Syntax


Nothing is official as yet, but I’m fairly confident that I can informally announce to the world that I will be commencing a Ph.D. next year.

My topic will be Classical Tiwi, an Australian language that seems to have escaped the radar for serious documentary research of late. This is especially odd, given that Tiwi1 is one of the country’s most populous languages with somewhere around 2000 native speakers2. Of course this is not quite the case when it comes to Classical Tiwi, which may have only around 250 speakers, many of them elderly.

I’ve been interested in Tiwi for quite some time, as a relative of mine married a Tiwi Islander, right when I started becoming interested in Australian languages. I even remember looking at the list of the authoritative publications for Australian languages, and noting that Tiwi was researched as far back as 1976. I somewhat facetiously told myself that I was going to do my Ph.D. on Tiwi and give Osborne’s 19763 description a surely-needed update.

Then, earlier this year, I was approached by a colleague who suggested for a bunch of reasons that I do Tiwi for a Ph.D., not knowing that I had Tiwi family connections and a previous interest. Quite serendipitous.

I’ll be enrolled at the University of Melbourne, so if all goes well throughout the application process, I should be looking to move to Melbourne sometime in early 2010.


  1. Officially there isn’t a difference between Modern Tiwi and Classical Tiwi, meaning that ‘Tiwi’ is considered one language still.
  2. The census numbers vary considerably. In 2006, 1724 people said they used Tiwi at home, while in 2001, the number was 2050, and I suppose people tend to overreport more than they underreport.
  3. Osborne, C. R. (1974). The Tiwi language : grammar, myths and dictionary of the Tiwi language spoken on Melville and Bathurst Islands, northern Australia. Canberra: Australian Institute of Aboriginal Studies.

My brother this morning uttered a sentence that I think deserves a bit of syntactic analysis. The context, if you can’t recover it from the sentence itself, was essentially my brother swapping a telephone cable, which resulted in the new cable sagging a bit with the slack. There is, however, a hook whose purpose is to take up the slack, except that it wasn’t in quite the right spot. Thus:

That hook could use moving.

This amused me somewhat, and much to their chagrin, I let everyone present know¹. It makes perfect sense to me, even if it’s a little difficult to see how the whole is composed by its parts, so I’m interested in how it came about.

I see the influence, and intersection, of a couple of other idiomatic syntactic constructions here, which I’ll refer to as the could use construction and the needs verbing construction.The could use construction was, I reckon, more originally said of animate subjects and refered to tangible things, such as:

I could use a torch

From here, it’s only a short journey to more abstract arguments, although the subject would still be an animate, as in:

You could use a break

This then would be taken to be euphemistic version of something like ‘I need a break’. Which brings me to the next construction of which this sentence was reminiscent: the needs verbing construction. I believe Language Log addressed this construction a while back, at least once, but I can’t find any record of it. The basic idea is, take a full sentence of the format x needs to be verb-en, and reformulate it such that it becomes x needs verbing. So your dog needs to be washed (unequivocally transparent syntax there) becomes:

Your dog needs washing

If we consider the lexical specifications of the quasi-modal verb need, then I hope we can agree that in its canonical form, it takes a complement, which usually surfaces as an object, as in:

I need a taco

It’s also possible for need to take as its complement an S (sentence) beginning with a to-infinitive verb, whose subject is functionally controlled by the subject of the matrix verb, or, if there is one, the object².

Let me put that another way: take a sentence like:

I need to do the washing

The person who does the washing here is the same person who does the needing: I. Whereas in:

I need you to do the washing

The person who does the washing is you instead (if you accede to my request, that is), so the controller of the subject of wash, in each instance, is the nearest argument. I’m getting slightly off-track, so ignore these little tangents relating to LFG and recall what I said about need in its canonical sense taking an object as its complement (I need a taco). Morphosyntactically speaking, a direct object is a noun, so it could be filled by a gerund; the -ing form of the verb that acts as a noun, as in his doing the dishes impressed me. This might be a red herring, but is it possible that the verb in the need verbing construction is in fact a gerund?

This analysis is probably getting a little bit too big for its boots by now, so I might wrap it up. I believe what my brother intended to say was that hook needs to be moved, which, on account of the entirely common needs verbing construction, becomes that hook needs moving. Finally, taking the rough synonymy in this instance of could use and needs, he came out with a slightly more euphemistic sentence that on one hand, implied that I should in fact move the hook while, on the other, cushioning the imposition on me to actually do something³, and produced:

That hook could use moving

Brilliant. Is this how people do construction grammar?


