What counts

Recently, I discovered that there are a lot of English speakers who think that equipment can be countable. This seems to be especially common in India, so you might dismiss it as a dialectal thing, except that various Government of Canada websites use equipments quite prominently, as do Concordia University and a host of other reputable users of standard English.

This came as something of a shock for me as it would never occur to me to use it that way. Even the Cambridge Grammar of the English Language seems to have overlooked this. On page 326 it asserts that "equipment, for example, has no plural counterpart." The OED, on the other hand is more savvy: "Used in the pl. to indicate the articles severally, in the sing. collectively."

So, what's next: shrubberies? furnitures? Apparently so! Surely crockery is safe since we have crocks for the individuals (not to mention the fact that it's hardly a common word.)

When it comes to being uncountable, it really is hard to decide what counts.

T

I was wrestling with my daughter and my nephew when he said, "this place is T," pointing to the sofa. T is short for TO, which, in turn, is short for time out, but neither of these two youngsters seemed to have ever heard time out except in the context of an adult-imposed opportunity for reflection on misbehaviour. The older kid's I've talked to were aware of the source, but the link appears to be fading.

Not at the cost of ~

There has been a long discussion on the ETJ forum (see for example here) about whether or not the distinction between English gerunds and present participles is arbitrary. I've been argue that it is, because gerunds do not in fact consistently act like nouns. This is the position taken in the Cambridge Grammar of the English Language.

Most of the arguments have been based on syntactical properties, but we coincidentally came across a construction that shows another way in which gerunds are unlike nouns. Where one writer had suggested that it would be nice to come to some consensus, I replied "not at the cost of internal inconsistency". The other writer said, "don't you mean at the cost of internal consistency?"

A google search shows that the normal way to express what I wanted to say would indeed have been "not at the cost of internal consistency." In other words where of is followed by a noun, the normal expression is "not at the cost of (good thing)."

• Our readers enjoy interesting words, but not at the cost of readability.
• Individual autonomy is attained but not at the cost of relatedness with close others
• Defence deals not at the cost of transparency
• 6200 ftServer offers solid fault-tolerance--but not at the cost of performance
• Most Pittsburghers want a more civil society, but not at the cost of fairness.
• Water-sharing pact not at the cost of fraternal ties

On the other hand, where of is followed by a gerund, the normal expression is "not at the cost of (bad thing)."

• We wanted to grow, but not at the cost of sacrificing quality.
• Healthcare systems need policies to keep costs down so that we can afford our insurance premiums, but not at the cost of losing our basic rights.
• the respondents said they wanted power -- but not at the cost of losing respect
• benefits of xenotransplantation realized, but not at the cost of compromising public welfare.
• to develop public programmes that would address the interests of their varied audiences, but not at the cost of ignoring their primary audience.

Lack of data vs. Poverty of the stimulus

NewScientist recently published an article about the subconscious that included the following claim:

"Infants do not need tutoring to acquire their native language; they pick it up subconsciously. What's more they do this with remarkably little linguistic data - what the Harvard University linguist Noam Chomsky has called the "poverty of stimulus" - suggesting that this subconscious learning allows youngsters to use information very efficiently."

The nativist language acquisition argument from the "poverty of the stimulus" seems as though it may have been misconstrued. It is not a quantitative argument but a qualitative one. Miller and Chomsky claimed in 1963 that children's input included many grammatical errors and disfluencies. How were children to discover which input to attend to and which input to ignore?

More recent arguments suggest that, errors aside, given the lack of negative evidence, it is hard to see how children could learn what is ungrammatical. We have evidence for children overapplying rules such as regular past tense endings. Given that, it would be logical to assume they would overapply other patterns such as the following:

• She sent a letter to him. / she sent him a letter.
• She explained the letter to him. / *She explained him the letter.

Without evidence that this last construction is not possible, the 'poverty of the stimulus' argument goes, how are children to know that they merely have not yet heard it?

The idea that children simply don't get enough input, however, is odd. Research suggests that average North American children get somewhere in the range of 26 million (plus or minus many million) words of input in their first four years of life. It's hard to see how this can be called "remarkably little linguistic data."

