IP: Ethics Questions for CMU, Rimm, and Sirbu part 1 of 2 (limits

[David Farber <farber () central cis upenn edu>]

[http://sun.soci.niu.edu/~cudigest/rimm/rimm2]


SOME THOUGHTS ON CARNEGIE MELLON'S COMMITTEE OF INVESTIGATION


            Jim Thomas / Department of Sociology Northern
                         13 September, 1995
           Illinois University (jthomas () sun soci niu edu)




((BACKGROUND: As an undergraduate at Carnegie Mellon University this
past year, 30 year old Martin Rimm published a research project in
the Georgetown Law Journal on "pornography" on the Net.  Time
Magazine featured the project as a cover story in its July 3
issue. The project was attacked for intellectual and ethical
improprieties, and CMU has begun an investigation into the project.
What follows are just a few of the questions I had after following
the project and its media coverage for these past two months--jt))




The Martin Rimm "Cyberporn" study, while thoroughly discredited on
intellectual and other grounds, remains a problem for those involved
in it.  CMU sources indicate that the CMU provost has formed a
committee with the faculty senate to investigate questions that have
been raised about the study's procedures and ethics.  (See CuD #7.58
and 7.59; A complete background, including the full study and a
critique by Donna Hoffman and Tom Novack, can be found on the CuD
homepage links at http://www.soci.niu.edu/~cudigest).


Some might wonder why there exists a need to pursue an investigation
of a discredited undergraduate project. To some, it may even appear
that continued inquiry into its procedures and the background of
Rimm, its "principal investigator," constitute an unnecessary
witchhunt, reflecting a lynch-mob mentality.


Such a view is erroneous and short-sighted.


Continued questioning of the study is not an attempt to "disprove"
or minimize Rimm's "finding" that 83.5 percent of Usenet images are
"pornographic," to deny that there is sex on the Net, or to minimize
the very real concerns of parents and others about limiting
children's access to undesirable material.


An airing of the study and its procedures should be pursued for
several reasons. First, the results of the study continue to be used
uncritically, especially by those who would exaggerate the prevalence
of objectionable Internet material.  Whether the figures approximate
reality is irrelevant. The issue is that there is no basis in *this*
study to give confidence in the figures. Normally, this would be no
more than an intellectual dispute to be resolved by additional
research. However, it is how the data were acquired and manipulated
that cause concern.


This leads to the second reason for pursuing questions about the
research.  There is sufficient evidence to suggest that serious
ethical improprieties underlay the study.  This, too, would normally
be an in-house matter best left to a University and the subjects
involved. However, at least some persons involved in the study turned
an intellectual exercise into a highly visible public media event. As
a consequence, the public has a right to address troubling questions
that subvert both the intellectual credibility and the procedures by
which claims were made and continue to be defended.  Third, this is a
cyberspace issue. It's no secret that researchers have found
computer-mediated communication a rich source for ethnographic and
other data collection.  The apparent ethical improprieties are
relevant to the broader intellectual community, as well as to the
on-line public, to the extent that they suggest several ways by which
scholars and others can go astray in violating basic human subjects
canons.  Finally, if sloppy research based on questionable data and
practices produces claims that are used to subvert First Amendment
rights in cyberspace, and when that research has been explicitly
identified as the product of one of the leading U.S. research
institutions, that institution has the responsibility of assuring the
credibility of that which was done in its name.


Pursuing questions raised by the study is not, or at least ought not
become, a mechanism for public humiliation of the participants or an
attempt to try them in public. Instead, raising further questions
provides exactly what academic scholarship requires:  An examination
of procedures and methods of public claims in a public forum in a way
that allows those expected to accept such claims an opportunity to
assess the credibility and biases of the researchers.  To that extent,
as any good ethnographic scholar knows, questions about how data were
gathered, about scholars' potentially biasing background experiences,
or about interpretative procedures, are of direct relevance to the
public. As a consequence, the CMU Committee of Inquiry might consider
the following questions about the study as a way to facilitate
independent assessment of the research.


                   WHAT WAS THE "CMU/RIMM STUDY?"