¹It’s quite normally the case that my occasional bursts of intense amusement in totally minor linguistic curios solicit sighs of impending boredom from everyone within earshot. That is, until I met my nibulin⁴, who is also a linguist and is similarly amused, just as intensely, by such things.
²I might be wrong about one or two points of terminology here, such as anapahoric versus functional control as it’s been quite a while since I’ve done any lexical-functional grammar. If you spot anything, let me know.
³There’s an awful lot of speech act theory and conversational politeness theory bound up in that which I don’t really have the time to go into, but it’s interesting nonetheless.
⁴I’m not going to define this for you – if you really desperately want to know what it is you can find the online Wagiman dictionary and look it up.

or, On the Grammar Wars1

Over the weekend, and extending into the week thus far, a debate has been steadily growing in the blogosphere, both here and in the US, about a controversial set of guidelines for teaching English published last year by the English Teacher’s Association of Queensland (ETAQ).

Before I go on, I might say that the breadth of this debate is such that I barely know where to begin, so logically, I might try beginning at the start.

The model upon which ETAQ’s guidelines were based is called Systemic Functional Linguistics (SFL), a framework of textual analysis devised by M.A.K. Halliday back in the 60s, that focuses on the interpersonal and ideational functions of language; language as its used in a larger social context. The name indicates that it is concerned with ‘systems’, i.e., texts of indefinite size, and entire communicative practices, and with ‘functions’ as opposed to ‘forms’. This is important and I’ll come back to it.

The terminology of SFL is quite specialised, and for a good reason. Since it concerns functions and not forms, overall meanings rather than constituency, SFL had to create a whole new set of terms that differed from formal categories, so as to not have the same terms meaning something different to someone who uses a different framework. SFL is, in my opinion, commendable for this, otherwise things would certainly be confusing.

The author of the guidelines, Dr Lenore Ferguson, took a rather confused version of SFL’s already specialised terminology, and went on to write it up as the definitive model for English structure teaching. This has resulted in serious formal errors that have led critics, such as Geoffrey K. Pullum and Rodney Huddleston, to describe the guidelines and other related publications as Not just a little bit ropey, but absolutely incompetent, full of utter howlers2. Here is The Australian‘s description:

A TEACHERS’ guide to grammar circulated by the English Teachers Association of Queensland is riddled with basic errors, leading an internationally respected linguistics professor [Rodney Huddleston] to describe it as “the worst published material on English grammar” he has seen.

What sort of ‘utter howlers’ are they talking about?

Here are a couple of the now well-publicised errors that Huddleston took as exemplary and vocally objected to:

  • Won’t in The boy won’t eat his lunch is labelled an adverb; it is actually a modal auxiliary verb,
  • Capable of in The boy is capable of eating his lunch is also described as an adverb; it is really an adjective followed by a prepositional phrase headed by of, and
  • A set in A set of bowls is called an adjective; it isn’t even a grammatical unit at all. It’s half a noun phrase – the rest of it is a prepositional phrase of bowls. A set on its own is just a noun phrase, or a determiner followed by a noun.

These are pretty indefensible errors, one might assume, and it’s almost impossible to imagine anyone even rudimentarily trained in linguistics making them. But I’ll show later that, although they’re still indefensible, they’re completely explicable based on what Dr Ferguson was trying to do. I’ll also hopefully conclude that SFL has been copping too much flak in the whole debate.

Okay, just a little on the rest of the debate. Language Log caught onto the issue early on when Geoffrey K. Pullum posted this summary of the controversy, and he and other LanguageLoggers have since posted several follow-ups, the last of which is brilliant. Closer to home, Larvatus Prodeo, one of Australia’s most widely read political/social/cultural blogs, had this post, which has solicited a massive reader response; 163 comments and counting.

As you might expect, most of these comments, as well as the various letters to The Australian in response to this issue, and much of the other online coverage this issue has received, comprise people crying fowl of prepositions at the end of sentences, conjunctions at the beginning, blah blah blah. Apart from that, there have been some slightly more informed people arguing the differences between prescriptive grammar and descriptive grammar, which is totally irrelevant to the ETAQ’s teaching guidelines, Huddleston’s response or anything3.

