Saturday, 25 November 2017

John Bateman On "The 'Space' Between A Grammatically-Induced Semantics And Any Contextualised Use Of Involved Grammatical Constructions"

John Bateman wrote to sysfling on 2 November 2017 at 19:10:
I guess I should say something since the word 'space' came up.... for variety, spice and funny symbols, see:
  • Bateman, J. A. Language and Space: a two-level semantic approach based on principles of ontological engineering. International Journal of Speech Technology, Springer, 2010, 13, 29-48.
and for this as a further refinement of Michael and Christian's Construing Experience:
  • Bateman, J. A.; Hois, J.; Ross, R. J. & Tenbrink, T. A linguistic ontology of space for natural language processing Artificial Intelligence, 2010, 174, 1027-1071
One of the main points of these papers (derived from our 12 year collaborative research center on space ...) is to open up the 'space' (ha ha) between a grammatically-induced semantics and any contextualised use of involved grammatical constructions so that the enormous flexibility of ranges of interpretation is both made visible and constrained sufficiently so that it can be worked with productively. In the simple cases, one *might*, along that path, get to descriptions of locations-in-space (measurable, movable-in, etc.); in most cases, however, the constructed context is more interesting. One is in any case dealing at least with places rather than locations (i.e., social constructions and not geometry).

Blogger Comments:

[1] See John Bateman Positively Judging Himself.

[2] For a grammatical analysis of this clause nexus, see The Transitivity Of Bamboozlement.

[3] To be clear, "a grammatically-induced semantics" is the SFL model in which semantic potential is construed by lexicogrammatical potential; and "contextualised use of involved grammatical constructions" refers to the grammatical structures that realise grammatical systems instantiated in texts that realise situations.  The 'space' between the two is not a theoretical space; the two are related along three theoretical dimensions simultaneously:
  • stratification (semantics realised by lexicogrammar),
  • axis (system realised by structure),
  • instantiation (text as instance of system).

[4] Here Bateman follows Martin (1992) in misconstruing ideational semantics (the meaning construed) as ideational context (the field of culture construed).  For Martin's misunderstandings of context see here; for Martin's misunderstandings of stratification, see here.  For Bateman's endorsements of Martin's theoretical misunderstandings, in general, see here.

[5] To be clear, on the SFL model, socially constructed "places" and geometrical "locations" are construals of experience as meanings of socio-semiotic systems. For more on Bateman's epistemological assumptions, see John Bateman Denying The Existence Of Text.

Monday, 6 November 2017

David Banks' Reasons For Not Identifying 'Accompaniment'

 David Banks replied regarding Annabelle Lukin's query on sysfling on 2 November 2017 at 22:09:
It seems to me that if you say "between Iraqi and British forces" is a circumstance, that means that it is functioning at the level of the clause, and it is a circumstance of the predicator "have been". But I don't feel that holds water. If "between Iraqi and British forces" is related to a process, then that process is "clashes", which is a nominalized process, and the relationship is not circumstantial but agentive, so in the unpacked variant "Iraqi and British forces clashed", "Iraqi and British forces" functions as actor. This is not unusual; it is common in cases of grammatical metaphor, for the putative actor of a nominalized process to be encoded in a prepositional phrase functioning as qualifier of the nominalized process. It is the reciprocal nature of the process in this case that gives us the process "between" rather than "by".

If we simply ask: what is "between Iraqi and British forces" doing, surely the answer is: telling us about the nature of "clashes", that is, it is describing "clashes". Since "clashes" is a noun, an element which describes it is a modifier or qualifier. 
The only complication arises because "clashes" is a nominalized process. Who is doing the "clashing": answer: "Iraqi and British forces". So the qualifier encodes the putative actor of the nominalized process.

and on 3 November 2017 at 19:22:
Among the numerous variants we have:
"Iraqi forces clashed with British forces"
"British forces clashed with iraqi forces"
"Iraqi and British forces clashed"
So, I feel, you can't get away from the fact that "Iraqi forces" and "British forces" are the participants in the process of "clashing", and the grammatical metaphor (used for reasons we do not know since it's out of context) means that the participants are encoded as a qualifier of the nominalized process.

I agree with almost everything David wrote, with the possible exception of needing to agree to disagree! :-)

Blogger Comments:

Reminder: the clause under discussion is
there have been more clashes between Iraqi and British forces.

[1] Trivially, a circumstance is attendant on a Process (experiential), not a Predicator (interpersonal), and have is Finite and been is Predicator.

[2] This is incorrect.  The Process of this clause is have been, not clashes; clashes functions as the Thing/Head of the nominal group more clashes which functions as the participant Existent.

[3] This is true; clashes is a nominalised process, which is why it doesn't function as the Process of this metaphorical clause.  It is in the more congruent agnates, which Banks cites, that it functions as Process.

[4] The agency of both the metaphorical clause and its congruent agnates is 'middle', not 'effective';  that is, there is no Agent.

