Friday, 4 December 2020
⁶ Let me give an example of the use of these criteria when introducing a new class of unit. The quantity group was only introduced to the present description of English syntax when detailed studies of this area of the grammar showed that it could not reasonably be handled in terms of an existing unit, such as the quality group. In other words, it became clear that there are units with 'quantities' as their pivotal element (e.g., over sixty) as well as groups with 'qualities' as their pivotal element (e.g., quite clever) — and that the elements of structure in the two classes of unit serve different functions.
Blogger Comments: To be clear, for Fawcett (p196):
… the different classes of syntactic unit that are recognised in the description of a language are to be identified by their internal structure, i.e., by the elements of structure of those units.
Tuesday, 1 December 2020
Since our topic in this book is syntax, I shall now re-express the semantic view taken in Section 10.2.1 in terms of concepts that are at the level of form. Then, since 'class of unit' is a key category in the present theory, I shall examine the extent to which the criterion for recognising 'classes' of unit proposed in "Categories" has been supported by other scholars.
In terms of criteria at the level of form, then, the different classes of syntactic unit that are recognised in the description of a language are to be identified by their internal structure, i.e., by the elements of structure of those units. This concept will be exemplified in the classes of unit in English described in Sections 10.2.3 to 10.2.12. For a fuller discussion of the concept of 'element of structure', see Section 10.5.⁶
 To be clear, "the semantic view taken in Section 10.2.1" is the mutual determination of class of semantic unit and class of syntactic unit (p193):
… the class of a syntactic unit and of the semantic unit that it realises are mutually determined … each such class exists to express the specific array of meanings that are associated with each one of the major classes of entity in the semantics. … Each of these semantic units corresponds directly to one of the five major classes of syntactic unit that are recognised in the present syntax of English.
 To be clear, the validity of a theoretical proposition rests on reasoned argument, not on the extent to which it is supported by other scholars. This is another instance of the logical fallacy known as argumentum ad populum.
 To be clear, "the criterion for recognising classes in 'Categories' " is with reference to the structure of the unit above in the compositional rank scale. Halliday (2002 : 49, 50):
The class is that grouping of members of a given unit which is defined by operation in the structure of the unit next above. … Class, like structure, is linked to unit: a class is always a class of (members of) a given unit: and the class–structure relation is constant – a class is always defined with reference to the structure of the unit next above, and structure with reference to classes of the unit next below. A class is not a grouping of members of a given unit which are alike in their own structure. In other words, by reference to the rank scale, classes are derived “from above” (or “downwards”) and not “from below” (or “upwards”).
 To be clear, as can be seen from the quote immediately above, Fawcett's recognition criteria for classes of unit directly contradict the criteria used in the theory he is developing: Halliday's superseded Scale & Category Grammar. Whether or not this raises theoretical problems will be examined in future posts.
Sunday, 29 November 2020
The nearest that the present grammar comes to a generalised concept of a 'modifier + head' relationship is its recognition of the fact that other elements of a group typically depend on the presence of the 'pivotal element'. Thus when the grammar generates a "common nouns" as the head of a nominal group, other elements realising other types of meaning typically get brought into play as well. Thus it is preferable to characterise the nominal group as a unit for expressing the wide range of types of meaning associated with a 'thing', rather than in terms of an over-simple series of 'modifier + head' relationships.
So far I have been explaining the concept of 'class of unit' in terms of the concepts of the Cardiff Grammar. In very broad terms, the concept is the same in IFG — except that the account of 'class of unit' given above is more directly connected to the level of meaning here than it is in IFG. However, the criteria used in the present theory for setting up different classes of unit are completely different from Halliday's, and the concept of 'class of unit' is therefore also significantly different. Indeed, the result is the recognition of a different set of classes of unit for English, as we shall see in the rest of Section 10.2.
 This is misleading. To be clear, Halliday's SFL Theory models the nominal group as both an experiential structure ("types of meaning associated with a 'thing' ") and a logical structure ("an over-simple series of 'modifier + head' relationships"), whereas Fawcett's Cardiff Grammar only models the nominal group as an experiential structure. SFL Theory therefore has the added explanatory advantage of accounting for nominal group structures where the Thing and Head are not conflated, as in measure expressions (Halliday & Matthiessen 2014: 391-2):
 This is seriously misleading, because it is the direct opposite of what is true. In SFL Theory, the class of a unit, such as a group, is distinguished by taking the view 'from above', the function it potentially serves at the higher rank. Halliday & Matthiessen (2014: 363):
A class is not a grouping of members of a given unit which are alike in their own structure. In other words, by reference to the rank scale, classes are derived “from above” (or “downwards”) and not “from below” (or “upwards”).
Friday, 27 November 2020
Once we recognise that each element of a syntactic unit makes a unique contribution to realising the meaning of that unit, we can dispense with the traditional, over-narrow characterisation of the internal structure of groups as a series of 'modification' relationships. In its extreme form, this model presents groups as simply the 'hypotactic expansion' of the word class that functions as the 'head', in a series of 'modifier-head' relationships. This concept is central to 'sister dependency' grammars such as that of Hudson (1976), and it is reflected in Halliday's suggestion that there should be a 'logical' as well as an 'experiential' structure in the nominal group and the supposed 'verbal group'. (For a critique of the concept of a 'logical structure' in the 'verbal group' see Fawcett 2000.) However, to see groups as little more than the 'hypotactic expansion' of a word is to lose the important insights that come from approaching each group on its own terms, as a unit whose elements function to express meanings. Yet this is what Halliday is at times inclined to do. (See IFG pp. 180-1 and 196, and for a specific case where he treats the group very small as a 'word complex', see IFG p. 184.)
