Thursday, May 27, 2010

A Weekend at the Beach

I'm getting in early with this week's post, as I'll be away tomorrow and for the rest of the weekend. I'm off to the NSW central coast for a weekend with my project team. It's something like school camp, but much more nerdy, much more fun and no one to tell us to go to bed at 9pm.

I'll be showing the team what I've been up to for the last year, which is somewhat nerve-wracking and exciting. Exciting, because this time last year I had nothing, or less than nothing, to show them, and this year I've got lots of my own data. Nerve-wracking because showing people your analysis - even a tentative one - is always stressful. Fortunately they're all very nice people so I won't be too worried.

And in language news this week, we see that computational manipulation of language can actually have some good.

Thursday, May 20, 2010

Time to play the name game

For the last couple of months I've enjoyed sharing Friday thoughts on a range of topics beyond my own work. This doesn't mean that I haven't been doing any of my own work, far from it. I'm still tidying up some final transcriptions from the last trip, thinking about what I need to do on the next trip (lots) and reading, reading, reading.

One thing I have been thinking about is that I am tired of referring to the language I'm working with as K. As I mentioned while I Nepal, there was some confusion over which language I was working on, and it eventually transpired that it is a previously undocumented dialect of a larger language. I maintained references to the language as K. until now for the sake of consistency.

The reason for this convention has nothing to do with the fact I'm trying to keep you from finding out who I'm working with and where. I'll happily tell you the name of the language if you email me/come to any presentation I do or read anything I (eventually) publish on this topic. It's more that the internet has a merciless memory, and if, in the future, people do want to find out about this language I don't want this blog to be the first thing that pops up.

I have been pondering a solution for a while and have decided to fabricate a pseudonym instead of continuing to resort to an initial. Therefore, for all future references in this blog to the language I'm working with and the people who speak it I'll use the word "Tam". This is their word for 'language' or 'speech' and is sufficiently short and snappy. For those that care it's a high tone word.

So no more Poe-like references to the language K., from now on all my rantings will be in relation to Tam - as well as whatever other linguistic matters pique my interest.

Friday, May 14, 2010

Dictionary Fail Fail

Headline making news from LexicographyLand this week:

Gasp! Egads! Someone found an error in the Oxford Dictionary and now the world as we know it is about to end!

Who would have thought that a dictionary with over 301,100 main entries compiled by mere mortals (and the mentally unstable would have an error! I would be impressed if we only found one.

Saturday, May 8, 2010

Scarlett's a Flop

While getting my usual dose of trashy celebrity news a few days ago, I came across an article about Scarlett Johansson and her problems with sweating:

Rather than muse over why I find it necessary to waste my time reading about the bodily functions of people I don't know, I want to look at something in particular Johansson was quoted as saying in the article

"I like the getting ready part. I like the hair and the make-up and the pretty dress... It's just getting on the red carpet I just instantly flop sweat."

What the heck is "flop sweat"? It's certainly a phrase I've never heard before. For once the Oxford English Dictionary wasn't able to come to my rescue - It had no information on the phrase. Webster's online was much more helpful, unsurprising perhaps, given it was used by an American - it appears that only Americans "flop sweat" ( ). I was somewhat surprised to see that it's earliest cited reference was in 1953. Clearly Americans have been suffering this problem without telling us for a while.

Webster's tells us that this phrase acts as a noun in English, but I'm pretty unconvinced by that. Given the context above, it's very much functioning as a verb - there's no other verb in " I just instantly flop sweat" and it's being modified by the adverb "instantly" - nouns modified by adverbs don't really happen in English.

Even though it's clearly not a noun, but some kind of verb, as an Australian English speaker I find it difficult to parse. I'm not very sure which part is the core verb of the construction, which is made rather difficult by the fact that both "flop" and "sweat" can be either nouns or verbs. Option a) is that "flop" is the verb, and matches with things like "I drip sweat" and "I shed skin". Option b) is that it is "sweat" that is the verb and "flop" is a modifier, like "heat sweat" and "stress eat".

If we were to turn it into a progressive aspect (that's basically an -ing form in English) then we would have to place the progressive suffix on the verb of the sentence:

Option a) "I was flopping sweat"
Option b) "I was flop sweating"

I thought that option a) was more likely before I did this test, but perhaps it is "sweat" that is the verbal element here. I'm still confused. If anyone else has heard this phrase before, or has any thoughts on it I'd love to hear.

Friday, May 7, 2010

Colourful Results

The team at xkcd have published the results of their massive colour name survey (be warned, some of the language results is a little colourful!):

Names for colours are not the same across different langauges, some languages for example don't have separate names for what we call blue and green (these are called 'grue' languages), while some may have more specific basic colour terms than we have (Russian, for example, has two different basic blues they distinguish which we'd just call 'light' and 'dark').

Lots of work has been done by linguists over the last forty years looking at colour term variation - but the volume of data collected in this online survey is staggering. over 5 million colours were named over almost a quarter of a million individual sessions. And because they are free of the pressures and constraints of institutionalised research they've made the data available for you do play with at home!

The only real variable they've examined to date is gender, but given that participants were asked to give their native language as metadata there is potential for some cross-linguistic analysis.