[5] This part is true, but irrelevant, because the query is about the metaphorical clause, not its more congruent agnates.

[6] An instance of this type would be there have been more attacks by British and Iraqi forces.  In this type, the prepositional phrase is indeed functioning as Qualifier, since, unlike the clause under discussion, it can't be thematised without more attacks; see previous post on determining constituency.

[7] Here Banks repeats Bartlett's misinterpretation of 'joint participation in the process' as 'reciprocality'; see previous post.  Since the process is not 'reciprocal' — the British and Iraqi forces didn't "clash each other" — this is not the reason for the minor Process being between rather than by.  The function of between is to construe the joint participants as a circumstance; see [8].

[8] The prepositional phrase metaphorically construes the joint participants in what is congruently a material Process, 'clash', as circumstantial to the existential Process have been.  The type of circumstance that construes joint participation is 'Accompaniment' (Halliday & Matthiessen 1999: 174; Halliday & Matthiessen 2014: 324).

[9] This correctly identifies agnates of the metaphorical clause, demonstrating both Accompaniment (with British forces, with Iraqi forces) and its relation to 'joint participation' (Iraqi and British forces), but still fails to make the theoretical connection.

[10] What we can say is that the grammatical metaphor reconstrues a very material process as one of merely existing, and reconstrues the participants of that Process, both Agent and Medium, as merely circumstantial.  Given that it was Britain that illegally invaded Iraq, and not the contrary, we can take a wild stab in the dark and suggest that one reason the metaphor was used was to hide the agency of British humans in the "clash", along with the fact that their actions impacted on Iraqi humans.

[11] See [1] to [10] above.  See also Thoughts That Didn't Occur…

Saturday, 4 November 2017

Tom Bartlett's Reasons For Not Identifying 'Accompaniment'

How about this:
There have been more clashes between Iraqi and British forces
The process type is simple. But what kind of circumstance is "between Iraqi and British forces"?

I'd say it's a qualifier in the ngp.

My feeling here is that it could usefully be analysed as a marker of reciprocality, i.e. where processes are two way, such as:
they married/fought/loved/hated each other 
These have nominal equivalents of
the marriage/fighting/love/hate between them
For that reason I would also stick with the qualifier analysis. 
It's a bit difficult to talk about agency though as love and hate are non-agentive processes (just Medium and Range).

Blogger Comments:

[1] The question of whether the prepositional phrase between Iraqi and British forces functions as a circumstance or a Qualifier is one of constituency: does it realise an element of clause structure (circumstance) or does it realise an element of group structure (Qualifier)?  Halliday and Matthiessen (2014: 270-1) provide a simple means of determining this:
To differentiate [circumstance from Qualifier] in analysis, we can apply textual probes: in principle, being an element of the clause, a circumstance is subject to all the different textual statuses brought about by theme, theme predication and theme identification. … In contrast, a Qualifier cannot on its own be given textual status in the clause since it is a constituent of a nominal group, not of the clause; so it can only be thematic together with the rest of the nominal group it is part of.
Bartlett's analysis of the prepositional phrase as Qualifier is ruled out by the textual agnate:
it is between Iraqi and British forces that there have been more clashes.
cf the Location agnate:
it is between the Tigris and Euphrates rivers that there have been more clashes.

[2] This confuses reciprocality with joint participation in a process.  On the basis of Bartlett's examples, the omitted clause they clashed each other falsifies the interpretation as reciprocality.  On the other hand, a more congruent agnate of Lukin's clause
Iraqi and British forces have clashed again
demonstrates that the circumstance between Iraqi and British forces is agnate to joint participation in a process.  The type of circumstance that is agnate to joint participation is Accompaniment.  Halliday & Matthiessen (2014: 324):
Accompaniment is a form of joint participation in the process and represents the meanings ‘and’, ‘or’, ‘not’ as circumstantials;
Halliday & Matthiessen (1999: 174):
The circumstance of Accompaniment does not correspond to any one particular participant rôle. Rather, it corresponds to an extending of the participant itself, by addition or variation… Grammatically, the analogous type of participant is one represented by a nominal group complex.
A transitivity analysis of Lukin's clause can be viewed on the Sysfling blog here.  Other instances of non-spatiotemporal uses of 'between' can be viewed on the Sys-Func blog here.

[3] These nominal groups provide no grammatical reasoning for interpreting Lukin's clause, since they are not agnate to it.  Being nominal groups, they are merely instances in which the prepositional phrase does actually function as a Qualifier.  As an argument, this can be seen as an example of the logical fallacy known as begging the question.

[4] To be clear, emotive mental Process clauses are potentially 'two-way': emanating ('like' type) or impinging ('please' type).  It is only the emanating type that is 'middle' in terms of agency ("just Medium and Range").  Impinging mental Process clauses are 'effective' in terms of 'agency'.  However, 'effective' agency can also be realised in analytical causatives, as in you made me love you: