There’s a microblogging service that is allowed by the Chinese Government, who have armies of censors making sure it doesn’t allow the same freedom of expression as Twitter. More on this can be read in the recent Economists Article.

But what’s really great is running it through Google Translator. Google Translate works by looking at statistical correlations, using massive bodies of text. This means that it often gets the right meanings, but the language is always strangely garbled. In the case of microblogging, this equates to something like raw, unfinished poetry. Here are some examples of what I mean.

  • Household mold: Hungry dizzy. . .
  • Late Anna Na: In fact, one is quite good. cold go out into the yard to hide tight spots. late Anna Na.
  • RONALDOlp: I am sad, but you hear me
  • Yangfear: I want to be able to wear sunglasses a winter city. . . But for the damn paper, I can go to LA next week, 4 days the sun drying. . .
  • wisney: Grass your mother actually has a girlfriend … how do you do … do not die you horse egg roll …

Try if for yourself!

Posted in Linguistics Musing | Leave a comment


Anyone who reads Language Log regularly probably has noted that Mark Liberman repeatedly references the song ‘Oxford Comma’ by Vampire Weekend. A few days ago I found out that another famous indie band (which I like), Yeasayer, have a song called ‘Mondegreen.’ A mondegreen is a misheard lyric: an example being something like “You can’t take this guy from me” instead of “You can’t take the sky from me” (In the Firefly theme). Somewhat fittingly, no site that I can find has a full list of the lyrics that doesn’t include at least a few ‘?????’. See what you can make of it.

The singer was never the best at enunciating. I sadly couldn’t identify any specific mondegreens while I was listening – if I had, they probably wouldn’t have been proper mondegreens anyway.

Posted in Linguistics Musing | Leave a comment

Case ‘n Point

I  identified this lovely little eggcorn a little while ago. In case you didn’t already know, an eggcorn is an idiosyncratic substitution of a word or phrase for a word or words that sound similar or identical in the speaker’s dialect. The word was coined by our very own GK Pullum, who heard of an American woman who thought that ‘acorns’ were ‘eggcorns.’ (I can’t really tell the difference, either.) For me, a speaker of American English, ‘and’ often reduces to a syllabic ‘n’, while ‘in’ does the same in certain environments. Because of this, there is no distinction between ‘case in point’ and ‘case and point’. I’ve never really known which one is the correct usage. ‘Case in Point’ has 2,600,000 hits on google. ‘Case and point’ has 3,590,000. So, both are in pretty rampant use. Are they both acceptable? I am guessing yes, as I could understand either, and I use both (since there’s no difference.) Is anyone here British: if so, which do you use? And which one was there first? I have no idea. But that’s what eggcorns do – they reflect a lack of a priori knowledge about something. And this is just another …wait for it… case’n’point. (You saw that coming, didn’t you? I know: I should’ve tried to use QED! Oh well. Too late now.)

Posted in Linguistics Musing | Leave a comment

The Ideophonic Glede

This morning I picked up The Ring of Words again, as I mark the pages and stop reading whenever I get an idea for something that should go on here. I had stopped at the word ‘gleed,’ which is an obsolete term for an ember or coal. ‘Gleed’ is another word, now sadly ignored as it has passed out of use, which has the gl- segment which marks various words to do with sight or light: glow, glimmer, glitter, gleam, glint  – there are plenty more. These sorts of words are typically known as ideophones. While I was researching around the web for a way to justify my folk etymology for these words – which I had hypothesized had evolved from the older word ‘gleed’ (which isn’t true – *gl words like this go back at least to ON, and probably back to PIE) – I came across a fairly good blog on ideophones.  Since that blog contains more information than my ramblings about the possibility of cultural imitation or derivation as the evolutionary source for ideophones could possible hold, I’m just going to refer you there. Definitely worth at least a cursory look.

Posted in Linguistics Musing | Leave a comment


From time to time, I find myself desiring to read books which are not related to linguistics in any possible way. Mostly, this means that I want to read science fiction or fantasy books. So, yesterday, instead of working like the diligent student I am, I sat down and read Ender’s Shadow by Orson Scott Card. It was a fascinating novel about Bean, a secondary character in the main book which preceded it, Ender’s Game. The premise of the story is that there is an alien invasion, and Earth has united to fight it by sending their best and most brilliant children to Battle School in a satellite circling the earth, to train them as admirals to fight in the coming war. I know, it sounds awesome, right? In reality, it is, because it’s not a novel about children – it’s about strategy. And the use of strategical theory is brilliant in both books. So much so, that when I found at the end of Ender’s Shadow a short page about influences and thanks, I immediately went to the library to find a collection of essays called Makers of Modern Strategy: from Macchiavelli to the Nuclear Age. I justified this distraction by claiming that as a linguistics student, I must improve myself in order to get into an MSc program, and ultimately, into journals and a tenure position. How better to improve myself than to analyse the lives of greater men who had been through things which probably had analogues to my own life? (This justification is of course, ridiculous, and therefore worked incredibly well.)

