Home » Articles » NG19. Interrogatives – polar and pro-adverb

NG19. Interrogatives – polar and pro-adverb

InterrogativesNG analysis is now applied to yes/no questions and to questions with how, why, when, where etc – i.e. where the answer would be an adjunct rather than an argument of the predicate (as with who or what).

A digression concludes that all auxiliaries can be treated as the same for the purpose of allocating arguments to main verbs – not just in questions but in all sentences.

Is Nero…?

A polar (yes/no) interrogative is formed by fronting an auxiliary.  The (noun)__is junction already seen must have a corresponding is__(noun) junction, the latter delivering an additional proposition signalling that the clause-type is a question.

Sentence (5) becomes:

(55) Is Nero giving Olivia to Poppaea?

This approach seems to work with most of the declarative variants listed in LS15.  Those with fronting need the subject brought forward to follow the auxiliary.  For example, (9) becomes:

(56) Is Nero to Poppaea giving Olivia?

Sentences like (56) with a preposition phrase before the verb are at best marginal.  Interrogatives derived from declaratives with two bare nouns before the verb don’t work at all.  For example, (13) becomes:

(57) * Is Nero Olivia giving to Poppaea?

The reason could be the ‘garden path’ at Olivia, with Is having been treated as copular.   The Is__(noun) junction must allow for:

(58) Is Ganymede Rosalind?

It must also allow for polar interrogatives like (55) with only one bare noun before the main verb, while disallowing (57) and similar.

Revisiting LS12 should explain how sentence (58) must create the following from Is__Ganymede and Is__Rosalind.

LanguidSlog

The shared node represents ‘something instantiated by GANYMEDE and by ROSALIND but by nothing else’ – which can only mean GANYMEDE and ROSALIND are the same thing.  The leading verb also creates a proposition (not shown) indicating a question.

That ‘garden path’

In (57) everything would be OK if the sentence ended at Olivia.

The puzzle is why sentence (13), Olivia Nero is giving to Poppaea, doesn’t also fail.  This starts with Nero__is and Olivia__is and should create a pair of propositions like those pictured above.  Obviously the verb-created tense/aspect proposition (again not shown) will signal a declarative rather than an interrogative but the sentence should fail at giving.

Clearly Is X Y? is a far more common type of sentence than (13).  It’s therefore reasonable to assume that whatever makes the difference happens in (13) – something not shown in the analysis in LS14.

At this stage in developing NG, my best guess is that (noun)__is creates something different when it is prosodically distinctive as it would be in (13).  To be topicalised the noun needs emphasis as well as fronting.

That can’t easily be done in (57) and the sentence fails.

Can Nero…?

Generally the approach illustrated above for sentence (55) should work for other auxiliaries.

(59) Can Nero give Olivia to Poppaea?

This analysis is exactly the same.  The difference is yet again the unseen proposition.  This time, modality for can is important as well as tense/aspect and clause-type (‘interrogative’ in this case).

Sentence (57) is ungrammatical and its equivalent with can is certainly awkward.

(60) ? Can Nero Olivia give to Poppaea?

There is no garden path at give.  However I propose to analyse all auxiliaries in the same way as is has been analysed in LanguidSlog thus far.  That means everything distinctive about the auxiliary must be bundled into the ‘unseen’ proposition it creates.

The LS piece in which those propositions get defined is not yet on the schedule.  I hope you can bear with me.  It’s impossible to do everything at once.

Of course, under this proposal, (60) is deemed to be ungrammatical.

I offer no suggestion as to why modals take an infinitive main verb while be and have forms take participles.  Therefore language knowledge can’t be generalised across all (auxiliary)__(main verb) junctions. In any case the ‘unseen’ proposition will be different for each of the small number of auxiliaries.

Why is Nero…?

The closed interrogatives discussed above can be turned into open interrogatives by preceding them with how, why, where, when.  Each of these stands for an adverbial and so they can be called pro-adverbs (by analogy with pro-nouns).  Any of these can be in a junction with the auxiliary verb as parent, for example why__is.

An adverbial adjunct is optional and cannot fill any of the thematic roles of the main verb.  Thus Why is Nero giving Olivia to Poppaea? has AGENT, THEME and GOAL as well as the pro-adverb.

Who is Nero…?

Well, the story about pro-adverbs was short and sweet.  Interrogatives with who or what appear similar but are much more challenging – as we’ll start to see next week.

Mr Nice-Guy

Comments