[Sysfling] Leaving Medium implicit?

David Rose david.rose at sydney.edu.au
Fri Dec 1 21:39:56 GMT 2017


Hi Gordon

> As for where (lexical) items reside, all I can add or reiterate is that for us in Old South Wales, the 'item' is a unit of (arbitrary) form that 'expounds' (realises/lexicalises/lexifies) an element of structure in a grammatical unit. So for example, in 'national anthem' the lexical item 'anthem' expounds the head of the nominal group. The particular lexical item selected will be that corresponding to a feature or set of features in the system network. As you know, in CG systemic features are semantic in nature, but even the features that Ruqaiya posits for distinguishing processes in her 1987 paper also seem almost completely semantic in nature.
> 
> Your turn, now David. Where does the lexical item come in in your model's architecture?

Oh alright ;) a different but hopefully complementary perspective. I’m interested in social context as a semiotic plane that is organised systemically and realised by language, including grammar and discourse systems. Grammar and discourse construe register (field/tenor/mode) in different ways, partly by configuring lexical items. Eg IDEATION configures items in taxonomies, figures and sequences, TRANSITIVITY in processes, participants and circumstances. Each of these contribute general types of patterns to unfolding construals of a field. Lexical items contribute more specific features of register.

We can often categorise lexical items according to grammar or discourse patterns they tend to instantiate (as Michael and Christian like to do in IFG). But these categories are only one dimension of their meaning. Their more specific meanings can only be described in terms of register systems (the function of dictionary and thesaurus). So my instinct is to locate them at the level of register. This is hard to conceptualise because they only manifest as wordings, so we usually talk about them as if they are words. But the items themselves are meanings (at register level), that are realised in language as wordings. In text, they co-instantiate with grammar and discourse features to construe patterns of register. These probabilistic co-instantantions may be the basis of much linguistic description.  