Tuesday, 24 November 2020
Fawcett (2010: 194, 194n):
I have just stated that each semantic unit has an associated "array of meanings". Such arrays of meaning are in effect the 'elements' of the higher unit that is constructed in the planner and that corresponds to each syntactic unit. They are realised — though not in a one-to-one relationship — in the different elements of structure of which the syntactic unit is composed.⁵
⁵ One example of the lack of a one-to-one relationship between the higher units in the planner and syntactic units of English is that a single element of the nominal group (the head) typically realises the two semantic variables of (1) the 'cultural classification' of things (which are realised by the 'common nouns' of a language) and (2) 'number' (i.e., 'mass' or 'count' and, if 'count', 'singular ' or 'plural', to slightly oversimplify).
 As the term 'planner' makes clear, Fawcett's Cardiff Grammar is a model of text generation by computer, not a theory of language as embodied in human interactants.
 To be clear, like common noun, the categories of mass noun, count noun, singular noun and plural noun are grammatical, not semantic. Halliday & Matthiessen (2014: 75):
Consider for example the class of ‘noun’ in English. A general definition would involve both grammatical and semantic considerations, with some of the grammatical features having an overt manifestation and others not:
(semantic:) expresses a person, other being, inanimate object or abstraction, bounded or unbounded, etc.
(grammatical:) is either count or mass; if count, may be either singular or plural, plural usually inflected with -s; can be made possessive, adding -’s/-s’; can take the in front; can be Subject in a clause, etc.
See also Halliday & Matthiessen (2014: 384-6).
Sunday, 22 November 2020
Thus the Cardiff Grammar's approach to the concept of 'class of unit' in syntax is to recognise that each such class exists to express the specific array of meanings that are associated with each one of the major classes of entity in the semantics. For English these are what we might term situations, things, qualities (of both situations and things) and quantities (typically of things, but also of situations and qualities). English also has a semantic unit that can be termed a minor relationship with thing. Each of these semantic units corresponds directly to one of the five major classes of syntactic unit that are recognised in the present syntax of English. These are:
nominal group,prepositional group,quality group, andquantity group.
To be clear, the classes of semantic unit in the Cardiff Grammar are all experiential meanings, despite the fact that the Cardiff Grammar locates the textual system of THEME and the interpersonal system of MOOD, along with the experiential system of TRANSITIVITY, at the level of semantics. This discrepancy is not addressed anywhere in this publication.
Friday, 20 November 2020
To summarise: the two mutually defining concepts of 'unit' and 'rank' have no explicit role in the theory of SF syntax that is set out here. Moreover, although they are heralded as central concepts in most works that describe the Sydney Grammar, neither Halliday nor Matthiessen make much use of them in practice — either in their theoretical statements or in their descriptions of English (except in their accounts of 'rank shift', where Halliday's preference is now for the term "embedding"). The twin concepts of 'unit' and 'rank' play no part in the operation of the grammar, and the centrality of the concept of 'unit' in "Categories" is replaced by the centrality of the concept of class of unit (as described in Section 10.2 below). The concept of the 'rank scale' is replaced by a statement about the probabilities that a given class of unit fills a given element of the same or another class of unit, as discussed in Section 11.2 of Chapter 11 and as exemplified in Appendix B.
 This is misleading, because, as previous posts have demonstrated, Fawcett explicitly ranks his units on a scale from higher to lower.
 This is very misleading indeed. In SFL Theory, the rank scale provides the formal units that are modelled in terms of their functions in realising meaning. Paradigmatically, each unit on the grammatical rank scale is the entry condition for the systems that are realised as structures. Syntagmatically, each function structure at a higher rank is realised by a syntagm of forms at the lower rank. For example, a clause structure such as Sayer ^ Process ^ Verbiage ^ Location is realised by the syntagm nominal group ^ verbal group ^ nominal group ^ adverbial group.
 To be clear, if there is a class of unit, there is a unit of which there are classes.
Tuesday, 17 November 2020
Finally, I should comment briefly on the fact that the concepts of 'word' and 'morpheme' are not even considered as candidates for any possible 'rank scale'. The explanation is connected with the meanings of the terms item and exponence, as defined in this theory, and these will be explained in the sections dealing with those concepts (Section 10.5 of this chapter and Section 11.5 of Chapter 11). The present theory regards the relationship of words and morphemes to the 'higher' units on the supposed 'rank scale' as different from the relationship between, let us say, groups and clauses. Indeed, it is the relationship between clauses and the various classes of groups that lies at the heart of understanding English syntax, as Appendix B demonstrates.
 To be clear, Fawcett's 'item' (p226) includes both 'word' and 'morpheme', and it expounds the lowest element of syntactic structure:
The third of the three major categories in the present theory of syntax (with 'unit' and 'element') is the item. This term includes both 'word' (in its traditional sense) and 'morpheme'. … In the present theory of syntax, the lowest syntactic category on each branch of the tree in a tree diagram representation of a sentence is an element (e.g., the head of a nominal group). And each such lowest element is expounded by an item — or as we shall see shortly, by items (in the plural).
That is, Fawcett's model confuses function structure (head of a nominal group) with formal units (items as words and morphemes).
 To be clear, as the quote above demonstrates, the exponence relation does not obtain between items (words and morphemes) and higher units (e.g. nominal groups), but between items and an element of group structure (head). That is, Fawcett's model confuses formal constituency (clause-group-word-morpheme) with function structure (head as element of the nominal group).
 To be clear, if there are higher and lower (formal) units and (functional) elements, then both units and elements are ranked on (formal and functional) scales.
