7.4 lang science | exploringlanguage

7.4 Language Analysis Is Science

You might not think of language when you think of science, but people who study language are linguists – and they are also scientists; they investigate language scientifically.

Linguists use the SCIENTIFIC METHOD to analyze language

Linguists, and other scientists, begin with a description that they get from making observations.
Then they make a hypothesis about what’s going on with the thing they’re studying.
Then they must show that their hypothesis is right by testing it with new data, and they should look for counterexamples that would disprove their hypothesis.
If counterexamples are found, then they revise the hypothesis.

Throughout these language lessons, you have often used scientific methodology to figure out things about language. Let’s look at one more example of this from the Native American Language Lushootseed, spoken in Washington State.

Lushootseed uses a word formation process called reduplication in various parts of a word are doubled to create a word with a different meaning. There are several different types of reduplicative processes. For each set of data, answer the following question:

How does you form the words in column B from the words in column A?

Data Set 1 (adapted from Hess and Hilbert, 1995)

A B

ʔálʔal house ʔáʔálʔal hut

ʔúqw’ud pull out ʔúʔúqw’ud pull part way out

híwil go ahead híhíwil go on ahead a bit

q’íxw upstream q’íq’íxw a little bit upstream

Data Set 2

A B

gwəčəd look for something gwəčgwəčəd several people look for something

dəšəd be on side dəšdəšəd set many things on side

yəcəd report him/it yəcyəcəd always talking about him

tədəd put them in row tədtədəd put them in rows

t’əq’əd patch it t’əqt’əq’əd patch it up

čəgwás wife čəgwčəgwás seeking a woman to marry

Data Set 3

A B

ʔáɬ fast, quickly ʔáɬáɬ hurry up!

dzáq’ fall, topple dzáq’áq’ stagger, totter

čəx̌ split čəx̌əx̌ cracked to pieces

Do the words below, another word pair of the Data Set 3 type, conform to your hypothesis for Data Set 3? If not, revise your hypothesis to account for this pair.

yubil starve yububil tired out, sick

gwədil sit down gwədədil sitting for lack of anything else to do

saxwəb jump, run saxwaxwəb scurrying about ineffectively

Revised Hypothesis:

[Teacher Notes: Data Set 1: Double the first two sounds/letters. (ʔ is a consonant called a glottal stop that makes the sound in the middle of the Enlgish world 'uh oh.' Students may be curious about some of the other letters. Their pronunciation can be found here: http://archive.org/stream/rosettaproject_lut_ortho-1/rosettaproject_lut_ortho-1_djvu.txt ) This process seems to change the meaning to a smaller version of the original word. It's a diminutive. Data Set 2: This process doubles the first three sounds/letters in order to make a new word (resulting in some kind of distributive meaning). Data Set 3: This process appears to double the last two letters/sound (in order to form what has been called the 'out of control' affix). However, upon expanding the data set, we see we need to reevaluate. Have students state what the problem is: that doubling the last two would result in yubilil. Revising to account for all of the data: Take the first syllable, remove the first letter/sound (called the onset of the syllable), and double the rest. Lushootseed speakers do this without even thinking about it and likely woulnd't be able to say what they're doing. They know this word formation process unconsciously.]

key words: Native American languages, science, hypothesis, morphology, scientific methodology

Here is this lesson as a pdf.

Exploring Language

Daily Language Investigations for English Language Arts

Linguists, and other scientists, begin with a description that they get from making observations.

Then they make a hypothesis about what’s going on with the thing they’re studying.

Then they must show that their hypothesis is right by testing it with new data, and they should look for counterexamples that would disprove their hypothesis.