You may remember that I posted about the origin of Nostalgia back in the spring. Well, a similar thing occured which brought linguistics back into my mind. While I was reading about how Sébastian Le Prestre de Vauban, a famous 17th century French strategian, completely innovated the way sieges were done by using advancing levels of trenches, I came across the following sentence:

Saint-Simon, not content with dubbing [Vauban] a Roman, applied to him, for the first time with its modern meaning, the word patriote. [pg 74]

And there you have it. The first use of the word patriote with it’s modern meaning. Of course, the OED disagrees, and traces itback, in English no less, to 1577, a full 125 years before this usage. But that’s not really the problem: the problem is that I managed to find something that reminded me of linguistics, and so I felt obliged to work on my essay at 1:30am last night. And I didn’t pick up any strategical analogues to writing it, but one: don’t delay in doing something. Maybe I’ll do better next time.

Posted in Linguistics Musing | Leave a comment

Ken More

Ken More – Graffiti on Old College

I was walking by Old College yesterday when I saw this graffiti on the side. I didn’t make much of it, but my flatmate pointed it out and gave a perfunctory ‘heh.’ “Know more,” he said, in his upper-class Scottish accent. And so I started thinking. It’s possible that this was more than just standard idiotic gang graffiti (if anyone else gets annoyed at seeing ‘œ’ all over town, you’re not alone.) It might be a statement about the Scottish-English dialect. ‘Ken’ is a notoriously Scottish word, meaning ‘know’. So, when one translates this statement into Standard English, there’s a pun: No more & Know more. The homophones would probably be glossed over by most, to be honest – I suspect that the pun wasn’t the original intention of the author . But if they were intended, I think this is a pretty witty way to say that the Scottish dialect, or culture, or what-have-you, is being forgotten (hence the imperative) and must be either revived or relearned or studied. The statement is about as vague as my argument for it, which is a bit of a problem, so I’ll stop trying to explain it.

It’s actually a pretty clear cut translated pun, I think. Another good example of a pun like this would have been used by the Cyclops in the Odyssey. Odysseus, not wanting to be known by his real name, gave the name ‘ou tis’, meaning ‘no one’. However, another way to say this would be to use the other negative: ‘metis’. This is a pretty witty pun, as that is a homophone with Metis, the god of cunning. So, when the cyclops has lost his eye and is calling for his kinsmen, he’d be saying in effect that both ‘no one’ and ‘cunning’ caused the issue. Quite…well…cunning. (The actual text may not use this form, but it was certainly known to the Greeks. Or, at least, to the Greek teacher at Edinburgh.)

Of course, another possible reading of the graffiti and it’s polysemic meaning comes from where it was drawn: Old College. Given the cuts that are being proposed to higher education, and the march today at 11:15 from East Market Street, I find the building it was placed to be a very suitable choice, really.

Posted in Linguistics Musing | 3 Comments

Rosetta Stone 300

The Rosetta Stone company has a new project out. Here’s the details, copied verbatim from the Linguist List email.

If you have 10-30 minutes and a keyboard or a microphone, please
consider making a submission to The Rosetta Project’s latest
volunteer-based linguistic documentation project:

The 300 Languages Project is a special effort by The Rosetta Project
(, part of The Long Now Foundation
(, to begin the construction of a universal corpus of
human language by collecting parallel text and audio in the world’s 300
most widely-spoken languages. The resulting collection will contain
thousands of volunteer-contributed public domain text documents and
audio recordings which will be made available to researchers and the
public alike via The Internet Archive, a free online digital library.

The 300 Languages Project seeks to develop an extensible protocol
and a set of scalable, low-cost (i.e., volunteer-based) methods and
standards for language documentation via the building of a “seed
corpus” – a corpus which starts small but is designed to grow.

The 300 Languages Project is collecting translations and recordings of
three important texts: the Swadesh List, the Universal Declaration of
Human Rights, and Genesis chapters 1-3. These texts were chosen
primarily for usefulness in research (e.g., the Swadesh list) and
breadth of existing translation (Genesis and the UDHR).

The 300 Languages Project is made possible through the support and
sponsorship of Distinguished Career Professor and speech technology
expert Dr. James K. Baker and is conducted in partnership with the
ALLOW initiative of the Center for Innovations in Speech and
Language at theLanguage Technologies Institute.