(det) hijab

What with the tragic filicide of Aqsa Parvezes, the word hijab has been much in the news. What I noticed that I had not before is that it is almost always the hijab, as in "she was killed over her refusal to wear the hijab." The OED cites the following instances:

"1980 Associated Press Newswire (Nexis) 15 July, She said the wearing of the hijab, or veil, is a matter of choice. 1985 Times 5 Jan. 8/3 Every woman on the street wears either the traditional chador..or the more practical hijab, a dark scarf pulled over the forehead. 1994 J. I. SMITH in A. Sharma Today's Woman in World Relig. 306 Many Egyptians who do not adopt the higab are deeply respectful of those who do. 1995 New Yorker 30 Jan. 60/3 A young girl..who was dressed in a black djellabah and wearing the traditional head scarf, or hijab. 2005 Asiana Spring 278/1 (heading) Wearing a hijab is no barrier to success."

Notice that only the last one makes the unmarked choice of a instead of the. This suggests that it is a symbol rather than merely a piece of clothing. Other somewhat analogous items include: veil, burka, turban, kirpan, cross, and robes of office.

Unfortunately, none of these is entirely satisfactory. For a nun, taking up the veil, entails more than simply wearing it. To a certain extent, the same could be said for hijab, but once a nun takes up the veil, it doesn't matter whether or not she is actually wearing it at a given time. As this story shows, the same is not true of hijab. Also, burka and hijab should, theoretically, pattern identically, but they don't. Although the OED has examples such as, "1929 Daily Express 15 Jan. 1/1 The Queen [of Afghanistan] is wearing the boorka--a heavy shapeless garment which effectually hides her beauty," current usage seems to favour "a burka". Here are the counts from the Google news archive:

burka hijab the 818 4850 a 1380 2340

With turban, we find 4,540 for the vs. 12,800 for a. And kirpan is almost in a dead heat.

Clearly this data is very muddy, but I think it is at least suggestive that there is something different going on with hijab.

Past tense cut in US schools. Is Canada next?

Underfunded Schools Forced To Cut Past Tense From Language Programs

With American schools cutting the past tense from the language curricula, it can't be long before Conservative provinces in Canada follow suit.

Double modals

A number of people who attended my TESL Ontario presentations on Thursday and Saturday seemed skeptical that modal auxiliary verbs co-occur in southern dialects. Ben Zimmer comes to the rescue with a perfect example, and from a US presidential candidate nonetheless. As far as I know, however, there are no dialects that allow multiple coordinator constructions like "and but" (which was the point I was making in my presentation).

More large-scale testing problems

I've commented before on problems with large-scale test problems at ETS. Now, Sam Dillon, writing in the New York Times reports on a case in which incorrect page numbering on test booklets led to the disqualification of 5,600 reading test scores, an embarrassment that led to a half-million dollar penalty for RTI International, the company that handled the tests.

The grammar of physical exhaustion

I wonder if Shatner ever felt that he was hyphenating his words. Likely this is more of a cartoonist's thing.

And the worms ate into his brain

In his new book, Musicophilia, Oliver Sacks seems to have created a new word for those sticky songs that you can't get out of your head. In German, these things are called ohrwurms, their word for earwigs. This has come into English as earworm. Now Sacks has decided that he prefers brainworm. You can watch a video of him talking about it on the book's page on Amazon.com.

John Terauds, writing in the Toronto Star on the weekend, used Sacks's new word, but now says via e-mail that it was a brain fart.

Smaller than small

The ETJ list is often a source of interesting questions. Recently, Peter Warner observed:

"Listening to a British presenter, I realized that his sense of little and small were different. His general order of descending relative size went roughly in this sequence:

huge, big, small, little, tiny

while my American background views little and small as basically synonymous. Speaking with him after his wonderful talk confirmed his different sense of those two words."

Peter then asks, "is this a British vs. North American English usage difference?"

If you check the British ESL dictionaries, they typically define little as meaning small and the other way round. Similarly, the Concise Oxford defines little as "small in amount, size, or degree..." There is no mention of one being less big than the other.

There are, however, other differences. Least interesting among these is that small is only an adjective (the small of your back excepted), while little exists in two flavours: adjective & determiner. More interesting is that they are both marked, but little seems to be more marked. That is, when I want to know the size of something, it's typical that I would ask you how big something is, not how small it is. In this sense, big is unmarked. Yet, it is even less common to ask how little something is, about 10 times less common in the frame "how ~ is it?'.

Because of the problems parsers have distinguishing between adjectives and determiners, it's hard to get a good corpus view of all aspects of the difference, but here's a neat one: Search for (adj) + little in the spoken section of the BNC and you get the following (the first number is raw hits, the second number is hits per million words):

1 NICE LITTLE 161 15.58
2 TINY LITTLE 36 3.48
3 LOVELY LITTLE 35 3.39
4 POOR LITTLE 33 3.19
5 OTHER LITTLE 14 1.35
6 LITTLE LITTLE 13 1.26
7 SILLY LITTLE 13 1.26
8 CHEEKY LITTLE 11 1.06
9 PRETTY LITTLE 11 1.06

Try it again in the academic section and what do you get? Nothing. There are no such pairs.

