COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	.require "let.pub[let,jmc]" source
C00010 ENDMK
C⊗;
.require "let.pub[let,jmc]" source;
.turn on "→";
∂AIL Editor↓%2Creative Computing%1↓20 Lynnfield Drive↓Morristown, N.J. 07960∞

.once center
%3A PREJUDICED ANALYSIS%1

Dear Editor:

	The "Prejudice Analysis" program by Richard Kahn and Mark Gross
described in your September-October issue is a prize example of the misuse
of computers denounced by Weizenbaum and others.  It falsely claims to
analyze, disguises propaganda as science, and intimidates its subjects.
To whatever extent Dr. Siegel of Tufts is responsible, he has shown
neither the honesty required of a scientist nor the concern for subjects
required of a psychiatrist.

	Here are the counts of the indictment:

	1. %3The computer is used to mystify the user, disguise what is being
done, and lend scientific authority to opinion%1.  It claims to "analyze"
the extent of a person's racial prejudice.  It simply counts the extent
of his agreement with the "right answers".  The more strongly he holds the
right opinions the less prejudiced he is said to be.
That cannot honestly be called analysis.

	2. %3The questionnaire itself is a propaganda document%1.
Many of the obviously "prejudiced" statements are caricatures of the
actual beliefs of most of Siegel's foes - e.g. the opponents of busing.

	3. %3The program expresses fanatical intolerance%1.  Each item
has a right response, and disagreeing slightly is counted "ANSWERED IN A
RACIST MANNER".

	4. %3The program teaches that objectivity is racist%1.
Many of the statements concern matters that vary from area to area, e.g.
whether a neighborhood is open to Negroes, whether welfare families would
soon follow Black families into the neighborhood, whether particular
schools would be better with more minority groups, and whether property
values would go up or down if minorities entered.  The student is taught
that circumstances are irrelevant; the non-racist answer is always the
same.  Many of the questions are only tenuously related to race, but
you're racist unless you conform.

	5. %3The word %2prejudice%3 is misused%1.
Its use started when it was observed that many anti-Negro and
anti-Semitic views were based on hearsay and were usually abandoned on
acquaintance.  Such a view is a prejudice, but many views called
prejudices here are held by people as much acquainted with minorities as
Siegel, Kahn and Gross.  They still may be mistaken, but they are not
prejudices unless thoughtlessly held.

	6. %3The program instigates and manipulates guilt feelings
to browbeat students into conformity%1.  The student soon recognizes that
he is in the hands of fanatics.  If he wants the good will of his teacher
and his right-thinking classmates, he will know how he must answer all but
question 9.  Unless he is rather subtle, he will believe he ought to
disagree with %2"City riots are a threat to our suburban life"%1 just as
he must disagree with the more traditional %2"Propery values will go down
if minorities enter the neighborhood"%1.  But this is the racist response.

	Perhaps not agreeing doesn't take rioters seriously enough.  The
effect is to trip the unwary and worry the wary.  Almost no
subject will get a clean bill of health, and each will go away with a
feeling of guilt.

	6. %3Using the program in a classroom violates the civil right of a
student not to be swindled and browbeaten by his teachers and
university researchers%1.  Considered as a scientific experiment
involving human subjects, it violates every pertinent code of ethics
and may violate Massachusetts or Federal law.

	I fear that such abuse of questionnaires
and computers has become widespread in the social sciences, but it is
rarely described as clearly and concisely as Messrs. Gross and Kahn have
done.  Since the program is dated 1970, one can hope they have recovered
some obectivity, honesty and fairness.  Still it would be interesting to
know how many high school classes were subjected to this computerized
indoctrination.

	The reader may think it foolish to have put
even this much effort into attacking a high school students' hack.
Maybe no class was ever subjected to it, and if one was,
maybe the students were not intimidated, thought it was silly,
and said so.
My excuse is that it is rare and refreshing to find all these common
intellectual crimes concentrated in two pages and
a computer program whose unambiguous behavior leaves no room for
the authors to claim they were misunderstood.
.skip 3
.nofill
→John McCarthy
→Computer Science Department
→Stanford CA, 94305