Looks quite cool, if I do say so myself.

Posted in Linguistics Musing | Leave a comment

“I don’t know the meaning of the word!”

We’ve probably all heard them before: ”X isn’t in my dictionary/vocabulary!”, or  ”I don’t know the meaning of the word X!”, and variations on that theme. Language Log is notorious for fighting campaigns against the “no word for x” construction, used by poor journalists and politicians to equate lack of a term for something (often wrong) to ignorance about that thing in general. So I was surprised when I was unable to find a single post talking about the usage of the above trope. “The word X isn’t in my X” comes up with over 2.5 million hits on google. “I don’t know the meaning of the X!” comes up with another 2.5 million (although both could be used to mean literally not knowing something.)

Is this the same as “X language doesn’t have a word for X?” No. It’s not pejorative, for one. Generally this trope is used to intensify the quality of someone by negating the opposite quality: He’s fearless, as he doesn’t know the meaning of the word fear. Secondly, it’s generally not used to refer to more than one person. But it is similar in that it falls into the old trap of assuming that vocabulary items determine knowledge about a said thing, and that without a proper word, there can be no knowledge of its meaning. This is, of course, ridiculous. “Gravity?!” I might say, “I don’t know the meaning of the word!” Were I to jump off a cliff, I would come to a swift conclusion regarding whether or not I know what gravity is (as well as a swift conclusion anyway), even if I never learned the word gravity in the intervening 12.4 seconds when I realised that I was falling.

Seeing as how this idiomatic construction isn’t used normally to refer to insult and condescend an entire language group, however, I don’t think it’s as serious a sin. The only occurance in the field I can currently think of is in a copyrighted Calvin & Hobbes strip, where they are standing in a field, looking like a pair of pathetic peripatetics:

  • Hobbes: We’re lost, aren’t we?
  • Calvin: Lost? I don’t know the meaning of the word lost!
  • Hobbes: How about the word mommy?
  • Calvin & Hobbes: <shouting> MOMMY!

The joke doesn’t really transfer over in text. But you get the idea.

Posted in Linguistics Musing | 2 Comments

Look Not To This Post

My friend Jack was remarking on our fish tank. It is remarkable in that it has no fish, but rather a half inch of water and an apparently autochthonous snail population. (This is the first time I’ve used this, my favourite word, correctly and not metaphorically.) He was looking for them on the landward end of the tank, and I said: “Look not to the land.” Everyone in the room (around eight people) spontaneously began to laugh.


I immediatelly assumed because it was as if I had quoted something. But looking over google, I can’t find any usage of the “look not to x” that I would have seen before. There are a few examples from the King James Version of the Bible, which suggests archaic usage (of which I am not surprised). I may have sucked it up from there. There’s one usage of it in Lord of the Rings, my standard goto for strange archaisms I’ve internalised. On google, there are some wonderfully hilarious examples: look not to the Karzai government, look not to the social stereotypes too often reinforced by social psychologists, look not to his high linguistic level of speaking, etc. ‘Look Not To The Mountains’ is a new short film about hunting in Africa in 1904… And, strangely, ‘look not into the abyss’ comes up, although there is merely a mind-shatteringly small 9 hits for that, which is a misquotation (or non-standard translation) of a quote attributed to Nietzche.

So I don’t know where it came from. But this is merely a case of modal dropping and negation moving: ‘We look not’ instead of  ’We do not look’. Why is that inherently humorous? Well, I’m guessing not because it was a strange grammatical usage, but rather because I used it referring to snails, and spoke as if the fish tank were an entire microcosm. The context was everything. If I had said this in class, I doubt a single person would have laughed, least of all the lecturer. And so, I’ll end with two questions: should we posit context as an influence on grammaticality judgements, since the response differs based on context? And, secondly, how in the hell would one draw a tree of this?

Posted in Linguistics Musing | Leave a comment

Bilingualism’s Cognitive Effects in Children

In a recent Discovery News Article, Bilingualism Good for the Brain, three points were used to sum up a recent article in Science related to bilingualism.

  • “Speaking two or more languages appears to enhance executive function — the ability to focus on the information needed to complete a task.
  • Bilinguals with Alzheimer’s disease retained brain function longer than those who spoke only one language.
  • The “cost” of bilingualism is that bilinguals may have smaller vocabularies in each language.”