Despite my continued efforts, I have been unable to force the debate back on track, and before I give up, the rest of this post is how I might characterise the main issue and add my perspective, for whatever it’s worth.

As I mentioned earlier, SFL deals with language as it’s used in society and larger contexts, and it deals with its function in those contexts; not, crucially, its forms. So instead of dwelling too much on nouns, noun phrases, prepositions, clitics, perfect participles, structural categories and so on, SFL talks about participants, processes and circumstances as the basic sentential units.

That is, sentences are analysed in SFL as to who is being talked about, the participants; what is happening, the process(es); and optionally, any other adjunctive information such as location, the circumstances. SFL still has formal units at the heart of this; sentences and the constituent words are all labelled (mostly correctly) as to their parts of speech, but the main priorities are the  discourse functions of language that operate at level different from the parts of speech. It assumes that a structural analysis of the individual sentences using ‘traditional’, or generative grammar in this context, has already taken place.

To emphasis one important point already made, central to SFL, and many models of language, is the independence of form and function; the difference between phonological shape and meaning. This is one thing that Ferguson’s English teaching guidelines, based heavily on SFL, omitted, and it would thereby be regarded by probably every practicing linguist today as an inadequate theory of the structure of language.

Since form and function were collapsed, and since SFL prioritises the function of language, the result is a framework that, when applied as a structural analytical tool for parsing sentences, mislabels parts of speech. To take an example from those cited above, the sentence The boy is capable of eating his lunch would have a well-defined structure that many linguists would easily be able to draw for you. Here’s the bracket notation (simplified) that you can insert into the Syntax Tree Generator to see it4:

[S [NP [DET The] [N boy]] [VP [V is] [AP [A capable] [PP [P of [VP [V eating] [NP [DP his] [N lunch]]]]]]]]

SFL would then go on to analyse the sentence as to its functional aspects. Taking for instance the complex adjective phrase capable of eating his lunch, SFL would see the actual process as ‘eating’, while the beginning of the adjective phrase in which it is embedded, would be seen to contribute an adverbial element, since it arguably modifies the way in which the boy’s eating habits are seen.

Thus, SFL would defensibly analyse the words capable of as having an adverbial function. Dr Ferguson, in collapsing forms and functions, has therefore attributed capable of to the formal category ‘adverb’. In essence, when form and function are collapsed and functions take priority, then functional categories erroneously become formal categories.

That is how I would explain the ‘utter howlers’ that Huddleston identified; the logical result of using a functional analysis to identify forms, and I believe it also may explain all the other aspects of these guidelines – or at least those that I’ve come across, I haven’t been able to see a copy yet.

I should add that these differences are not mere differences of terminology; the approach suggested by ETAQ gives ‘grammatical unit’ status to strings of words that cross boundaries between phrases, such as, again, capable of, which is not a single unit at all, but a fragment of an adjective phrase.

The issue then, instead of how these guidelines managed to come up with these monstrosities of formal analysis, should be what purpose SFL, or even this corrupted version of it, can serve as a structural analytical tool. My response to that question would be something along the lines of ‘not much’, though I do think it has a reasonable place in education as a textual criticism tool. That is, the focus on interpersonal interpretations, ideational content and context would help students to critically evaluate actual and intended meaning in prose, performances, plays, speeches and the like. However, before such contextual criticism takes place, students must be able to analyse a sentence into its constituent phrases, their interactions with each other, and the individual grammatical and lexical units, words, that form the syntactic basis of a language like English – all using terminology that is at least internally consistent, but consistent with the standard set of terms used by just about everyone else.

I don’t intend any of this to be a defence of, or an apoogy for Systemic Functional Linguistics, that’s another debate altogether, but I do think that as a framework for literary criticism, it is being grossly misused in this educational context, and the students who will not learn to correctly identify parts of speech and structurally analyse sentences because of it, will potentially be at a disadvantage later in their schooling.

SFL cops enough flak already; it shouldn’t have to defend itself here, when the real culprit is Ferguson’s awful corruption of it.


  1. Thanks to Jane Simpson and Mark Harvey for the translation.
  2. Pullum at Language Log, Queensland grammar brouhaha
  3. As an aside, it appears that as linguists, we haven’t yet achieved our tacit aim of convincing the masses that our profession is not about telling them how to write. I fear we have many more years of putting up with misplaced Grammar Nazi over-sensitivity.
  4. I’ve heavily simplified this. Especially when it comes to the analysis of ‘his’, which I’ve just left unanalysed as a determiner phrase (DP). Merely calling it a determiner would be too simplistic.