Now do the same with small, this time with academic section first:

1 PROXIMAL SMALL 28 1.81
2 OTHER SMALL 20 1.30
3 DISTAL SMALL 6 0.39
4 NORMAL SMALL 6 0.39
5 UPPER SMALL 6 0.39
6 ENTIRE SMALL 4 0.26
7 INDIVIDUAL SMALL 4 0.26
8 NUMEROUS SMALL 4 0.26

And in conversation? Zip.

So it would seems that small is more academic, at least in this kind of pairing.

Now, this is stretching things quite a bit, but if we consider the general prejudice that conversation is trivial while academic language is weighty, long, and substantial, there may be an argument, if only a metaphorical one, for little being smaller than small. I think I'll put it to Lynneguist over at Separated by a Common Language.

Problematising Problematic

About a month ago, Russel Smith, writing in the Globe & Mail, discussed a tendency for the meaning of technical terms, such as price point and deconstruct, to drift as the words become mainstream.

"It's inevitable that technical words and phrases will be imported into everyday language from specialized jargons. It's also inevitable that those terms will then change their meanings slightly. They usually lose some of their specificity, a bit of their subtlety, and become synonymous with some other everyday term."

I think he overstates the case here, but still, there is a case to be made. And notice how the paragraph above is simply descriptive, rather than evaluative. Perhaps his time at And Sometimes Y has relieved Smith of some of his more prescriptive tendencies, though, making it clear that he hasn't entirely shuffled off the curmudgeon's burden, he writes, "My favourite example of a corrupted technical term has to be 'deconstruct'." He explains the Derrida's use of the word like this:

"The aim of the reading was to show how the text's meaning is elusive, how it contradicts itself. It's part of a larger view of language as something essentially problematic. In other words, it hardly means an elucidation, as we use it so casually to mean now, but almost the opposite"

Still, Smith remains admirably neutral throughout. And the changes that he describes actually seem to be supported by evidence.

Ironic, then, that he is taken to task in a letter that appeared a few days later:

Posted on 08/09/07

Problematic problem

PETER FERGUSON

Kimberley, Ont. -- It is striking that in discussing the demotic use of jargon (Technical Terms And Mainstream Meanings - Review, Sept. 6), Russell Smith employs "problematic" in just such a manner. Having originated in the field of logic, the word means doubtful or questionable and has only recently come to be used in describing something that poses a problem. That Mr. Smith is, surprisingly for him, unaware of this history may be a problem, but it is definitely not problematic.

Here, Ferguson indulges in a number of fallacies (but they're so fine, you see):

1. language must not change, so the older meaning is the only true meaning

2. language always moves from the erudite to the debauched, so the technical meaning must be older

3. whatever I think a word means is what it means
   

If we consult the OED, we find that it lists more than one meaning for problematic. Yes, polysemy is alive and well. We also find that the term of art from logic dates from 1610, but the more general sense is attested from 1609. From this, it would be hard to argue that one preceded the other in English. Finally, the OED disagrees with Ferguson's definition, giving the following instead:

2. Logic. Of a proposition: that asserts that a state of affairs is possible rather than actual or necessary.

If they will only listen

Believe me, English really doesn't have a future tense. Some may think this is simply a labeling preference, but it goes beyond that. The belief that English has a future tense gets people all twisted up in ad hoc explanatory knots that vanish once you accept a future-tenseless English into your heart.

Take, for instance, this question, which recently showed up on the ETJ list.

I was asked if the sentence "if he's not going to change his mind
there is no use talking to him" was correct.
I said yes.
"But you can't use if with the future tense" came the reply
"quite right" said I.
"so why is this ok?"

In case you're asking where these two came up with this "rule", here are a couple of versions for you. The first one is from an TESL site called englishpage.com.

Like all future forms, the Simple Future cannot be used in clauses beginning with time expressions such as: when, while, before, after, by the time, as soon as, if, unless, etc. Instead of Simple Future, Simple Present is used.

And here's another from Wikipedia (now changed):

Future tense forms are not used in the condition clause (protasis) in English: *If it will rain this afternoon, …

If you subscribe to the future-tense school of English grammar, it is completely natural to look for regularities in how such a tense would be realised. This is what we're seeing above. The people who are promulgating these "rules" are looking at a specific instance in which it seems to be true (e.g., *if it will rain this afternoon) and generalising from there.