Sunday, 15 November 2020
Fawcett (2010: 192n):
³ However, there is still the major difference between the two models in what types of relationship they permit between two (or more) clauses that make up a sentence. While the Sydney Grammar allows for "paratactic" and "hypotactic" relationships between clauses, the Cardiff Grammar recognises only the first of these. The prototypical type of "paratactic" relationship is that of 'co-ordination', as in My brother has arrived but his girlfriend will be a few minutes late, and all grammars recognise this relationship. In contrast, Halliday's "hypotactic" relationship between two units is not recognised in most other grammars. It is said to be one of 'dependency' without 'embedding', an example being He told us that he would be there. In Halliday's model the unit that he would be there is said to be 'projected by' the superordinate clause He told us, rather than being an element of it. In the Cardiff Grammar it would be regarded as a 'Participant' in the Process of 'telling', and it would be treated as an embedded clause that fills a Phenomenon that is conflated with a Complement. Thus the Cardiff Grammar treats virtually all of Halliday's "hypotactic" relationships as types of embedding.
 To be clear, in SFL Theory, the relation of clause to clause complex ('sentence') is not constituency ('make up'). In each clause nexus, one clause is the expansion or projection of the other.
 To be clear, recognising one type of interdependency, parataxis, but not the other, hypotaxis, introduces theoretical inconsistency, and prevents the recognition of agnate clause complexes (and their further agnates), as will be seen in later discussions.
 To be clear, in SFL Theory, 'co-ordination' involves two dimensions of choice: interdependency (parataxis) and expansion (extension).
 To be clear, here Fawcett implies the logical fallacy known as argumentum ad populum: a proposition is claimed to be true or good solely because a majority or many people believe it to be so.
 To be clear, in SFL Theory, the Range of a verbal Process is Verbiage, not Phenomenon. Phenomenon is a participant of a mental clause, not a verbal clause.
 To be clear, this reduces the explanatory potential of the theory. For example, it fails to distinguish embedded facts, which are clause participants, from projected ideas, which are not.
Friday, 13 November 2020
The present model takes broadly the same position as Halliday on this matter, i.e., the clause is taken to be the "highest" unit of English syntax. The term "sentence" is used here simply as a place-holder for the function served by the clause (or clause complex) in discourse. Thus it operates at the interface between the grammar and the 'higher grammar' that specifies the structure of discourse. Thus it is more like an "element" (see Section 10.5) than a "unit".³
 To be clear, if there is a highest unit, then there is a scale on which the unit ranks highest. That is, contrary to Fawcett's claims, his model is indeed formulated on the basis of a rank scale of units.
 To be clear, the term 'sentence' labels a form, not a function, and so cannot be a "place-holder" for any function served by the clause.
 To be clear, this is the only use of the term 'higher grammar' in this publication; but see later posts.
 To be clear, in SFL Theory, the term 'element' refers to a function, such as existential Process, whereas the term 'unit' refers a form, such as a verbal group. In Fawcett's Cardiff Grammar, however, an 'element' is the lowest syntactic category in the representation of a sentence, in a model that has "no rank scale" (p226):
In the present theory of syntax, the lowest syntactic category on each branch of the tree in a tree diagram representation of a sentence is an element (e.g., the head of a nominal group).
Tuesday, 10 November 2020
Halliday's formalisation of the concept of the 'rank scale' is, as with a number of his concepts, the formalisation of concepts that are found in traditional grammar. We therefore find broad equivalences to his proposals in works such as Quirk et al (1985), with the presentation of the concept that the units of the sentence (but often with reservations) clause, group or phrase, word and morpheme are typically related to each other in a 'constituency' relationship. Interestingly, however, Halliday later changed his view on the status of the "sentence", redefining it as a "clause complex", and limiting the use of the term "sentence" to clause complexes that occur in writing. Since his framework also allows for "group complexes" (and also, though very much less frequently, for "complexes" of "words" and "morphemes"), it is clear that a "sentence" is not a unit in the same sense that a clause or group is. For Halliday, then, the "sentence" is simply another type of "unit complex". The significance of this change of position is that it makes the clause the highest unit of English syntax.²
² It should be noted that, while the concept that a sentence is a "clause complex" underlies all of Halliday's analyses of texts in IFG, he nonetheless still includes "sentence" in the list of units in his discussion in Chapter 1 (1994:23) of the theoretical concepts use in IFG — while warning that this will be "re-interpreted" later in the book.
This is misleading. Halliday (1994: 23) explicitly uses the term 'sentence' as a way of connecting his functional grammar to 'folk linguistic theory' and 'traditional school grammar' for the benefit of the reader who is unfamiliar with his model:
Later, rather than "change his position", Halliday (1994: 216) explains why the term 'clause complex' is theoretically preferable to the term 'sentence' :
In later editions of IFG (2004, 2014), 'sentence' is explicitly listed as a unit of written expression (graphology); Halliday & Matthiessen (2014: 21):
Sunday, 8 November 2020
Fawcett (2010: 191):
The unit of the clause and various classes of group (or "phrase", as the group is termed in traditional grammars) are well-established in all theories of syntax. However, so far as I am aware no linguists other than those who use the Cardiff Grammar have yet recognised the unit of the cluster (though Quirk et al (1985:1276) come close to doing so in the case of the genitive cluster). For a brief discussion of this matter and an account of the various classes of cluster, see Sections 10.2.10 to 10.2.12.
 To be clear, in SFL Theory, there is an important distinction between 'group' and 'phrase', though both are positioned on the same rank, because both realise functional elements at clause rank. Halliday & Matthiessen (2014: 362-3):
… a group is in some respects equivalent to a word complex — that is, a combination of words built up on the basis of a particular logical relation. This is why it is called a group (= ‘group of words’). …
A phrase is different from a group in that, whereas a group is an expansion of a word, a phrase is a contraction of a clause. Starting from opposite ends, the two achieve roughly the same status on the rank scale, as units that lie somewhere between the rank of a clause and that of a word.
