Monday, 15 July 2019

Fusilli bolognese

I had the good fortune to be on a train full of children, maybe about nine or ten years old, the other day. I mean, literally full, there were two of them in the seat next to me.

They were off on a residential visit somewhere so meals were involved, so they started talking about their various fussy eating foibles. One said she couldn't eat steak, to which the teacher responded with admirable restraint 'It's OK, we're not having steak'.

Another was alarmed when she heard one of the meals would be spaghetti bolognese (or 'skabetti', as the teacher called it, which I always find an adorable pronunciation), because she doesn't like spaghetti. Fortunately for her, it turned out that the spaghetti to be used was in fact fusilli (the spiral one), which she did like. For me, spaghetti is specifically the long, thin, solid, cylindrical ones, and any other pasta shape has to be called by its own name (or just 'pasta'). But spag bol is special, because it's the pasta meal most British people encountered first, so it's kind of a meal in itself. I'm not quite sure how to describe its cultural role, actually, for people that don't already have the same cultural knowledge of it. I mean, Heinz sells it in tins, is probably the most meaningful thing I can tell you about it. So I suppose for her, spaghetti isn't a thing in its own right and it could be used as the generic term for pasta as long as it's with that particular sauce.
Image result for spaghetti bolognese tin
Tin of Heinz spaghetti bolognese

Pasta words are one of those semantic fields where there's a lot of overlap and variation, by the way - 'noodles' for me has to be in an Asian dish, like chow mein, so even though they're long and thin, I couldn't call spaghetti noodles. For some people 'noodles' covers all kinds of pasta-type foods. 

Thursday, 11 July 2019

Yo, Samity Sam!

Brilliantly, the tweet I am going to be famous for (at 200+ likes at the time of writing), is one in which I present myself as an idiot.
Lane Greene said that American kids learnt how to pronounce 'Yosemite' because of the Looney Tunes cartoon character 'Yosemite Sam'. Despite plenty of exposure to this character, I did not similarly acquire the correct pronunciation of the word, because I interpreted the name as 'Sammity Sam', a sort of reduplicated nickname form of the name Sam. This is not unreasonable, as I'd not seen it written down and never heard of the place Yosemite. Other people did the same thing, and basically heard it as 'Yo, Sammity Sam'.

But here's the thing: I didn't interpret that first syllable as anything at all. 'Yo' wasn't specially in my vocabulary as a child in 1980s UK. So I just kind of heard it as a meaningless syllable. I'm not an acquisitionist so I could be wrong here, but I'm pretty sure this is NOT how language works. There are not meaningless syllables that consistently occur adjacent to a specific word. I should have definitely called on my pattern recognition skills and interpreted that sound as part of the name, or as being part of the phrase, as others did ('Yo!'). But I didn't.

Wednesday, 19 June 2019


Hooray! Another article about how 'textspeak' is bad for kids is out! (Daily Mail link, so don't click it – you can get the idea from just reading this post.)

It's really a shame that the experts they asked were not experts in the thing they asked them about. They're experts in children's potential and curriculum development, both important things, but not actually language, which is the thing we're concerned about the harm of here. It seems comparable to asking experts in primary education if mobile phone masts are harming children's concentration or something. They'll have relevant things to say about concentration but they won't actually have the expertise to say if it's the mobile phone masts that are the problem.


In the image above, Prof. Mellanby says 'these sentences do not contain grammar', of the following:
OMG ikr
Yo dude r u still coming to party Friday
I'm just going to take the second one. It contains, among other things, the following grammar:

  • a vocative (Yo dude)
  • subject-verb agreement (r for 2nd person singular)
  • question inversion (r u rather than u r)
  • a verb phrase with a prepositional complement (to party) and adverbial (Friday)
  • present tense (r rather than were)
  • progressive aspect (coming)
  • pre-verbal adverbial (still)
  • knowledge of which words can be omitted in this context (the, on)
It also has 100% correct spelling, if one allows that r u is an abbreviation rather than an error.

Tuesday, 11 June 2019

Southern privilege

One of my favourite writers is Justin Myers, who among other things writes The Guyliner. He reviews the Guardian Blind Date every week and I look forward to it eagerly. It's funny, but it's also clever and insightful and often poignant. This week, he threw in a comment about 'doing funny accents' that was so spot-on in its identification of the problem of centring privilege.