But this disallows perfectly fine sentences such as "if you'll excuse me, I've got a bus to catch" or even "If you'll be using a wheelbarrow frequently, then make risers low". It also catches instances such as the one that initiated the question above.

On the other hand, it ignores other problems such as "it could rain tonight" becoming "if it could rain tonight...", or "that may be the right one" becoming "if that may be the right one", or even "he might have been there" as "if he might have been there".

A belief in the future tense does this because it leads people to look for regularities in some hodge podge notion of the future tense where there are none to be found rather than within the system of modal auxiliaries where they actually exist.

In fact, the problem only resides with certain modals (mainly 'will', 'may', 'might', and 'could') when they are used to express probability (i.e., epistemic uses). Thus, "it could rain tomorrow" doesn't work as *"if it could rain tomorrow", but "you could help me tomorrow, couldn't you" easily becomes "if you could help me tomorrow" because 'could' here is denoting ability (or willingness, i.e., it is deontic) rather than probability. Note that this also holds true for present and past time as well as future time.

Rethinking rethink

A student (college-level native speaker of English) writes, "It is time to rethink the path society is on." Something in my brain went sproing when I read this. Well, not really sproing, more like twing. Anyhow, when I went over it again, path just didn't feel quite right.

I ran it by my family who were enjoying the weekend while I wade through this pile of papers. My brother found it unremarkable, but my mother immediately rephrased it as "rethink where society is going". It then occurred to me that rethink seems to have a different set of complements from think. After some googling and corpus searches, I'm confident that this is so.

You'd never say, "*think your plans/policy/approach", but "rethink your plans/policy/approach" is fine. It seems that, for whatever reason, rethink now patterns with reconsider rather than think. I wonder when that happened or if it has been thus since it came into the language around 1700. I wish I had access to the OED from home. [Now that I'm back at work with access to the OED, I find that rethink has been transitive from its earliest recored use.]

CBC blowin' in the Burmese wind

CBC started the week reporting on the marches in "Myanmar, formerly known as Burma". For the last two days, however, it's been "Burma, also known as Myanmar".

[Actually, a quick web search shows that it's a policy change.]

Promises are not verbs

In today's Toronto Star, Andrew Chung writes,

"Promises are the currency of elections. With no platform to judge, on what other basis could voters make a decision when casting a ballot? Promises are also part of a category of verbs that experts call 'performative speech acts.' These utterances actually cause the speaker to perform a certain act."

I'm pleasantly surprised that the topic of performative speech acts (PSAs) should show up in The Star. They're rather curious self-fulfilling things. When you say, "I promise x", you have done so and need do no more. The act is carried out through the utterance. Promise isn't the only word that works this way. Other examples are declare, sentence, damn, pronounce, etc.

But notice that though this is a list of verbs, the verbs themselves do not, contrary to what Chung says, constitute speech acts. It is the utterance, usually in the form of a sentence, that is the PSA. Nor are promises verbs. But I do appreciate Chung's giving it the old college try.

An ESL policy update from Annie Kidder

Today, Annie Kidder at People for Education posted a comment to a previous post. It's substantial enough that I'm reposting it here.

In early September, the province quietly posted its new English as a Second Language (ESL/ELD) policy on the Ministry of Education website. The province has been promising new funding and policy for ESL since the spring of 2006, and while the new policy addresses some of the concerns raised over the last three years by Ontario’s Auditor General and others, it is difficult to see how it will alleviate many of the ongoing issues in ESL programs.

• Many school boards report using a substantial portion of their ESL funding to cover the costs of things like heat, light and building maintenance. The new policy does not protect ESL funds.

• Students’ ESL support is reduced, or eliminated altogether, when the funding runs out, as opposed to when the student has sufficient English skills to function academically. The new policy says this should not happen, but does not commit to supplying funding for students who are not ready to be withdrawn from ESL programs.

• There is no measurable English-proficiency standard that ESL students should attain before ESL services are discontinued. No standard was set.

• The Ministry does not ensure that the ESL/ELD funding targets students most in need of assistance. The policy suggests that boards use the funds where it will be most effective – but there is no specific direction given.

• Funding for ESL does not differentiate between students who arrive in Ontario as refugees and have little or no formal education in their first language, and students who have attended school in their home countries. The policy notes that there are different levels of need among ESL/ELD students, but does not provide for differentiated funding.

• According to the Ministry of Education, students usually take from five to seven years to become fluent in English. But funding for ESL/ELD support runs out after four years. Funding has not been extended beyond the four years.

