Tuesday, August 30, 2011

language organelles in the human brain


Out of the nine regions they analyzed -- four in the left frontal lobe, including the region known as Broca's area, and five further back in the left hemisphere -- eight uniquely supported language, showing no significant activation for any of the seven other tasks. These findings indicate a "striking degree of functional specificity for language," as the researchers report in their paper.

Future studies will test the newly identified language areas with even more non-language tasks to see if their functional specificity holds up; the researchers also plan to delve deeper into these areas to discover which particular linguistic jobs each is responsible for.

Sunday, August 28, 2011

Chinese without tones

This was inspired by a Lingomi Blog.
Many famous scholars write about how characters will (or should) be replaced by a romanized alphabet (link). But I’ve been wondering about whether Chinese could lose something else it’s famous for. Could Chinese ever lose its tones?

Chinese without tones? But aren’t tones kind of essential to Chinese? Yes, they are. If you’re learning Chinese and don’t know your tones, frustration will ensue. Tones are important (a subject I’ve been known to blog about), but there is one situation where tones don’t matter: music.

Why don’t chinese songs have tones, and how can Chinese speakers understand the words without them? (Ask antimoon has a few good answers to this question). The fact that Chinese people do understand music lyrics proves that it is possible for Chinese speakers to understand tone-less Chinese.

But could tone-less Chinese become more widespread?

D - I could not figure out how to post in their comment section.

I was thinking about the Bantu pitch register system.
Chinese is not conveniently sorted in a way that each tone denotes a certain grammatical category.

With 'ma', you might use:

•妈 mā mother
•麻 má bother
•马 mǎ horse
•骂 mà scold
•吗 ma (question tag)
If it helps, try a sentence like: "Mom is bothered by the horse's scolding - yes?" Not an entirely sensible sentence, but it at least catches all the five tones (including neutral) and in the right order.

The Thai language is a tonal language. This means that the same word said with a different tone, can mean totally different things. For example the word “mai” said with a low tone (l) means “new”. “Mai” said with a rising tone (r) makes a question. ”Mai” said with a falling tone (f) means “to burn”. “Mai” with a rising tone (r) spelt a different way means “silk”. So if I ask a Thai person the following question “Mai (r) mai (l) mai (f) mai (r)…. aka mai mai mai mai, I have asked the question “does new silk burn?”

D: if Chinese had a tone that corresponded to, say, verbs, then losing tone would be less ambiguous.
Take the example of "ma", meaning horse, hemp, mother and scold.
That is 3 nouns and 1 verb.
Now if we had ma mean horse, scold, big and quickly, we'd have something to work with.
Rigid SVO word order would ensure that the position of ma in a sentence would denote what grammatical function it served, even without tone.

So I guess I'm saying that while Chinese cannot avoid a vast # of homophones without tone, a designed language reminiscent of Chinese COULD avoid this. It could be used with variable word order by native speakers (or fluent ones) by using the tones in lieu of a 'latinate' affix system.


Thursday, August 11, 2011

Advice on apostrophes 'n possessive


To answer the name question first: It varies. Some journals use the apostrophe alone, with a singular name ending in s, some use an apostrophe-s. I vaguely remember being taught to avoid the possessives on names with one syllable (“Keats’ poetry”) and to add it to multisyllabic names (“Davis’s hernia”) – or was it the other way around? Some houses do in fact have variable rules within the same publication. Jesus and Moses, for example, and Greek names such as Ulysses and Socrates, are often exceptions, and are written in their possessive forms without an extra s (“Jesus’ name”, “Socrates’ argument”). The Globe and Mail is admirably consistent on this one: We add the logical apostrophe-s to everything,

But The Globe and Mail’s style guide also notes that recasting these sentences to avoid too many sibilants can itself end up in awkward contortions, and that it is preferable to write “the hostess’s gown” than “the gown of the hostess.” The latter just sounds stilted. And there will always be exceptions to everything: When it comes to odd team names like the White Sox, we don’t add an extra s for a possessive form; we write “the Sox’ losing streak.” How is it possible to keep all this straight?

Friday, August 5, 2011

net search considers relationships


Etzioni proposes that instead of simply looking for strings of text, a web search engine would identify basic entities -- people, places, things -- and uncover the relationships between them. This is the goal of the UW's Turing Center, which he directs.

The Turing Center has developed an open-source tool called ReVerb that uses information on the web to determine the relationship between two entities.


ReVerb is a program that automatically identifies and extracts binary relationships from English sentences. ReVerb is designed for Web-scale information extraction, where the target relations cannot be specified in advance and speed is important.

Open Information Extraction (IE) is the task of extracting assertions from massive corpora without requiring a pre-specified vocabulary. This paper shows that the output of state-of- the-art Open IE systems is rife with uninformative and incoherent extractions. To over- come these problems, we introduce two simple syntactic and lexical constraints on bi- nary relations expressed by verbs. We implemented the constraints in the ReVerb Open IE system, which more than doubles the area under the precision-recall curve relative to previous extractors such as TextRunner and WOE-pos. More than 30% of ReVerb's extractions are at precision 0.8 or higher— compared to virtually none for earlier systems. The paper concludes with a detailed analysis of ReVerb's errors, suggesting directions for future work.

Wednesday, August 3, 2011

The Globe and Fail. Semicolons.


"The company is counting on the new devices to help it weather the rough transition to the more advanced QNX software platform next year. In the meantime, all of the new BlackBerrys – a full touchscreen device; a touchscreen version of its popular Bold smart phone; and a more powerful version of the Torch – run on a sleeker, updated version of the company’s operating system."

(From Wiki)
Applications of the semicolon in English include:

Between items in a series or listing containing internal punctuation, especially parenthetic commas, where the semicolons function as serial commas:
She saw three men: Jamie, who came from New Zealand; John, the milkman's son; and George, a gaunt kind of man.

Between closely related independent clauses not conjoined with a coordinating conjunction:
I went to the basketball court; I was told it was closed for cleaning.

Between independent clauses and semi clauses linked with a transitional phrase or a conjunctive adverb:
Everyone knows he is guilty of committing the crime; of course, it will never be proven.

Of course, the simplest solution would be to write that passage as two separate sentences.

I am learning Office 2010 right now.
I wonder if it can detect those mistakes?
Grammar checkers have progressed since my U days.