 To be clear, one reason why Fawcett's 'cluster' is not recognised in SFL Theory is that it is not necessary. To explain, Fawcett (pp212-3) provides the following examples of 'genitive' cluster:
- a girl's bike (glossed as 'a bike that is suitable for a girl'),
- my sister's most precious doll, and
- the dog's back legs;
and the following example of a name cluster:
- This is my father's
Friday, 6 November 2020
The closest that it is possible to come to the 'rank scale' concept in the theory to be described here is to extract from a wider, probabilistic set of statements (which we shall come to in Section 11.2 of Chapter 11) a statement that the various classes of cluster always fill elements of groups (though they are in fact virtually limited to one class of group, i.e., the nominal group), and that groups quite frequently fill elements of clauses (though they also frequently fill elements of every class of group and cluster, and there is one class of group — the quantity group — which fills elements of groups very much more frequently than it fills an element of the clause). Or we might take a different approach and, ignoring the cluster, we might say that, for each word in a text that is being analysed, there is a fairly good possibility that it will be functioning as an element in a group (though many high frequency words such as forms of be function directly as elements of clauses), and also that, if one has identified a group, there is a fairly good possibility that it is functioning as an element of a clause (though many groups do not).However, these heavily hedged statements are as far as one can go, in the present theory, in trying to state generalisations about the syntax of English in 'rank-like' terms. Notice, moreover, that we cannot turn such statements on their head and say, in the manner of "Categories", that clauses "consist of' groups and groups "consist of' clusters or words, because in the present model it is only some elements of clauses that can be filled by groups; it is only a very few elements of groups that are filled by a cluster (see Section 10.2.12); and it is only some elements of groups that are always expounded by words. (In any case, in the present theory we would not wish to say that groups "consist of' words, because the phenomena typically described as "words" are treated as a type of 'item', and so not as 'units'. See the discussion of this point in Section 10.5.) Thus, while some of the phenomena that originally gave rise to the concepts of a 'rank scale' of 'units' have their place in the new theory, the concept of the 'rank scale' itself plays no part in it.
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.
 To be clear, Fawcett's 'item' includes both 'word' and 'morpheme', and is said to lie 'outside syntax' in his model of syntax. Fawcett (p266):
The third of the three major categories in the present theory of syntax (with 'unit' and 'element') is the item. This term includes both 'word' (in its traditional sense) and 'morpheme'. Strictly speaking, the concept of 'item' lies outside syntax, since items are a different manifestation of meanings at the level of form from syntax.
Moreover, the 'item' is itself related to a rank scale of functions, on which 'element' is the lowest rank. Fawcett (p266):
In the present theory of syntax, the lowest syntactic category on each branch of the tree in a tree diagram representation of a sentence is an element (e.g., the head of a nominal group). And each such lowest element is expounded by an item — or as we shall see shortly, by items (in the plural).
That is, Fawcett's model also applies a model of formal constituency (rank scale) to functions (e.g. head of a nominal group), despite Fawcett denying that a rank scale is a component of the theory.
Tuesday, 3 November 2020
Fawcett (2010: 190n):
¹ In other words, the topmost unit in a representation of syntax is always a clause — even though the topmost unit in a given text-sentence may appear to be a nominal group because of ellipsis.
To be clear, if the clause is the topmost unit, and a nominal group is not the topmost unit, then there is a scale on which there are higher and lower units. That is, despite Fawcett's claims, his model makes use of a rank scale — though inconsistently, as will be seen.
Sunday, 1 November 2020
Fawcett (2010: 189-90):
One way in which a clause may be said to be a 'higher' unit than the others — as the 'rank scale' concept predicts it to be — is the fact that the unit that is shown as the 'highest' on the page in any tree diagram representation of a text-sentence is always the clause.¹ In a generative SF grammar, provision for this fact is built into the probabilities on the features in an early system in the overall network, so ensuring that a clause is always generated first. However, we may note that the informal summary of clause syntax for text analysis purposes shown in Appendix B shows that a clause is the only unit that can fill the "element" of "sentence", and this fact is given to the computer parser of texts as a relevant part of its knowledge of syntax. However, this is a fact about clauses in relation to all other units, and on its own it does not constitute evidence that there is a 'rank scale' of units in the "Categories" sense of the term.
The units of grammar form a hierarchy that is a taxonomy. To talk about any hierarchy, we need a conversational scale; the most appropriate here might seem that of size, going from “largest” to “smallest”; on the other hand size is difficult to represent in tables and diagrams, and may also trap one into thinking in substantial terms, and a vertical scale, from “highest” to “lowest”, has advantages here. For the moment we may use both, eventually preferring the latter. The relation among the units, then, is that, going from top (largest) to bottom (smallest), each consists of one, or of more than one, of the unit next below (next smaller). The scale on which the units are in fact ranged in the theory needs a name, and may be called rank.
In this view of linguistics description is, as already emphasised, a body of method derived from theory, and not a set of procedures. This has one important consequence. If description is procedural, the only way of evaluating a given description is by reference to the procedures themselves: a good description is one that has carried out the right procedures in the right order, but for any more delicate evaluation external criteria have to be invoked. Moreover, every language has to be treated as if it was unknown, otherwise procedural rules will be violated; so the linguist has to throw away half his evidence and a good few of his tools.
A theory on the other hand provides a means for evaluating descriptions without reference to the order in which the facts are accounted for. The linguist makes use of all he knows and there is no priority of dependence among the various parts of the description. The best description is then that which, comprehensiveness presupposed, is maximally grammatical: that is, makes maximum use of the theory to account for a maximum amount of the data.
 As this clause demonstrates, Fawcett's model is concerned with text generation by computer, not with language as spoken or written by humans.
 This is misleading. As Appendix 2 (pp306-7) demonstrates, Fawcett's model includes the following units, all of which are, in effect, constituents of the clause:
- nominal group,
- prepositional group,
- quality group,
- quantity group, and
- genitive cluster.
