Monday, September 16, 2013

"Harvarding" a book

Here, according to Kristin Luker, is how to "Harvard" a book:

1. Look at the table of contents and index, focusing your attention on topics you care about

2. Skim the introduction and conclusion

3. Skim the chapters that seem relevant

4. Decide whether the book is worth reading.

5. (Optional) Read 4-7 reviews of the book.

I would add that you can/should "Harvard" the book reviews, too.
So here's my attempt to Harvard a book that I have already read part of: The Empirical Base of Linguistics: Grammaticality Judgments and Linguistic Methodology. I'm giving myself 20 minutes to do it. Go.

Preface: purpose of the book is "to demonstrate the the absence of a methodology of GJs in linguistics constitutes a serious obstacle to meaningful research" (p xi). I would say this is similar in sociolx/app lx, WEs, though the solutions will be different from those proposed for Lx.

"grammars of intuition" is a problem.

GJTs are "pseudoexperimental" but have "no experimental controls" and often only the linguist him/herself is the "subject"

NB: Schutze is critiquing GJs in "theoretical syntax" studies.

GJs are "not pure sources of data" but "instances of metalx performance."

4 reasons for using GJs:

1. examining sentences types that rarely occur
2. examining “negative information” about “strings that are not part of the language"
3. distinguishing grammatical knowledge from accidents of speech
4.  minimizing other factors in the study of the mental nature of grammar 

lx intuition is 'shifty and variable' Householder 1965 (awesome name!)

"I do not believe one can defend the sufficiency of judgments alone" p3

Lx judgments do play a fairly central role in our day to day lives. p4

GJs are just as much performance and confounded by just as many factors as other kinds of lx performance data p6

--- Note: Ultimately Schutze wants to reform empirical data-gathering processes of elicited metalinguistic performance for Linguistics qua Linguistics -- this is traditional linguistics in the sense that it is looking at language from a more or less structuralist, Chomskyan, cognitivist perspective. Right? Not that there's anything wrong with that! But it's not what I'm doing -- I want to re theorize AJTs from a sociocultural perspective. -- 

"A working hypothesis'

"we should start from the position that te entire behaviour of making GJs is the results of interactions between primarily language faculties of the mind and general cognitive properties, and crucially does not involve special components dedicated to linguistic intuition." (fine!) p14

language is like other behaviours p15

Cites Chaudron and Birdsong as previous treatments of the topic, though notes they are both L2 focused. Says no one has done what he has done "within the basic framework of generative grammar." p17

--- Note: skim Birdsong again! --

p25 "acceptability" isn't the same as "accepted" 

Defn of acceptability:

1. any particular instance of a speaker accepting or rejecting a sentence is an act of performance
2.ANy sort of generalization across many such instances is a generalization about performance
3. A judgment of one's disposition towards accepting or rejection a sentence is itself a type of performance.

--Sure -- again , performance is not a problem for what I am proposing.--

The first 20 pages of Aspects of the Theory of Syntax sets up all future GJ/AJ studies, no? (p28)

p212 "linguistics has nothing to lose by taking data collection, particularly judgement collection, more seriously, both with regard to the insights that will be gained and the theoretical issues that will be clarified, and with regard to the standing of the field as a scientific endeavor or in the larger academic setting. The realization seems to be growing that the psychology of GJs can no longer be ignored."

Again, I want to do the same thing -- but not for theoretical lx, and not about psychology -- I want to understand this data collection from a qualitative / sociolingusitic / discursive / sociocultural / etc perspective.

No comments: