Linguistics/Psychology 315/515
Prof. Suzanne Kemmer
Fall 2010

Course information
Reading list
Links to research

Assignment 1

Lexical Semantic Features
and Lexical Relations

Posted: Wed Sept. 1 2010
Due: Wed. Sept. 8 before class

Honor Code policy: You are free to discuss the assignment orally with others taking the class (making sure you are contributing and not just absorbing others' ideas). But don't share written work with one another until after the assignments are graded. Also don't use written work from students in previous years of this class. If you wish, you can test out data on native speakers of English from outside the class (as long as they haven't taken the class and as long as they are NOT linguistics grad students, profs or other linguistic professionals). Do not consult a dictionary for this assignment. Careful thought, probing your own and/or other speakers' native intuitions about what it is normal to say in a given situation vs. what is semantically anomalous, is the main vehicle to an insightful analysis.

Assignments should be uploaded to Owlspace. It will probably take at least 5 pages of text (if printed out) to answer these questions well. Total points 100.

Suggestion: allow yourself time to think about the assignment before writing, but leave plenty of time for writing. An analysis that seems perfectly straightforward when you talk or think about it may be hard to 'put into words'. Part of the exercise lies in sharpening your ability to clearly get across what you mean about meaning--not a simple task.

This is an exercise in "semantic feature analysis" or "componential analysis". The usual methodology for these types of analysis, which you can use here, is the intuitions of native speaker(s) (including the analyst, i.e. you) arrived at by controlled elicitation (cf. Tyler). Controlled elicitation for semantic features involves presenting examples to find which of them 'work' and which lead to semantic anomaly (meaning a clash of semantic components). By careful comparison and contrast of example sentences you should be able to arrive at some conclusions about meaning components of these words.

Asking native speakers (including yourself) how they/you would define the word is not a good methodology as it does not tap unconscious knowledge of meaning revealed by specific types of usages.

1. What semantic features (or components or meaning properties; terminology is flexible at this point) are present in the meanings of the words in (a)-(b) below? (80)

a. plod
b. scurry

For each case, give justifications for your features. In your justifications, support your claims by citing normal, odd, and abnormal uses of the word in phrases or sentences (as for example the following sentences with assassinate in them:

Several U.S. presidents have been assassinated while in office.
??The cat assassinated the bird.
???He slowly assassinated the prime minister.

Use some reasonable way of indicating degree of normality/acceptability, e.g. question marks as above, or words such as ODD).)

You can also support your analysis by citing other words which contrast with the target word with respect to particular features. Focus on the literal, motion verb senses of the words.

A tip: Don't be scared of 'semantic fuzziness'. Try to identify the meaning properties that seem to be most important at getting at the essence of the word. These are the ones that, if omitted, would give a distorted idea of the word's meaning.

If you notice other properties that seem to you to be conventionally associated with the word, discuss them too. These may be more variable across speakers, but you can argue for their validity in your speech (or the speech of other native speakers you consult) by citing relevant examples of normal, odd and distinctly abnormal usages.

2. For the pair of partial synonyms in (a) below, determine the semantic features or properties that are shared by members of the pair, and determine the features/properties that distinguish the words in the pair from one another semantically. Provide evidence for your features as above. (20)

a. float -- drift

Your analysis for this problem will be limited to shared and distinguishing properties of the words (supported by examples of use), rather than a full semantic description as in the previous problem.

© 2010Suzanne Kemmer
Last modified 1 Sept 2010