Friday, 30 October 2020
10.1.2 The remaining traces of the concept of a 'rank scale' of 'units'
It would be surprising if the concept that was the backbone of Halliday's original theory of syntax were to disappear without trace from the theory proposed here. After all, the phenomena in language that led Halliday to recognise the concept in the first place are still there. We should ask, therefore, whether the concepts of 'unit' and 'rank', in Halliday's original senses of the terms, have any correlates in the framework of concepts proposed here. Let us examine this question for a moment.
 To be clear, the validity of theorising is not a matter of whether or not it is 'surprising'.
 This is misleading, because it is untrue. The rank scale was not the 'backbone' of Halliday's superseded theory, Scale & Category Grammar. Halliday (2002 : 41):
The fundamental categories for the theory of grammar are four: unit, structure, class and system.
Traditionally these terms have usually referred to “grammar above the word” (syntax) and “grammar below the word” (morphology); but this distinction has no theoretical status. … But it seems worthwhile making use of “syntax” and “morphology” in the theory, to refer to direction on the rank scale. “Syntax” is then the downward relation, “morphology” the upward one; and both go all the way.
 To be clear, the theoretical question is how to model such phenomena to the best advantage.
 In anticipation of Fawcett's misunderstandings of Halliday's original senses of 'unit' and 'rank', it may be useful to cite Halliday's own words, as a point of departure. Halliday (2002 : 43):
The category set up to account for the stretches that carry grammatical patterns is the unit. The units of grammar form a hierarchy that is a taxonomy. To talk about any hierarchy, we need a conversational scale; the most appropriate here might seem that of size, going from “largest” to “smallest”; on the other hand size is difficult to represent in tables and diagrams, and may also trap one into thinking in substantial terms, and a vertical scale, from “highest” to “lowest”, has advantages here. For the moment we may use both, eventually preferring the latter. The relation among the units, then, is that, going from top (largest) to bottom (smallest), each consists of one, or of more than one, of the unit next below (next smaller). The scale on which the units are in fact ranged in the theory needs a name, and may be called rank.
Tuesday, 27 October 2020
Fawcett (2010: 189):
Taken together, the above facts suggest (1) that those who work in the framework of the Sydney Grammar do not in fact find the concepts of 'unit' and 'rank' useful when they are engaged in the nitty gritty work of describing a language or of describing a text; (2) that they only find the concept of 'rank' useful when embedding occurs, and (3) that in any case the term "embedding" is to be preferred to "rank shift". However, this apparent down-grading of the theoretical importance of the 'rank scale' and 'rank shift' is accompanied by a far stronger claim about the limitations on when 'rank shift' can occur than the claim made in "Categories", as we shall see in Section 11.8.5 of Chapter 11.
Finally, I should perhaps point out that, while the term "unit" is used frequently in the writings of Cardiff grammarians, it is always with the meaning of 'class of unit' (this being the topic of Section 10.2).
 As can be seen from the two preceding posts, 'the above facts' are a collection of misleading untruths.
 To be clear, this is a non-sequitur. Even if what Fawcett wrote about the concepts of 'unit' and 'rank' in Halliday's current framework had been true, it would still say nothing whatsoever about what Systemic Functional grammarians find useful.
 To be clear, each rank is the entry condition to a set of functional systems; for example, the rank of 'clause' is the entry condition to the systems of THEME, MOOD, and TRANSITIVITY. That is its paradigmatic function. By the same token, each unit is the syntagmatic domain that is structured; for example, the unit 'clause' is usually structured as Theme^Rheme. Moreover, the elements of function structure at a higher rank are realised by units of a lower rank; for example, Subject, at the rank of clause, is realised by a nominal group.
 This is still misleading, because it is still untrue; see the previous post.
 This is misleading, because it is untrue. It is the rank scale that models syntax (and morphology) in SFL Theory. It is for this reason that Fawcett misrepresents its theoretical value in SFL Theory. Rank-shift, of course, is important in accounting systematically for instances in which a higher ranked unit realises a structural element of the same rank, or below, as, for example, when a clause realises the Subject of a clause: what you see is what you get.
 As we shall see in the examination of Section 11.8.5 of Chapter 11, Fawcett misrepresents Halliday (1994) on this matter, as well.
 As previously noted, the term 'class of unit' presupposes the notion of 'unit' (of which there are classes).
Sunday, 25 October 2020
Fawcett (2010: 189):
Furthermore, in the brief section in IFG on 'rank shift', Halliday quickly introduces the terms embedding and embedded as alternatives to "rank" and "rankshifted", and he then drops the original two terms completely. So it seems that Halliday himself appears to prefer the terms "embedding" and "embedded". (This is my preference too — but Halliday would not share the reason for my preference, which is that the term "embedding" expresses the key concept in the new theory that a unit frequently occurs within a unit of the same class without invoking the notion of the 'rank scale'. Again, see Section 11.1 of Chapter 11).
 To be clear, what Halliday (1994: 188) actually writes is:
 To be clear, as can be seen in the quote above, Fawcett here confuses the notion of 'rank', a level in the rank scale of forms, with 'ranking', the status of a form that is not rankshifted.
 This is misleading. As can be seen from the quote above, Halliday does not "drop these terms completely". Instead, he explains that he uses the term 'embedded' only as an alternative to 'rank-shifted', excluding 'hypotaxis'. The reader is invited to predict whether or not Fawcett's use of 'embedded' includes the notion of 'hypotaxis'.
 This is misleading, because it flatly contradicts the Halliday quote, as can be seen above.
 To be clear, this proposition is not discussed in Section 11.1, as will be seen when Chapter 11 is examined.