Just to take a step back: we normally think about male privilege, white privilege, or straight privilege. It doesn't meant that (say) men don't have problems, or never face discrimination, but that they don't face the kind of systematic discrimination that those who aren't men face, while of course they may face systematic discrimination of other kinds (e.g. you might be a gay man, or a black man, and have male privilege but not straight or white privilege). In the grand scheme of things, northerners aren't who you think of as facing the worst discrimination, but nevertheless, there is a sort of 'southern privilege'. (North and south here refer to England, by the way - in itself this minimises the existence of the other UK countries and especially Scotland.) This shows up mostly in accent discrimination, which can be a proxy for class discrimination.

Justin talks about the way that people frequently imitate a 'northern' accent when he tells them that he is from Yorkshire. (UGH by the way - 'the north' is a big place with a lot of different accents.) He specifically mentions the way that they say 'oop north', and the way they think this must pronounced like 'poop' because they don't realise it's just the vowel like in 'book' but written with a double 'o' to emphasise the difference from the southern pronunciation. And here comes the part that I'd never even thought about before, which is that having this special spelling to indicate the northern pronunciation is in itself a staggeringly southern-centric way of doing things. As Justin points out, there is no northern equivalent of an approximation of the southern pronunciation (he writes it as 'ap', which is pretty close to the IPA for the RP pronunciation). The word 'up' spells both the southern and the northern pronunciation; the letter 'u' represents both the sounds /ʌ/ and /ʊ/. To write 'up' as 'oop' leaves it as representing only the southern pronunciation, ignoring the northern one altogether.

In this situation, as always, the people in the position of relative power fix the language in a way that positions the less powerful ones as 'others', not the norm.

Friday, 17 May 2019

Unexpected adverbials

Liliane Haegeman has been here at Kent all week on an Erasmus teaching mobility visit. She's been giving four seminars on adverbial clauses, including among other things the difference between what she calls 'central' and 'peripheral' adverbial clauses. Central adverbials modify the event itself, so in this case the 'while' clause tells you the time during which the main event happened:
I took out the rubbish [while you were watching telly] 
Peripheral adverbials, meanwhile, don't tell us about event time or anything like that, but are more about the relation between the clauses or speaker attitude. In this case, the 'while' clause gives a contrastive or concessive meaning: 
I’m quite active, [while he is a total slob] 
These clauses have various properties that distinguish them. One of these properties is the tense of the verb. Central adverbial clauses (e.g. expressing time) have the same tense as the main verb, or if they don't, they're interpreted as doing so. Here, there is a future marker 'will' in the main clause, and the 'while' clause has present tense 'watch', but it's interpreted as happening at the same time as the main clause.
I'll take the rubbish out [while you watch telly]
A peripheral (e.g. contrastive) adverbial with different tenses is interpreted as being two different times:
I was fortunate to get full funding for my degree, [while he has to borrow a student loan]
The 'full funding' happened in the past (with past tense 'was') and the 'borrowing' is now (present tense 'has').

All of the above is paraphrasing what Liliane told us in the first seminar. She also said this this is something that you specifically have to tell learners of English, that they have to use present tense in this kind of clause. 'If' clauses are the same, in that they have two meanings:
If you don't understand this part, you won't be able to follow tomorrow's seminar. [if = condition]
If you didn't understand, why didn't you raise your hand and ask? [if = assumed background]
The conditional one is the one where the tense should be present even if the main clause is future, as it is here. Then, at the end of the seminar, Liliane was talking about the following one the next day, and she said this:
Even if you won't come back to the class, you have the handout. 
For this to be a normal conditional, it needs to be present tense 'Even if you don't come back'. But Liliane used the future tense marker 'won't', and then all of a sudden it was forced into the assumed background interpretation: 'Even though I understand it's the case that you won't come back...' where the implication was that she knew that we wouldn't return, and so she had given us the handout, when in fact the meaning was that she thought we would but if we didn't, we had the handout.

