Inferring:

Most literary elements of stories require us to *infer*. For instance, even character studies require us to infer by connecting characters' appearance, behavior, etc. to their personality traits. Often, in making these inferences, we connect clues from the text to our background knowledge. For instance, if we describe a character as "good-natured," we need to know not only what the character has said or done that marks him as good-natured, but also what good-natured means. The quality of our inference depends not only on our attention to the text but also on the quality of our background knowledge. When your background knowledge is accurate, your attention to the text is close, and your connections are logical, then your inferences are likely to be strong. When you are writing about your inferences and observations, it is often a good idea to state your background knowledge clearly and even back it up with sources, so that your reader will understand the source of your inference.

Literary elements often operate to make the specific textbound details more universal. When we are making inferences about symbols, for instance, we should usually look for abstract meanings. We should also usually look for meanings that are not limited to the specific place and time in which the story occurs, but may have relevance for other people living at other places and times. (If the story has no deep or universal significance, it is usually found in a newspaper or a history book, not a collection of literature).

As you begin to write paragraphs about your inferences, you will likely discover that you don't have enough details to support your inferences. Inferring and attention to detail go hand in hand. Once you have an inference, you should go back to the text and look for other details that support your idea. You should also read carefully, asking yourself questions and writing on the page, the first time through the story.

Since you should usually have several supporting details for each paragraph, your topic sentence will need to be broader than your examples, because it needs to include all of them.

Anastasia Krupnik books
Dave Barry books
business-y books
books about being a doctor
books about getting in to college
poems
s.e. hinton books
all newbury books

incorporating background knowledge into your essays

a) in the introduction

b) in the exposition of your supporting paragraphs

c) etc etc

inference as clues + background knowledge. . .

"what has happened here" elsa barkley brown

Lonesome Words: The Vocal Poetics of the Old English Lament and the African-American Blues Song
by Margaret G. McGeachy, M. G. McGeachy

Poetry of Mourning: The Modern Elegy from Hardy to Heaney (Paperback)

Science reading: the myths of innovation, berkun

They Say, I Say: "Letter From Birmingham Jail"; Op-Eds


Social Sciences

Language and Communications
History
Politics and Government
Cultural Critique

Humanities
(An Album of Styles)
Literature and the Arts
Philosophy and Religion

Narrative
Personal Report
People, Places
(Education)
Prose Forms: Journals

Natural Science
Nature and the Environment
Science
Human Nature

Irony
Prose Forms: Proverbs

specific-sounding choices that lack concrete details

inference: cause-and-effect info.

reading: unfamiliar text more actively

main idea should a) make claim about the topic and b) lack concrete detail

after 500s, try reading 1st, 2nd, and last sentences of supporting paragraphs; skim everything else.

300=range scorers: very weak scanning skills. Need to work on scanning skills.

inference is really all about cause and effect.

What would author A say about ref B is really inference. Compare main idea of passage A with ref and find connection, then answer question.

Interesting:

http://www.cs.nott.ac.uk/~ceilidh/papers/Morpheme.html

Google search: stripping suffixes prefixes


Also, it was supposed to be possible to delete all words that never occur in academic writing.

another possible sorting criterion:

occurrence within the TIME corpus (maybe more than 50 occurrences)?

What could I do?

Well, first of all, I could start by stripping down that list:

a) running it through the prefix-suffix remover;
b) running it through the list of criteria I already posted.

Then I could run it through WordNet and pick up some definitions and synonyms.

Then I could parse it against itself and remove all the synonyms that weren't in the list.

Then I could pull the example sentences and the collocates.

If only that site can be parsed! (!!)

Then I would want to yank:

a) the collocates:
for intransitive verbs, nouns one before, prepositions one after, and nouns two after;

b) for transitive verbs, nouns one before and nouns one after;

c) for adjectives, nouns one after;

d) for prepositions, nouns one before, verbs one before, and nouns one after.

Also: 3 example sentences per word, ideally based on collocates, maybe one for each.

The Wish List of Extremely Robust Searchiness:

1) Def.
2) Synonyms *without* slangy or obscene ones.
3) Robust collocation locator.
4) Example sentences:
a) with appropriate collocates;
b) at a good reading level.
5) Use of the following webpage:
http://www.unc.edu/~jelsas/Research/TextMiningToolkit/manual/