Friday, 23 October 2020
We should therefore ask how far the concepts of 'unit' and 'rank' are still actually used in Halliday's current framework. As we noted in Sections 5.3 and 5.4 of Chapter 5, the concept of a 'rank scale of units' is surprisingly absent from the list of "basic concepts" of Halliday's authoritative "Systemic theory" (1993), where the focus is mainly on the grammar as a generative device. However, we also noted that in IFG it continues to be used in virtually the same way as in "Categories". Thus Halliday writes, as part of the opening sentence of Chapter 1 of IFG (Halliday 1994:3), that "a passage of English [...] consists of larger units made up out of smaller units, [and] these smaller units, in their turn, are made up out of units that are smaller still." On the other hand, the index of IFG shows that the term "unit" is not used after the introductory chapters (as we saw in Section 6.2.2 of Chapter 6), and nor is "rank", except in the two brief discussions of 'rank shift'. Matthiessen (1995) goes even further, only mentioning the concept of 'unit' twice in his work of almost a thousand pages (though he has rather more to say about 'rank' and 'rank shift').
 This is still misleading, because it is still untrue. Halliday (1995: 272):
 To be clear, this is explicitly stated in Halliday (1993); see  above.
 This is still misleading, because it is still untrue. Halliday (1994) is organised on the basis of rank. Chapters 3, 4, 5 and 7 are concerned with clause rank, and Chapters 6 and 7 Additional are concerned with group/phrase rank.
 To be clear, 'unit' refers to form, whereas Systemic Functional Linguistics is concerned with identifying the functions of forms.
189 pages into this publication, Fawcett is repeating his previous misrepresentations of Halliday instead of getting on with outlining his own theory.
Tuesday, 20 October 2020
One reason for approaching the 'new' theory in the way that we now are is that it establishes from the start that there is a major change in this basic aspect of the theory. The precise nature of the concept that replaces the 'rank scale of units' will become clearer in the next section, and clearer still in Section 11.2 of Chapter 11. We shall find that the framework of syntax proposed here retains only a few incidental traces of the concept of 'rank'. There is consequently no role in the present theory for the concept of 'unit' in Halliday's original sense of 'unit on the rank scale'.
The major critical evaluation of the concept of the 'rank scale' will come in Section 11.1 of Chapter 11, under the more general heading of "constituency", and Appendix 4 provides some further notes on 'the rank scale debate'.
Even though the concept of 'unit' is not used here, a concept that was originally derived from it plays a central role in all modern SF grammars. This is class of unit, which we shall come to in Section 10.2. However, while all systemic functional grammars use the concept of 'class of unit', there is an important difference between Halliday's criteria for defining a class of unit and the criteria used by most other systemic functional grammarians who have written on the subject, as we shall see.
 To be clear, Fawcett's claim is that 'unit' plays no part in his theory, but that 'class of unit' does. Since 'class of unit' presupposes 'unit', just as 'class of bird' presupposes 'bird', this is nonsensical. The reason why Fawcett would jettison 'unit' as a theoretical concept is that he sees it as inextricably bound to the notion of a rank scale, and he rejects the notion of a rank scale because it is the grammatical rank scale that models syntax (and morphology) in SFL Theory, and which, therefore, makes his own model redundant.
 To be clear, as previously explained, a functional grammar relates the class of a unit to the function it generally realises; that is, classes are viewed 'from above'. Halliday & Matthiessen (2014: 76):
The class of an item indicates in a general way its potential range of grammatical functions. Hence words can be assigned to classes in a dictionary, as part of their decontextualised definition. But the class label does not show what part the item is playing in any actual structure. For that we have to indicate its function. The functional categories provide an interpretation of grammatical structure in terms of the overall meaning potential of the language. For example, see Figure 2.9.
As Fawcett has previously demonstrated, he defines 'class of unit' 'from below' — that is: not in terms of the function it realises — and it will therefore be interesting to examine the validity of the views of "most other systemic functional grammarians who have written on the subject" when Fawcett eventually relates them. Of course, any claim that a proposition is valid because a majority of people support it is an instance of the logical fallacy known as argumentum ad populum.
Sunday, 18 October 2020
Thus, while the concept of the 'rank scale' has survived into the reincarnation of S&C as Systemic Functional Grammar in the Sydney version, it has not in the Cardiff version. Its centrality in the Sydney version is given a visual manifestation in the well-known summary diagram of the lexicogrammar in which the two dimensions of organisation are (1) the 'rank scale' of units and (2) the four major 'metafunctions' (e.g., Halliday (1971/73b:141), Halliday (1977/78:132) and Martin (1992:18). A similar diagram with eight 'strands of meaning' can be found in Fawcett (1980:95). However, the crucial requirement of the 'units' dimension in such diagrams is not that the units should be arranged on a 'rank scale', but that the set of units should be complete. This may sound a small difference, but it is not, as we shall see.
However, while the "Categories" concept of the 'rank scale' is clearly still present in the Sydney Grammar, we have noted (in Section 5.3 of Chapter 5) that in "Systemic theory" Halliday defines 'rank' in a curious way that avoids mentioning the concept of a 'rank scale of units'. And we have also noted (in Section 6.2 of Chapter 6) that 'rank' has an apparently diminished role in IFG.
 This is misleading, because it is untrue. The arrangement of units on a rank scale is the theoretical point of such diagrams.
 To be clear, the question of whether the set of units is 'complete' or not depends on the criteria on which they are defined. Note that Fawcett's claim is that the concept of 'unit' plays no part in his Cardiff Grammar.
 To be clear, this does not sound like a small difference. The organisation of units in a compositional rank scale is distinct from the 'completeness' of a set of units.