…and I might just use them when teaching syntax in a few weeks’ time.

Via the Speculative Grammarian.

It seems that Gerard Henderson, former culture warrior, has coined a new euphemism in relation to Australia’s indigenous history. Henderson has always disputed the term stolen generation, because the population of stolen aboriginal people hardly comprised an entire generation, so it’s odd that this new euphemism of his retains this word.

Here’s an excerpt from Henderson’s column in today’s Herald:

Most of the separated generation had European or other genes – in addition to indigenous ones. This means that an abject apology would require that some of today’s indigenous Australians apologise to their indigenous predecessors for the actions of some of their European predecessors.

Leaving aside the fact that this is a spurious conclusion, the choice of the word separated is curious, but is glaringly deliberate – Henderson uses it twice in the same article, and all three instances of the term stolen occur either in direct quotes or, in one case, in the name of a lobby group, the Stolen Generations Alliance.

I think Henderson’s intention here is crystal clear; stolen is such a harsh term; it connotes malevolence when, in his view, clearly no such malevolence existed since it cannot be proved with official bureaucratic documentation. But as the rest of us know, absence of evidence is not evidence of absence.

Separated though, is far too light a term. Couples separate when they’ve had enough of each other. Potassium nitrate separates when it cools. Breasts are separated (and lifted) with the help of the appropriate undergarments.

Separate, the verb, when applied to humans, implies reciprocal volition, mutual agreement, as it were. The forced removal of aboriginal children from their parents in an effort to effect linguistic and cultural homogeneity, involved no such volition on behalf of either parent or child; these families didn’t actively separate, they were forced apart by racist policy.

Moreover, separated, in its unmarked form, is an unergative verb, it necessitates no agent, no one who causes the sepatation to occur. They separated is far more natural than he separated them.

Stolen however, necessitates that someone, an agent, willfully acted to cause the theft. The forced removals that led to the existence of the stolen generation had an agent; the Australian Government. Thus in my opinion, stolen is a perfectly accurate word to describe what happened to aboriginal children in Australia’s Colonial past.

Maybe it’s my less-than-prime cognitive state right now, but I’m beginning to notice little grammatical quirks and ambiguities that I’d normally have overseen (that was silly of me – thanks for pointing it out, David) overlooked completely.

This web page popped up when I opted out of a frankly unsolicited email advertising list:

You have been opted out.

Pardon? Is that an applicativised use of the phrasal verb opt-out? My understanding of this verb is that you opt out of something, you do not get opted out. Then again, if this use doesn’t strike you as odd; if it’s alright to you, to say that someone has opted you out of something, please feel free to digress.

Incidentally, corpus.google.com¹ shows that the strings have opted out and has opted out together generate about 188,400 results, while been and get opted out only generate about 2,000. Be opted out is surprisingly common though, with about 12,500 hits, so maybe it isn’t as ungrammatical as I thought.

The other thing I noticed today was the packaging on a salami from the supermarket, which read:

Ideal for entertaining.
For entertaining recipes, visit our website.

Honestly. Recipes are matter-of-fact, functional things. How entertaining do they have to be?

Seriously though, I was just having a conversation about a very similar thing in the linguistics room on irc.freenode.net. I was previously under the impression that the term operating system is a paraphrase of something like a system that operates, in which case you’d call operating a verb participle, I guess. But since an operating system is actually a system that pertains to operating, it’s accurate enough to call it a gerund.

~

In other news, I just upgraded my wordpress software from 2.3.1 to 2.3.2, because apparently there was a security fault with 2.3.1, and readers were occasionally able to see drafts, which are usually hidden. In fact I noticed a while back that my stats page showed many of my drafts as having been visited, which concerned me slightly. But it should be fixed now, so I can feel free to draft on.

~

¹I’ve mentioned corpus.google.com before, and I’ve been using it now for well over a year. In fact up until an hour ago I though I had originally coined it. But it has come to my attention that there was a blog post that antedates my first use by over 18 months. Still, I certainly came up with it independently, so it’s much like arguing over whether it was Newton or Leibniz who invented calculus.

Here’s the relevant bit:

I wonder if Google will eventually offer such a service themselves? “corpus.google.com”? (Apologies to those who thought this post was actually announcing such a service.)