• There are no minimum ESL/ELD training requirements for regular classroom teachers who have significant number of ESL students. The policy suggests it would be beneficial for new teachers to acquire ESL skills, but does not require it.

Over 600,000 foreign immigrants moved to Ontario between 2001 and 2006, most from non-English-speaking countries and many with school-age children. As a result, the proportion of ESL students in schools increased by 24%. Over the same period, the percentage of elementary schools with ESL teachers declined by 23%.

In schools with higher ESL populations (more than 10 ESL students) the percentage reporting ESL students but no ESL teacher has doubled since 2000.

The new policy does not require school boards to spend ESL money on ESL programs, it does not set standards for an acceptable level of English proficiency and it does not provide funding that recognizes the difference between refugee students needing substantial literacy support and ESL students who have strong literacy skills and only require support to learn the language.

To read the policy go to the Ministry of Education website.

Pirate speak in other languages

Since the good folks at Language Log reminded me that today is Talk-Like-A-Pirate Day, I asked my students if pirates have any particular linguistic tics in their languages.

• A German student came up with a sentence terminal ay! (read /Ei/). There's also this glossary of German pirate words.
  
• The Spanish speakers said Spanish pirates all have Castilian accents.
• Korean, Japanese, Iranian, Iraqi, Arabic, and Turkish students couldn't identify any particular pirate dialect.
  

Rabbit numbers

One of the readings included in our Comm 200 packet for students to respond to is "Dot-com this!" by Stephanie Nolen from The Globe and Mail, Aug 28, 2000. pg. R.1. It includes the following:

"Eric McLuhan, author of Electric Language, adding that English, as the language with the greatest flexibility and largest vocabulary, was the only language prepared for this shift.

But McLuhan, who is the son of the legendary communications theorist Marshall McLuhan, says the 15 years of the computing era have had drastic effects on the building blocks of writing.

Attention spans have declined sharply, and with them, sentence length. Twenty years ago, the average sentence length in a novel was 20 words; today it is 12 to 14 words. In mass-market books such as Harlequin Romance novels, the average sentence is only seven or eight words.

The stuff about English having the greatest flexibility and largest vocabulary is not even worth commenting on, but it seems pretty clear that McLuhan is just making up the sentence lengths too. Just in case, I did a quick tally to see how accurate his claims are. I took the top 10 selling books from 1985 and 2005 (from here) and pulled up the words-per-sentence stats from Amazon.com (e.g., here). Where they weren't available, I took a book from the same author published around the same time. Here are the results.

1985 WPS The Mammoth Hunters by Jean M. Auel 13.8 Texas (All We Did Was Fly to the Moon) by James A. Michener 14.6 Lake Wobegon Days by Garrison Keillor 16.8 If Tomorrow Comes by Sidney Sheldon 10 Skeleton Crew by Stephen King 12.5 Secrets by Danielle Steel 12.7 Contact by Carl Sagan 14.3 Lucky by Jackie Collins 8.7 Family Album by Danielle Steel 14 Jubal Sackett (Lando) by Louis L'Amour 14.2 average 13.16

2005 WPS The Broker by John Grisham 11.5 The Da Vinci Code by Dan Brown 11 Mary, Mary (actually Cat and Mouse) by James Patterson 9.6 At First Sight by Nicholas Sparks 11.6 Predator by Patricia Cornwell 11.1 True Believer by Nicholas Sparks 12 Light from Heaven by Jan Karon 9.9 The Historian by Elizabeth Kostova 19 The Mermaid Chair by Sue Monk Kidd 13.5 Eleven on Top (Lean Mean Thirteen) by Janet Evanovich 9.1 average 11.83

In this small unscientific study, there is a tendency for sentences to be shorter in 2005 but the difference of 1.33 WPS is nothing like the 10 WPS McLuhan is claiming. And when the largest average for a single book from 1985 is 16.8 WPS, it seems highly unlikely that you're going to find an average of 20 WPS for all novels published in that year. Maybe what computers have made us better at is pulling numbers out of a hat as if they were rabbits. Then again, maybe better is the wrong word.

Hofstadter on Pinker

Over the weekend The LA Times published a review by Douglas Hofstadter of Steven Pinker's new book The Stuff of Thought (The Times has put this behind a pay wall. You can still read it here.) Hoffstadter writes, appropriately, with the caution and insight of a scientist, disagreeing graciously where he sees fit. "Pinker exploits his wonderfully keen faculty for linguistic observation to pry open the human head and discover its secrets. Sometimes this technique works terrifically, other times not so well."