 This is misleading, because it is still untrue; see the earlier post Misrepresenting Halliday (1993) On Rank. What Halliday (1993: 273) actually says is:
Systemic theory retains the concepts of 'rank,' 'realisation,' and 'delicacy' from scale and category grammar. 'Rank' is constituency based on function, and hence 'flat,' with minimal layering;
That is, Fawcett does not understand that a compositional hierarchy of units (constituency) is a scale from the highest rank (clause) to the lowest rank (morpheme).
 This is misleading, because it is still untrue, and indeed, it flatly contradicts Fawcett's previous claim (p85):
the concept of the 'rank scale' is still reflected quite strongly in IFG
To be clear, each unit on the rank scale provides the entry condition to the systems of that rank. For example, the unit 'clause' is the entry condition for the systems of THEME, MOOD and TRANSITIVITY. Moreover, the following tables from IFG (Halliday & Matthiessen 2014: 20-1) put the lie to Fawcett's claim:
Friday, 16 October 2020
The concept of 'unit', in the sense defined in Halliday's "Categories", is inextricably bound up with the concept of 'rank' (as we saw in Section 2.3 of Chapter 2). In other words, there can be no concept of 'unit', in the "Categories" sense of the term, without the concept of a 'scale' of units that relates such units to each other in terms of their 'rank' on that scale — together with the accompanying set of 'rank shift' restrictions as to what 'rank' of unit may occur as an element of what other 'rank' of unit. Thus the concepts of 'unit' and 'rank' are inextricably intertwined in Halliday's theory of language, and together they make up the composite notion of the 'rank scale' that provides the backbone of Scale and Category (S&C) Grammar.
It is a tribute to the continuing influence of Halliday's founding paper "Categories of the theory of grammar" (1961/76) that the most helpful first step in explaining the new theory of syntax to be set out here to state the two basic Hallidayan concepts that it does not have. These are the precisely two concepts of 'unit' and 'rank'.
 To be clear, in both of Halliday's theories, Scale & Category Grammar and Systemic Functional Grammar, the term 'unit' refers to a unit of form — on the grammatical stratum: the clause, phrase, group, word and morpheme — and these are organised in terms of composition, such that a clause consists of one or more groups and phrases, which consist of one or more words, which consist of one or more morphemes. It is this compositional arrangement that constitutes the rank scale. Thus it is the rank scale that models what other theories model as syntax and morphology, and which, therefore, makes Fawcett's "Systemic Functional" syntax redundant.
 To be clear, here Fawcett claims that his rejection of Halliday's 'unit' and 'rank' is a tribute to the continuing influence of Halliday's theory. The question as to whether 'unit' and 'rank' are genuinely absent from Fawcett's theory will be explored in future posts.
Tuesday, 13 October 2020
There are two parts to a theory of instances of syntax: the categories and the relationships by which these are related. We shall begin in Chapter 10 with the more 'object-like' concepts of the theory, i.e., the categories. Then in Chapter 11 we shall examine the relationships. There is a fairly close parallel between this pair of concepts and the "categories" and "scales" of Halliday (1961/76). But it is important to emphasise that my term "relationships" includes a wider range of concepts than Halliday's term "scales".
The structure of what follows therefore broadly reflects that of both Halliday (1961/76) and Fawcett (1974-6/81). …
By the beginning of Chapter 12, therefore, we shall be in a position to summarise the concepts that are required for a modern systemic functional theory of syntax, and also to evaluate how far the seven "fundamental concepts" established in "Categories" are still valid today.
To be clear, Fawcett's "modern systemic functional theory of syntax" is a development of Scale and Category Grammar (Halliday 1961). This theory, which has long since been superseded by Systemic Functional Grammar, did not include system networks or metafunctions. On this basis, Fawcett's theory can hardly be called 'modern', and might be better termed a Scale and Category theory of syntax, were it not for the fact that even Halliday's first theory modelled syntax (and morphology) as a rank scale.
Sunday, 11 October 2020
In Chapters 10 and 11 we turn to the concepts that are required for the specification of 'instances of syntax'. As we have seen, these concepts are drawn on in a computer model of parsing such as that described in Weerasinghe & Fawcett (1993), Weerasinghe (1994) and Souter (1996). However, these concepts are also referred to in the realisation rules, and are in that sense presupposed by them.
As will by now clear, we shall be using the Cardiff Grammar rather than the Sydney Grammar as the baseline for constructing a modern theory of SF syntax.⁵
⁵ Apart from the reasons that follow from our findings in Chapter 7, there are two more reasons for this. Firstly, the Cardiff Grammar has taken the revolutionary proposals for changes to the theory made by Halliday in the 1960s (as summarised in Chapter 4) significantly further than the Sydney Grammar has. It has full implementations of (1) explicitly semantic system networks, (2) the concept of lexis as "most delicate grammar", (3) the integration into the system networks of the meanings realised in intonation and (4) the integration of meanings realised in punctuation. Secondly, the Cardiff Grammar provides a much fuller specification than the Sydney Grammar does of the syntactic concepts that are required, both for language in general and for the description of English in particular — especially in its recognition of classes of group and cluster that are not provided for in the Sydney Grammar.
 As previously explained, Fawcett's 'instances of syntax' are actually structures, not instances. Fawcett's model (Figure 4) confuses the realisation relation between the paradigmatic and syntagmatic axes with the instantiation relation between potential and instance.
 To be clear, Fawcett's Cardiff Grammar is his model of syntax. Halliday's "Sydney Grammar" — SFL Theory — models syntax (and morphology) as a rank scale, but Halliday (1985: xiv) explains why the term 'syntax' is inappropriate for a functional grammar:
 To be clear, Halliday's work in the 1960s was concerned with Scale and Category Grammar, not with Systemic Functional Grammar. That is, Fawcett's claim is actually that his Cardiff Grammar takes the proposals of Halliday's superseded theory further than Halliday's current theory does.