The second point is irrefutable, at least from my position. It concerns a recent test concerning Alzheimer’s in Canada.[2] The third and first are likely true, as well. But, having tracked down the relevant article in Science by Jared Diamond,[6] I did have a problem with the study that was held up as an example of this. Kovács and Mehler had done tests on bilingual versus monolingual pre-linguistic children (think 12-month-olds).[3] Diamond justified studying children this early by saying this:

Actually, infants learn to discriminate the sounds of the language or languages heard around them and to ignore sound distinctions not heard around them. For instance, Japanese infants lose, and English infants retain, the ability to discriminate the liquid consonants l and r, which the Japanese language does not distinguish.[1]

Since I am studying first language acquisition in relation to syllable processing, I wondered what exactly was being studied here: phonetic distinctions or the ability to notice unfamiliar phonotactics? Here are the details of the study, summed up by Diamond:

Kovács and Mehler devised a clever protocol in which infants looked for pictures of a puppet appearing on the left side of the computer screen. The infants were conditioned to anticipate the puppet by first hearing a nonsense trisyllable (e.g., “lo-lo-vu”). Within nine trials, both monolingual and bilingual infants learned to look toward the screen’s left side when they heard that trisyllable. But when Kovács and Mehler changed the rules and made the puppet appear on the screen’s right side after broadcasting a different trisyllable, the “bilingual” infants unlearned their previous lesson and learned the new response within six more trials. In contrast, the “monolingual” infants couldn’t learn the new response even after nine trials. Evidently, shifting frequently and unpredictably between hearing two parental languages made “bilingual” infants better able to cope with other unpredictable rule changes.[1]

Ok, so this is clearly not the /r/ and /l/ distinction, it’s to do with the cognitive abilities to recognise, internalise, and utilise systems. Putting aside the possible issues with using an eye-tracking experiment on one-year olds, the experiment seems to be pretty solid. I had two small problems with it: in the actual test, “infants listened to a trisyllabic speech item while watching a central attention-getter stimulus.” [3, 611] So, that’s merely a word of three syllables, in either AAB or ABA arrangement. But what was the attention-getter? And surely that would mean less attention to the auditory stimuli? The study, in that case, might show, either simultaneously with or instead of the conclusions drawn, that bilingual children are more able to internalise cues in the surrounding environment (if my thinking is correct).

Secondly, “Infants were first exposed to 36 AAB and ABA familiarization trials in pseudo-random order.” [3, 611] This intrigued me because what I wonder, again, is what this study was really diagnosing. I’m not sure that showing a three-syllable pattern, isolated, can really indicate different systems. This study would essentially show the ability for bilingual children to derive statistical analyses of syllable arrangement faster, and thus be able to better recognise variations or predict patterns, but only if one views it as a statistical analysis of syllable arrangement, such as A follows A follows B versus A follows B follows A. But one can’t really do that here, given the paucity of statistical analyses which the children were given. It’s been shown that it takes around, or more than, 3 minutes for children to do statistical analyses of uninterrupted syllable constructions.[5] (The methods used are actually variable as well; that’s what my thesis will be on.) Here, the infants are telling the difference between AAB and ABA – which must be something like spotting infixes versus suffixes (morphology). I wonder if there are different processes (or different variations on the same cognitive process) used to deal with morphology, syntax, phonotactic divisions. If so, then it might be possible that this study only dealt with an isolated occasion, and doesn’t necessarily hold true for general bilingual cognitive abilities. For my suspicion to be tested, however, there would have to be a wider range of systems used in the study. It would also be useful to have a control group of adults. Would bilingual adults score better on this test? Isn’t that the question we’re all asking ourselves, anyway: “Would I think better if I were Bilingual? Would I finish my Sudoku faster?” (Of course, this argument – that the results are not as wide-reaching as they might appear to be – is a favourite of naysayers, me included.)

The answer, at least for me, is who knows. But my hesitancy is, as usual, probably misplaced. The stats do show, in the end, that bilingual children really are better at this. I just wonder how ‘this’ is relevant to anything else.

As always, please shoot away at any holes in my arguments. This article was written in three stages, and I’m not sure that my initial biases didn’t win the day in the end.


  • 1. NYT:
  • 2. E. Bialystok, F.I. Craik, M.Freedman, Neuropsychologia 45, 459 (2007).
  • 3. A. M. Kovács, J. Mehler, Science 325, 611(2009).
  • 4. A.M. Kovács, J. Mehler, Proc. Natl. Acad. Sci. U.S.A.106, 6556 (2009).
  • 5. Aslin, Richard N. &  Jenny R. Saffran, Elissa L. Newport. Computation of Conditional Probability Statistics by 8-Month-Old Infants. Psychological Science, Vol. 9, No. 4. (July 1998), pp. 321-324.
  • 6. Diamond, Jared. The Benefits of Multilingualism. Science 330, 611-2 (2010).
Posted in Linguistics Musing | 16 Comments