theory:

Tucker (1998) provides a detailed, lexicogrammatical and semantic description of adjectives in English, with system networks showing the potential for construing qualities.

[8] To be clear, Fawcett (in press) is still unpublished, 21 years after the first edition of this work, and the fuller picture of the quality group in Appendix B (p307) is given as: 

It can be seen that Fawcett's quality group includes not only conjunctions ('&') and clause-rank interpersonal Adjuncts (indeed), but punctuation marks ('e') from the stratum of graphology. Moreover, in presenting the Numerative fiftieth as the apex ('a') of a quality group, it misconstrues it as representing a quality at the level of meaning.

Problems With The Term 'Quality Group'

Fawcett (2010: 206-7, 207n):

The term 'quality' is borrowed directly from the term used in the system network for the meanings that are realised in this unit. But, in a model of language in which the syntax is seen as the reflection of meaning, it is natural to introduce an explicitly functional label in this way. (The reason why this approach is not followed for the 'nominal' and 'prepositional' groups is that the labels that reflect the 'word class' tradition are so well established.) There is, of course, a long tradition in linguistics of using functional labels for categories that are in fact the realisations of meanings (as the term 'nominal" originally was), and there is certainly no reason to avoid using such labels — especially in an explicitly functional grammar such as the present one.¹² This discussion illustrates nicely the important fact — which is often overlooked — that it is not only the elements that are functional, but also the units.

¹² It has been suggested that we might call the group the "adjective-adverb group" to preserve the parallelism with the "nominal group", but it seems more natural, in the framework of an explicitly functional grammar, to use a label that directly expresses the type of meaning that this unit expresses.

Blogger Comments:

[1] To be clear, Fawcett does not provide the system network featuring the term 'quality'. Moreover, in terms of ideational meaning, his Figure 12 (p210) only features 'situation', realised by clause, and 'thing', realised by nominal group. Interpersonal and textual semantics are not featured.

[2] To be clear, in Systemic Functional Theory, syntax is not seen as a "reflection" of meaning, since syntax and meaning are different levels of symbolic abstraction, whereas a phenomenon and its reflection are not. This suggests that Fawcett does not understand the principle of realisation.

[3] To be clear, introducing an explicitly functional label for a unit of form creates a theoretical inconsistency. In SFL Theory, forms are modelled as a constituency hierarchy, a rank scale, with constituents assigned functions in the unit of which they are constituents, as exemplified by an adverbial group serving the functions 'Manner' circumstance or 'comment Adjunct' in clause structure.

[4] To be clear, in SFL Theory, an element is the function served by a unit; e.g. the element Senser is the function served by the unit nominal group.

[5] To be clear, a grammatical unit and its meaning are different levels of symbolic abstraction, and to label a grammatical form in terms of its meaning is to confuse levels of symbolic abstraction in the architecture of the theory. Moreover, units realise not just ideational meanings, but those of the interpersonal and textual metafunctions. But most importantly, this approach "leapfrogs" the more immediate grammatical functions realised by a 'quality unit', such as Attribute (participant), Manner (circumstance), Adjunct, and Theme or Rheme.

Fawcett's Argument Against 'Adjectival' And 'Adverbial' Groups

Fawcett (2010: 206n):

¹¹ In some earlier systemic grammars the concept of 'adjectival' and 'adverbial' groups was explored. But the fact is that the system network of meanings and the elements of structure in syntax that it is necessary to set up to handle the type of 'quality group' with an adjective at its apex includes essentially the same types of meaning and structure that are needed for the type with a manner adverb at its apex. It is therefore appropriate to capture this significant overlap by modelling the two phenomena in the same system network and the same syntactic unit. Compare He is less slow than he used to be and He climbed the stairs less slowly than he used to climb them.

Blogger Comments:

[1] This is misleading, since the adverbial group is a class of group in SFL Theory. 

[2] To be clear, Fawcett does not provide any system network of meanings to validate his claims.

