L103, Spring 2003: Examination 1

General instructions

Save your answers in a text file, putting two carriage returns after the answer to each question so that we have a place for annotations. Then upload the file using Annotate. You have until midnight tonight (Monday) to finish; the answers will be posted then, and no more submissions will be accepted. Start early. You may use any resources you want except other people. If you have questions about what a question means, send email to Mike, but do not wait till the last minute to do this.

Level-1 questions

1. Grammaticality, etc.
   You are a linguist studying some aspects of the pronunciation of English in Liverpool, England, and you have recorded several speakers, all of them university graduates. You notice on your recordings that one of the speakers pronounces the words book and buck exactly the same. You are trying to get more information about this behavior. For each of the following, say how it could help you, if at all, and explain your answer in a sentence.
   1. You check whether the speaker also pronounces other pairs of words the same, for example, stood and stud.

      This could be useful to you, first, because it can tell you whether the speaker's pronunciation of book and buck was just an accident (error) of some sort, or whether it is really general in the speaker's pronunciation, and second, because it can tell you whether this feature applies more generally in the speech of the person than to just those two words.

   2. You check how other people on your recording pronounce these words.

      This could be useful to you too because it could tell you whether the pronunciation is acceptable in this dialect, that is, whether this is a convention for them (rather than just a feature of the one speaker's idiolect).

   3. You take a poll of a group of high school English teachers in Liverpool, asking them whether it is correct to pronounce this pair of words the same.

      This would not be likely to help you at all. What English teachers think people should do is not relevant to what they actually do, which is what you as a linguist are interested in.

   4. You record the speeches of several Liverpudlian politicians, listening for examples of these or similar words.

      The speech of these policians could be expected to be relatively formal. It could tell you whether the pronunciation that you noticed is appropriate in all contexts or whether it is confined to informal registers within the dialect.

2. Dimensions and values
   Below are the personal pronouns of Lingala. Decide which dimensions are relevant for the personal pronouns in this language, and for each pronoun, say what the values on each dimension are. If a dimension is unspecified (irrelevant) for a pronoun, just leave it out. Be sure to provide enough dimensions and values to distinguish all of the words, but do not include any dimensions that are not relevant for this language. For example, if the language were English, you would write the following for the pronoun I:
   I -- PERSON: 1, NUMBER: SINGULAR

   Dimensions: PERSON (1, 2, 3), NUMBER (SINGULAR, PLURAL), ANIMACY (YES, NO)

   ngai 'I' PERSON: 1, NUMBER: SINGULAR
   yo 'you (one person)' PERSON: 2, NUMBER: SINGULAR
   ye 'he' or 'she' PERSON: 3, NUMBER: SINGULAR: HUMAN: YES
   yango 'it' PERSON: 3, NUMBER: SINGULAR: HUMAN: NO
   biso 'we' PERSON: 1, NUMBER: PLURAL
   bino 'you (more than one person)' PERSON: 2, NUMBER: PLURAL
   bango 'they' PERSON: 3, NUMBER: PLURAL

Level-2 questions

1. For each of the following semantic extensions, say whether it is based on metaphor, metonymy, generalization, or specialization, and explain your answer in a sentence.
   1. In describing chess games, people often refer to players by the color of their pieces: "Then Black captured White's queen."

      Metonymy. Each player has a particular color associated with them; the color belongs to the player in a sense. So we can use the color to refer to the player.

   2. The word pig is sometimes used to refer to a person who eats a lot or eats in a slovenly manner.

      Metaphor. A person who eats a lot resembles a pig (or at least one stereotype of a pig) in this way; it is the similarity between the two concepts that is the basis for the extension.



2. For each of the following properties of languages, say whether it is oriented towards the Speaker (production), the Hearer (understanding), or the Learner (long-term memory), and explain your answer in a sentence. (There may be more than one possible answer, so your explanation is very important.)
   1. No known language uses a sound for which the tongue touches the nose.

      Speaker-oriented. It is very difficult, if not impossible, for most people to touch their nose with their tongue; that is, production would be very awkward with a sound like this.

   2. When people speak to young children, they usually talk about the "here and now", that is, about things that are visible to the children, rather than about things that happened in the past or are expected to happen in the future.

      Two possibilities
      A. Hearer-oriented. Talking about things that are visible makes language easier to understand because the set of possible things being referred to is small, and the Hearer doesn't have to search memory to find what is being talked about.
      B. Learner-oriented. Talking about things that are visible makes it easier for learners to figure out the meanings of unknown words because the set of possible things being referred to is small; without knowing what is referred to, there is no way the child can learn what a word means.

3. A baby hears the word jello on two occasions, for a salad in which there are cubes of green, lime-flavored jello, and for a dish of green, lime-flavored jello (with the jello taking the shape of the dish).
   1. Give an example of an over-generalization or under-generalization (specialization) error that the baby might now make in the use of the word. In a sentence explain which kind of error it is.

      A. The baby uses the word jello only for green jello. This is under-generalization because the baby's category is narrower than the adult category (a sub-category of the adult category, which includes all colors and flavors of jello).
      B. The baby used the word jello to refer to all sweet food. This is over-generalization because the baby's category is broader than the adult category (a super-category of the adult category, which includes only one kind of sweet food).

   2. Explain why the shape bias, as described in the book, would not help the baby learn the meaning of the word.

      The shape bias leads the learner to generalize on the basis of shape, but in this case the two examples of the word have completely different shapes, so there is no shape-based generalization to make about the word jello.

   3. Describe an additional kind of bias, comparable to the shape bias, that might help the baby learn this word and similar words.

      As stated in the section on learning meaning in the book, the shape bias only applies to solid objects. Jello is only roughly solid. We can imagine a comparable "material bias" for non-solid things (masses), such as jello, whipped cream, clay, and butter for which the learner generalizes on the basis of texture or consistency rather than shape (or color or size). In the jello example, this bias would prevent the child from making generalization errors based on color or sweetness.