 To be clear, Fawcett does not present any of his 'explicitly system semantic networks' in this entire publication. In SFL terms, such networks are actually the grammatical networks, and can be found in Halliday & Matthiessen (2004, 2014). For genuinely semantic networks, see Halliday & Matthiessen (1999).
 To be clear, Fawcett has nowhere demonstrated, in this publication, how he models 'lexis as most delicate grammar', and it does not figure in his theoretical architecture (Figure 4). Moreover, since Fawcett locates grammatical systems at his level of meaning, his model is committed to lexis as most delicate semantics, not grammar.
 To be clear, Fawcett does not present any of the system networks of the meanings realised in intonation in this entire publication. For an SFL approach to intonation, see Halliday & Greaves (2008).
 To be clear, Fawcett does not present any of the 'integration of meanings' realised in punctuation in this entire publication.
 This misleading. The 'Sydney Grammar' (SFL Theory) does not provide any specifications 'of the syntactic concepts that are required, both for language in general and for the description of English in particular', largely because SFL Theory is not a theory of syntax; see  above.
Friday, 9 October 2020
Let me summarise. Leaving aside the "Split" and "Expand" operations of the Sydney Grammar, which are either unworkable or unnecessary, the Sydney Grammar has an equivalent for every realisation operation in the Cardiff Grammar except the first (though these are not always in a one-to-one relationship, as we have seen). These realisation operations are important concepts in the theory, as their treatment in both Halliday (1993) and Fawcett, Tucker & Lin (1993) clearly demonstrates.
However, these 'operation' concepts are a part of the grammar itself, so that they are relevant only indirectly to the outputs from the grammar — i.e., to a description of the structure of the text-sentences that are the instances of the potential specified in the grammar. Essentially, their function is to generate the relationships between the categories that we shall establish in Chapter 10. It is in Chapter 11 that we shall meet the relationships again. And it is perhaps significant that the first concept to be discussed there — that of 'rank' — has no equivalent among the realisation operations and will be rejected, while all of those to be considered in Sections 11.2 to 11.8 do have such a relationship.
 This is misleading, because it is untrue, as previously demonstrated.
 This is misleading, because it falsely presents Halliday's original model ("the Sydney Grammar") as if it were the derivative model (Fawcett's Cardiff Grammar).
 To be clear, this is both a non-sequitur and untrue. It is a non-sequitur because their being part of the grammar does not logically entail that realisation 'operations' are only indirectly relevant to the structures they specify; for example, the PROCESS TYPE system is also part of the grammar, and yet it is "directly relevant" to the experiential structure of the clause. And it is untrue because 'operations' (realisation statements) "directly" specify how system selections (paradigmatic axis) are realised structurally (syntagmatic axis).
 To be clear, here again Fawcett misunderstands the realisation relation between system and structure as the instantiation relation between system and instance — and he does so despite the fact that his term 'realisation operations' explicitly identifies the relation as realisation, not instantiation.
 To be clear, the claim that Fawcett's realisation operations (listed below) generate relationships between categories will be tested in the examination of Chapters 10 and 11.
1. Insert a unit (to fill an element).2. Locate an element at a place in a unit.3. Conflate an element or Participant Role with an existing element.4. Expound an element by an item.4a. Fetch a name to expound an element.5. Prefer certain features on re-entry to the system network , including preselection.6. For an element, re-enter the system network.
Tuesday, 6 October 2020
Fawcett (2010: 184-5):
Matthiessen & Bateman (1991) and Matthiessen (1995) call Halliday's "Split" operation "Expand", but otherwise there is, in principle, no difference between them and Halliday. In practice, however, they do not fully implement Halliday's concept that the MOOD structure consists primarily of "Mood + Residue" and only secondarily of "Subject + Finite + Complement" (or whatever it happens to be). See the examination of their account of generation in Section 7.4.2 of Chapter 7, which showed that their model is not a 'structure conflation' model, as the IFG model is, but an 'element conflation' model. As I argued there, this fact demonstrates that in the theoretical-generative strand of work in the Sydney Grammar the concept of 'structure conflation' is unnecessary, undesirable, and ultimately unworkable.
 This is misleading, because it is untrue. It is only Fawcett, not Halliday, who regards these structures as 'primary' and 'secondary'.
 This is misleading, because it misrepresents Halliday's model. As previously demonstrated in the examination of Fawcett's Section 7.4.2, and shown above in Halliday's realisation statement (b), in SFL Theory, it is only elements that are conflated, not structures.
Sunday, 4 October 2020
The other line of structure where Halliday frequently shows more than one line of structural analysis is in THEME, and this happens whenever there is a case of 'multiple theme'. For example, on page 55 of IFG there are three lines of structure for THEME in the analysis of Well but then Ann surely wouldn't the best idea be to join the group. On the pattern of the IFG approach to the structure of MOOD (as in Figure 7 of Chapter 7), Halliday's grammar would operate by first generating the 'multiple theme' of Well but then Ann surely wouldn't the best idea, since he analyses all of these items as different 'subcomponents' of the 'Theme', and it would then "split" it into its separate parts. It seems most unlikely that Halliday (or anyone else) would wish to treat all of these as parts of a single element, but that is the clear implication of the introduction of his "Split" operation. (See Section 10.3.4 of Chapter 10 for a discussion of 'primary' and 'secondary' structures.)
 To be clear, the example in Halliday (1994: 55) is:
 As previously explained for Mood, the grammar does not first "generate" a multiple Theme and then expand it into textual and interpersonal sub-components. This is to mistake a system network of relations for an algorithmic procedure. The identity relation between system and structure is intensive (elaboration), not circumstantial (enhancement: temporal): thematic structure realises systemic features selected from simultaneous networks, including those that specify textual and interpersonal Themes; Halliday & Matthiessen (2014: 106):
 Again, it is only Fawcett who regards these structures as 'primary' and 'secondary'.