Teaching science through language

« previous post | next post »

This morning, here at AAAS 2012, I'll be attending a symposium on "Teaching Science Through Language", organized by Anne Lobeck. The abstract:

There is a need for highly effective science education and for more successful ways to teach scientific inquiry. Work on language can play an important role in developing the concepts and skills necessary for understanding how science works. Language provides a wealth of data available from the students themselves — data with questions that beg to be asked, making everyday phenomena surprisingly unfamiliar and requiring explanation. Linguistics is at the core of cognitive science, offering incomparable ways to understand the nature of the human mind. The biological capacity for language appears to be shaped in part by genetic information and in part by information gained through childhood experience. Scientists have sought to tease that information apart, and this work has yielded good explanations in some domains and a body of understanding that can be made accessible to middle school and high school students. This symposium presents examples of linguistic puzzles that can be integrated into existing school curricula and that enable all children to understand elements of scientific work quite generally and to discover their own intuitive knowledge of language. (For example, how do we know that greebies is a noun in The greebies snarfed granflons, but a verb in Lulu greebies me?) All of this can be done without labs or expensive equipment by involving experimentation, observation, and testing of hypotheses.

Perhaps it's because we're Section Z, at the metaphorical end of the alphabet, that they schedule most of the language-related sessions at 8:00 am. This doesn't bother me, since I'm an early riser, but it's yet another reason to put make these sessions available to a wider audience via videos and slides on the web.

The "Science through language" symposium looks terrific. But as a phonetician, I'm sorry not to see something on the opportunities to teach natural science through the study of speech. Free software (Praat, Wavesurfer, Audacity, etc.) is easily available that will let students explore and measure speech sounds, learning about simple concepts like period, frequency, and wavelength, or more complex concepts like resonance and spectrum. Depending on age and interests, the explorations could branch off into statistics, or accent variation and sociolinguistics, or biometrics, or psychophysics, or whatever.

And here's a neat idea that it would be nice to replicate. Years ago, David Talkin used a lathe to shape wooden cylinders into a profile matching vocal-tract area functions. He then used these to cast plaster forms containing appropriately-shaped cavities. Blowing raspberries into these cavities then produced clearly-recognizable vowel sounds. It would be easy and cheap to replicate the positive or negative forms. And today, it's even easier to create a simple vocoder program that would do the same thing using arbitrary  digital recordings as excitation functions. (Praat contains all the functions that you need to do this and other experiments in the analysis/resynthesis of speech. And in fact, bright high-school students can learn how to write their own simple programs, say in Octave, to implement this sort of thing for themselves, thereby learning a bit of linear algebra and programming at the same time.)


  1. Andrew Shields said,

    February 18, 2012 @ 11:07 am

    "data with questions that beg to be asked": that phrase contains a nice little question that you've discussed at length yourself!

    What's really nice about this symposium is that it highlights that language is an ideal object of scientific study, that it has been extensively studied, and that it is not just a matter of thinking about how you use language yourself.

  2. Andy Averill said,

    February 18, 2012 @ 3:27 pm

    Speaking of non-electronic speech synthesizers, back in the 80's I owned a ridged plastic string that said "happy birthday" when you ran your fingernail across it at the right speed. It was a big hit at birthday parties. Unfortunately I seem to have lost it. Has anybody else run into this gadget? Made in Japan, I'm pretty sure.

  3. Maryellen MacDonald said,

    February 18, 2012 @ 9:03 pm

    At U. Southern California, the teaching science through language course has existed for almost 15 years. It's a General Education Science course that undergraduates can choose to fulfill part of their science requirement and has a substantial phonetics component with labs using speech analysis software, as Mark suggests. The class, Linguistics 275, Language and Mind, was originally developed by Barry Schein based on ideas for using language to teach science by I think Ken Hale, among others. The course was designed to be co-taught by a linguist and a psycholinguist, and Barry and I taught the first offering. The class now has its own textbook, by two of the other co-instructors, Dani Byrd and Toby Mintz, http://www.amazon.com/Discovering-Speech-Words-Mind-Dani/dp/1405157984/ref=tmm_hrd_title_0. Many students took the class because they were afraid of science and were quite surprised to find out just how much science there was in the class and the field.

  4. galthran said,

    February 19, 2012 @ 11:06 am

    We know no such thing about greebies! It's obviously an adjective in the first sentence (along with snarfed) and nonsense in the second.

  5. Joseph F Foster said,

    February 19, 2012 @ 11:29 pm

    Mr. Gathran,
    If The greebies snarfed franflons. is a "sentence" as you say and the author seems to have intended, then greebies is a noun. As to the second, Lulu greeboes me. being "nonsense", that's not at all obvious. If we knew what the verb greebie means, we'ld probably know what the sentence means, in which case it would not be nonsense.

  6. galthran said,

    February 20, 2012 @ 12:17 pm

    I guess the humor was indeed lost, but is (greebies snarfed granflons) = (purple ridiculous sausages) not a valid analogy? It is a guess at an interpretation that produces a result as grammatical as your guess. I know the -s and -ed endings are trying to imply otherwise but it's not something that I think is "known" with such obviousness.

  7. Maryellen MacDonald said,

    February 20, 2012 @ 7:58 pm

    Lulu greebies me sounds nonsense-y to Galthran presumably because greebies has a highly atypical sound pattern for a verb, even with the -s 3rd person singular ending and sentence context.

RSS feed for comments on this post