Predictably, I’ve also variously had to offer up similar apologies to some of my readers who were misled by my reference, such as David.

Earlier on this afternoon, I heard a cricket commentator, having heard about someone whose name he didn’t immediately recall, promise that he’d google him up. This would not be a natural usage for me, although it’s unequivocally clear what he means; it’s completely synonymous with (in my view) the more natural version to google someone, i.e. to search for them on Google.

Anyway, I started wondering how common the construction google up is, so I went and googled it… up, and here’s a breakdown of the returned hits on all permutations:

  google X google X up
him 106,000 7,510
her 92,900 3,490
them 151,000 11,300
it 5,600,000 513,000

Roughly speaking then, the non-phrasal variety google someone is far more common, but the phrasal variety google someone up has some substantial corpus.google.com¹ representation, about a tenth as much as the former.

I didn’t really have anything more to say about it, apart from pointing out that the phrasal verb google up is probably to be expected to occur on the basis of analogy from look up, as in I’ll just go and look them up (in the phonebook). Although curiously, a family member thought it meant to contact via Google rather than to merely find their details, as if on analogy from ring up.

In an age where Google has pretty much usurped phonebooks (of all colours), street directories, atlases, library catalogue cards, encyclopædiae and just about any other source of information, it may as well replace the linguistic idioms associated with them as well.

Holy crap! If someone googled up “Google”, do you think this post would be somewhere near the top?

~

¹No, corpus.google.com doesn’t exist as a URL; it’s just what I use to refer to the act of doing a quick Google search on a phrase using wildcards and quotation marks to back up one’s largely made up postulations about trends in modern English. Think of it as a snowclone, of the x.google.com template.

I just thought I’d point it out, since last time I used the term corpus.google.com, someone (let’s just call him Mr Nash – no wait, that’s too obvious; I’ll call him David N) wondered why they couldn’t find the corpus.google.com homepage.

There’s far too little linguistics on this blog, so in an attempt to rectify this:

Late last week on the bus I was having a conversation with a friend that, after a while, broke off on a tangent about the Roman Empire’s acronym SPQR.

It’s the sort of thing that young Roman men have tattooed on their arms, as if they were imperial Roman Gladiators, or Russell Crowe or something. Mussolini was similarly patriotic about it, as is my understanding, and put it on government buildings and manhole covers across the city.

My friend and I ended up discussing exactly what it meant, and as my friend is of that generation of people who were taught Latin all through high school, I was quite happy to accept that I was utterly wrong.

I had always heard the English gloss as The Senate and the People of Rome, and I thought the original Latin was Senatus Populusque Roma. Apart from that, I knew in the back of my mind that there was something funky going on with that clitic -que.

From these facts I more or less subconsciously concluded that it would parse as:

    [Senatus Populus]-que Roma
    [The Senate and the People] of Rome

which would very easily lend itself to the analysis (from someone who never did any Latin, if I might defend myself here) that the clitic -que was a genitive/possessive morpheme and was bound on the possessed entity, which in this case would have been the entire conjunctive noun phrase the Senate and the People.

However, I was wrong in my basic knowledge of the phrase. I learned that it was actually Senatus Populusque Romanus, and not merely Roma. So clearly then, the three noun roots, senat-, popul- and Roman- all take the same declension -us, meaning that they would be in the same noun phrase, or at least have the same semantic role, in which case a genitive construction would be unlikely.

My friend also told me that the clitic -que was not a possessive morpheme, but a conjunction ‘and’. It could then easily parse as a flat structure, a list of entities, The Senate, the People, and Rome, but this wouldn’t be congruous with the common translation into English, The Senate and the People of Rome.

Defeated, I looked up Wikipedia in the hope that it would have a morpheme-by-morpheme gloss and, while there was no such gloss to be found, there was another piece of the puzzle, an alternative translation. This time it was glossed as The Senate and the Roman People.

If this gloss is more accurate, then Roman People is one half of a conjunction, and The Senate is the other half. If that‘s the case, then why on Earth would the conjunctive -que (which I don’t even know whether to call a clitic anymore) be embedded inside the phrase Populus Romanus, since presumably it conjoins it with Senatus, rather than conjoining Populus with Romanus.

So at the end of the day, I’m not yet entirely sure how SPQR should be analysed, or even how it is best translated, but I’m sure some of my erudite and knowledgeable readers have studied Latin in their time and could shed some light on this…