in a nutshell
David


> 
> Very best wishes
> 
> Gordon
> 
> 
> 
> -----Original Message-----
> From: David Rose [mailto:david.rose at sydney.edu.au] 
> Sent: Friday, December 1, 2017 2:19
> To: Gordon Tucker <TuckerG at cardiff.ac.uk>
> Cc: sysfling at cardiff.ac.uk
> Subject: Re: [Sysfling] Leaving Medium implicit?
> 
> Hi Gordon
> 
> For the Royal Wedding, the UK needs a better national anthem, particularly since the Tories don’t want God to save Queen Megan. How about that rousing Monty Python classic “Spam, spam, spam. Spam, spam, spam.”
> 
> Love your evocation of co-occurrences. Reminds one of Jim’s cascading coalescences. Personally I like co-instantiation for integrating instantiation with stratification.
> 
> Only problem I have with Ruquaiya’s "only lexical item capable of functioning” is absence of probability. Isn’t the item/structure relation probabilistic? I’d be happy with probabilities of co-instantiation of strike (back) with either middle material or 1-Role Process. 
> 
> I’m still interested in where people locate lexical items in the architecture. Ruqaiya was trying to prove they could be delicate grammatical options. In 'that less well-known variety of SFL’, if Process is located in semantics, and Main Verb in syntax, where does 'the lexical verb 'strike’ reside? 
> 
> generally puzzled
> David
> 
>> On 1 Dec 2017, at 8:02 am, Gordon Tucker <tuckerg at cardiff.ac.uk> wrote:
>> 
>> Hi David,
>> 
>> I've just discovered your two posts, which I found in my SPAM mailbox!! Does my spam filter know something I don't?!
>> 
>> I too am puzzled when you ask if, in the kind of description I gave, ' we are actually classifying lexical items according to grammatical criteria'. 
>> 
>> For me a lexical item is that unit of form that enters into co-occurrence relations with other units of form, with strings of units of form, with co-text in discourses, with contexts etc. etc.  All grammatical and functional organisation is derived from our exposure to the range of co-occurrence instances. Indeed lexicalists like Michael Hoey, would say - I think quite reasonably - that our grammars emerge from our exposure to such co-occurrences. 
>> 
>> Yes, lexical items are classified (by many) according to 'grammatical criteria'. The various co-occurrences with a lexical item like 'strike' allow us to categorise the item (should we so wish to categorise!) as both a 'noun' and a 'verb', or more functionally/semantically as both a THING and a PROCESS/EVENT. 
>> 
>> So what is the relation between the lexical item and a functional structure? Well, the lexical item 'functions' as some functional category. Take for example a statement by Ruquaiya Hasan (1987:198) 'In English, the only lexical item capable of functioning as the Event in a clause with the above selection expression is 'scatter'. I find this kind of statement completely unproblematic. In the Cardiff Grammar, we would say something very similar. So the lexical item 'scatter' (as a chunk of form) functions as Event, Process, Main Verb (or Predicator) in the functional structure of the clause. 
>> 
>> Am I missing your point? If so, it could be because of the immense excitement I am feeling over next year's Royal Wedding in the UK, which I have been asked to contribute to!
>> 
>> Gordon
>> 
>> 
>> -----Original Message-----
>> From: d.rose at edfac.usyd.edu.au [mailto:d.rose at edfac.usyd.edu.au]
>> Sent: Thursday, November 30, 2017 3:05
>> To: Gordon Tucker <TuckerG at cardiff.ac.uk>
>> Cc: John Bateman <bateman at uni-bremen.de>; sysfling at cardiff.ac.uk; 
>> peter.strom at lnu.se
>> Subject: Re: [Sysfling] Leaving Medium implicit?
>> 
>> Hi Gordon and friends
>> 
>> Perhaps I'm being obtuse, but can I ask if, whether our perspective is transitive or ergative, when we describe items like 'strike' or 'strike back’ in terms like 1-Role Processes or 2-Role Processes (or ergative/transitive or whatever), we are actually classifying lexical items according to grammatical criteria?
>> 
>> By the term Process do we mean a lexical item or an element in a function structure, such as Actor+Process? If the latter, then what is the relation between the function structure and the lexical item?
>> 
>> many thanks
>> David
>> 
>>> On 24 Nov 2017, at 12:54 am, Gordon Tucker <TuckerG at cardiff.ac.uk> wrote:
>>> 
>>> I would tend to support John's analysis here.
>>> 
>>> Working in an SFG framework which does not use the ergative approach involving a Medium, the issue would be whether the Processes 'strike' or 'strike back', as here, are 1-Role Processes or 2-Role Processes.
>>> 
>>> There is arguably sufficient evidence to suggest that these are not Agent/Actor + Affected/Goal Processes but Agent/Actor/Medium Processes. Certainly, with 'strike back', in an IFG approach, in examples such as X struck back/against Y, the 'back/against Y would be a Circumstantial Role, rather than a Participant Role. 
>>> 
>>> Similarly, 'strike', as in 'the hurricane struck' might be considered a 1-Role Process, rather than a 2-Role Process, as in the hurricane struck the coastal area of Haiti'.
>>> 
>>> The test then, is which Participant Roles are inherently associated with a given Process - and indeed how do we establish that? The same test would apply to Processes such as 'fight', and 'fight back'.
>>> 
>>> If a Process is found to involve 2 Participant Roles, e.g. Agent + Affected, then it would be acceptable to posit an implicit non-realised second participant (i.e. the Affected). Whether one introduces an 'empty' Complement/Participant Role into the functional structure is a different matter. If it found to involve only 1 Participant Role, then on what grounds could one talk about an implicit second PR?
>>> 
>>> Gordon
>>> 
>>> 
>>> Dr Gordon Tucker
>>> Honorary Research Fellow
>>> Centre for Language and Communication Research School of English, 
>>> Communication and Philosophy Cardiff University John Percival 
>>> Building Colum Drive CARDIFF CF15 3EU
>>> 
>>> 
>>> 
>>> 
>>> 
>>> -----Original Message-----
>>> From: sysfling-bounces at cardiff.ac.uk 
>>> [mailto:sysfling-bounces at cardiff.ac.uk] On Behalf Of John Bateman
>>> Sent: Thursday, November 23, 2017 10:20
>>> To: sysfling at cardiff.ac.uk; peter.strom at lnu.se
>>> Subject: Re: [Sysfling] Leaving Medium implicit?
>>> 
>>> ... errr... but the Medium isn't implicit: it's the empire; as in 'John runs' ... which I will now do...
>>> 
>>> John.
>>> 
>>> _______________________________________________
>>> Sysfling mailing list
>>> Sysfling at cardiff.ac.uk
>>> https://mailman.cf.ac.uk/mailman/listinfo/sysfling.
>>> uk
>>> 
>>> _______________________________________________
>>> Sysfling mailing list
>>> Sysfling at cardiff.ac.uk
>>> https://mailman.cf.ac.uk/mailman/listinfo/sysfling.
>>> uk
> 



More information about the Sysfling mailing list