[3] To be clear, Fawcett argues for the theoretical validity of his quality group on the basis of both meaning and structure. And from the perspective of SFL Theory, in terms of meaning, both variants — adjectival and adverbial — do realise the same meaning, a quality, and in terms of structure, both variants do share the same logical structure:

However, what Fawcett's quality group fails to account for is the fact that the semantic element, quality, is realised as a different grammatical element in each case: Attribute (participant) vs Manner (circumstance). In SFL Theory, this distinction is "predicted" by the fact that the 'quality-as-participant' is realised by a nominal group, whereas the 'quality-as-circumstance' is realised by an adverbial group.

Problems With Fawcett's Quality Group

 Fawcett (2010: 206):

The quality group has as its pivotal element (the apex) an adjective or an adverb (typically a manner adverb). The quality group corresponds to the semantic unit of 'quality'. However we should note that the 'quality' may be a quality of either a 'thing' (e.g., clever) or a 'situation' (e.g., cleverly in He cleverly opened it with a paper clip). Thus quality groups that refer to the quality of an object have an adjective as their apex, and quality groups that refer to an event have, typically, a manner adverb at their apex — but sometimes other classes of adverb, such as the 'usuality' adverb often.¹¹

Blogger Comments:

[1] To be clear, here Fawcett applies a functional label to a formal unit, a group, thereby creating a theoretical inconsistency with regard to other groups in his own model, not least because he labels this group 'from above' (function) despite classifying it 'from below' (structure). In SFL Theory, formal units are assigned functions in the structure of the unit above them in the rank scale. For example, 

• the word clever is assigned the function Epithet in the nominal group a clever device,
• the nominal group (very) clever is assigned the function Attribute in the clause he's not (very) clever, and 
• the adverbial group (very) cleverly is assigned the function Manner: quality in the clause He (very) cleverly opened it with a paper clip.

[2] To be clear, in SFL Theory, qualities can be realised as circumstances (Manner), participants (Attribute) or a feature of a Process, as in he sped around the corner, where sped means 'moved speedily'. Halliday & Matthiessen (1999: 208):

[3] To be clear, in Fawcett's model, 'situation' is the semantic unit — regardless of metafunction — realised in syntax by a clause. Its nearest equivalent in SFL Theory is the ideational unit 'figure' (Halliday & Matthiessen 1999).

[4] To be clear, in SFL Theory, this variant of Fawcett's quality group is a nominal group with Epithet as Head, as in (she's) quite clever:

[5] To be clear, in SFL Theory, this variant of Fawcett's quality group is an adverbial group, as in (he does not plan) very cleverly:

[6] To be clear, in Fawcett's model (Figure 12), 'event' is not language, but a 'belief system' unit, that is typically realised in language as a 'situation' in semantics, and as a clause in syntax. As previously explained, this is inconsistent with the the epistemological assumptions on which SFL Theory is founded.

[7] To be clear, the 'usuality' adverb often does not realise a quality — not least because 'usuality' is interpersonal in function, whereas 'quality' is ideational. This highlights the general inconsistency here: many groups with an adverb as its apex do not construe qualities; for example (from Halliday & Matthiessen 2014: 313, 191):

• adverbs of place: abroad, overseas, home, upstairs, downstairs, inside, outside; out, up, down, behind; left, right, straight; there, here
• adverbs of time: today, yesterday, tomorrow; now, then
• adverbs serving as comment Adjuncts: naturally, inevitably, obviously, clearly, plainly, doubtless, indubitably, unsurprisingly, predictably, surprisingly, unexpectedly, evidently, allegedly, supposedly, arguably, presumably, luckily, fortunately, hopefully, sadly, unfortunately, amusingly, funnily, importantly, significantly, wisely, cleverly, foolishly, stupidly, rightly, correctly, justifiably, wrongly, unjustifiably, characteristically, typically, truly, honestly, seriously, admittedly, certainly, actually, really, generally, broadly, roughly, ordinarily, frankly, candidly, honestly, confidentially, personally, strictly, tentatively