Another example of the same thing is a little more complicated because you need to know about Dutch word order. Dutch has 'V2', which means that the finite verb is the second constituent in the clause. So it can look like English, where you have the subject, then the verb, then the object, or it can be some other part of the sentence before the verb, like the object or an adverbial clause like the ones we just talked about. Now here's the thing: only the central adverbials can be this first element before the verb. If it's the other kind, then it doesn't 'count' and you need something else to be there, like the subject. Look at this, where the verb is in bold (example from the seminar handout):
Dutch: [Als het je interesseert,] er zal in Parijs ook een vacature zijn.
Word-for-word: If it you interests, there will in Paris also a vacancy be.
Idiomatically: 'If you are interested, in Paris there is also a vacancy.'
The 'if' clause is the kind that gives you background info, that doesn't count, so you have something else (the subject 'there') before the verb 'zal', which is now the third constituent. If you swap the order and have 'Als... zal er...', namely 'if... will there' where the 'if' clause is the one element before the verb in normal second position, it magically gets forced into the interpretation where the 'if' clause is a real conditional - the vacancy only exists if you're interested in it!

(NB: I've massively over-simplified this, and much of the week was spent learning how lots of this has interesting exceptions, and I've conflated two types of clause, etc etc. I've also paraphrased Liliane's work to write this, so consider this a citation.) 

Monday, 13 May 2019

Ladies of science

The suffix -ist is added to a base to make a word that means 'someone who does something related to...', so an exorcist exorcises. It can mean someone who holds some belief system, so a Darwinist follows Darwin. And it can mean someone who practises some art or profession, so an etymologist studies etymology.

-ist is also helpfully gender-neutral, so it's never run into the same problems as -er/-(e)ress pairings like waiter/waitress, actor/actress. I learnt the other day that it was in fact invented specifically to refer to a woman, Mary Somerville, joint first female member of the Royal Academy, because she couldn't be called a 'man of science'.

Caption from exhibit at Turner Contemporary, saying that 'scientist' was first used to describe Mary Somerville

I'm pleased they went with this and not 'lady of science' or something.

Vaguely relatedly, there's a really nice essay by Laurie Bauer on why linguists are not called linguisticians, with some interesting insights on the connotations of each suffix (-ician was 'trivialising' at the time the words were being coined).

Tuesday, 7 May 2019

Toilet flushing instructions and recursive binary Merge

Spotted in the Duke of Cumberland loos in Whitstable this weekend, this instruction on how to flush the toilet:

Press hard
Both buttons
I interpret this as two instructions:
You could read it as a single instruction ('Press hard both buttons') but it's awkward in English. It would be normal in, say, Spanish, but in English a more idiomatic word order is 'Press both buttons hard', or verb (press) – object (both buttons) – adverb (hard).

The way that this is broken down into two phrases can be seen as support for the idea that syntax comprises a series of operations of recursive binary Merge. That's a technical way of saying that sentences are formed by combining two elements at a time, and combining the resulting component with a new element, still two at a time.

So, for example, we might think of a sentence like Birds eat seeds as being formed as follows:
eat + seeds --> eat seeds
birds + eat seeds --> birds eat seeds
Our loo-flushing example is a little bit more complicated. We don't have a subject, because it's an instruction so there is an implicit 'you' as the presser of the buttons. We definitely want to say that both buttons is a unit (a 'constituent'), which seems intuitively right (there are also ways to test this kind of thing). Then we want to say that Press both buttons is a constituent, with the verb press combining with its object both buttons. Then we would combine that whole phrase press both buttons with the adverb hard, telling us how the action of pressing both buttons should be performed. This makes more sense than saying that the verb press combines with a constituent both buttons hard, which doesn't seem intuitively right. Adverbs tell us how verbs are done, not what nouns are like. So now we have this structure:
both + buttons --> both buttons
press + both buttons --> press both buttons
press both buttons + hard --> press both buttons hard
The fact that the adverb refers to the verb, and not to the noun, also tells us why we get the broken-up instructions in the photo. The adverb, as I said, refers to the verb. We interpret it as referring to the whole verb phrase press both buttons as the thing that has to be done in a hard manner, but in fact it's really the pressing that is to be hard. The whole phrase involves the three levels of recursive Merge (recombining constituents) shown above, giving a final nested structure like this:
[3 [2 press [1 both buttons 1] 2] hard 3]
If we want to make it much simpler, one way of doing that is to remove the recursive part of the operation, and have things combine just once. This means, if you assume that Merge is binary (that things can only combine two at a time and not three or more), that the maximum number of words you can have in an utterance is two. And that's exactly what is happening in the photo: two pairs of words (both buttons; press hard), and their juxtaposition is what tells us that one applies to the other rather than their syntactic combination.