Time Magazine's July 3, 1995, issue featured as a cover story a
Georgetown Law Journal article by Rimm ("Marketing Pornography on the
Information Superhighway"), that uncritically reported the study's
findings in a sensationalistic manner (One of the story's fuzzy
graphics depicted a nude male presumably copulating with his
computer).  Although selective readers were given access to the study,
including Ralph Reed (the Executive Director of the Christian
Coalition), three journal commentators, Senator Charles Grassley, who
misrepresented the study's findings to hype his anti-pornography Bill
(S. 892), and Philip DeWitt, the Time writer who had access to the
study as an exclusive, others who attempted to obtain a copy were
refused. The reason: It was embargoed. Rimm claimed that the GLJ
embargoed it, but the GLJ claimed otherwise (see below).


The study purported to be an exhaustive analysis of "pornography" in
cyberspace, and it contained numerous methodological flaws and
demonstrably inaccurate claims (See the Hoffman and Novak critique).
Among the controversial aspects of the study was the implication that
it was a legitimate CMU-sponsored project. In fact, it was an
ambitious undergraduate research project. But, once the CMU/Rimm
connection was made, it became known as the "CMU study" in the media
and in Congress.


                      WAS THIS THE "CMU" STUDY?


Before the study appeared in Time or the GLJ, Rimm  appeared on
Nightline discussing the "CMU study," Senator Grassley alluded to the
"CMU study" (and planned to have Rimm testify in a Congressional
hearing), and those supporting the study (including Ralph Reed and
Catherine McKinnon) referred to the "CMU study." However, a recent
call to Don Hale, Vice President for University Relations at CMU,
said that CMU itself never claimed that the study was done under the
auspices of CMU.  "People were misinterpreting how we described the
study from the gitgo," he said.  And, he does make a compelling case.
Hale explained that he often uses the term "CMU study" as a
convenient shorthand to describe research projects done by faculty.
"I never thought about the implications," he said, indicating that he
would take more cautious steps in the future.  He was convincing, and
there is no reason to doubt his sincerity.  But, his words do not
reflect the actions of others, including some CMU personnel.


When I spoke to several CMU personnel about the study in the first
week of July before the controversy broke, they explicitly and
unequivocally associated CMU with the study.  There is often a thin
line between shorthand connotation and summary denotation, and to my
view, some CMU personnel crossed over that line.


Then, there is Rimm himself. In the study, he explicitly and
repeatedly refers to the study as the "CMU study". In his ABC
Nightline appearance, Rimm and others, including Ted Koppel, called
it the CMU study. The media, including the New York Times, called it
the CMU study. The commentators on the GLJ article called it the "CMU
study".


Despite the resulting brouhaha, CMU did not disassociate from or
officially respond to the study, until it issued a cryptic press
release in Mid-July:


     Carnegie Mellon University is responsible for the
     integrity of research conducted at the university.  As
     a community of scholars, in which truth and integrity
     are fundamental, the university generally examines
     carefully issues raised concerning the propriety of
     research conducted by members of the university
     community, taking due care to protect the rights of
     those members.


     Provost Paul Christiano already has informally sought
     and received advice from some faculty members about the
     study conducted by undergraduate student Marty Rimm and
     published by the Georgetown Law Journal.  He will soon
     form a committee of distinguished and knowledgeable
     faculty to examine in more detail the issues that have
     been raised about the study.  The committee will
     recommend the appropriate next steps, if any, that
     should be taken relative to this study and, if
     necessary, relative to policies on undergraduate
     research.


The release indicates that this was no longer accepted by the school
as the "CMU Study," but rather was now the work of "undergraduate
student Martin Rimm."


                           THE CMU INQUIRY


Later in  July, CMU Provost Paul Christiano adhered to CMU policy in
forming a three-person Committee of Inquiry to investigate whether
there existed sufficient grounds to form a five-person Committee of
Investigation to address allegations of research impropriety.  On
August 8, Provost Christiano issued the following memo:


     The Committee of Inquiry, which was formed to conduct a
     limited inquiry into allegations directed at the subject
     research, now has completed its work.  That committee has
     recommended, in accordance with the above-cited policy, that
     several allegations warrant the conduct of a thorough
     investigation, through a five-member faculty Committee of
     Investigation.  This committee, to be formed jointly by me
     and the leadership of the Faculty Senate, is expected to
     submit its recommendations to me, to the president of the
     university, to the leaders of the Faculty Senate, to the
     dean of student affairs, and to the researchers themselves.


     The specific recommendations that have been provided to me
     by the Committee of Inquiry remain confidential, according
     to the above-cited University policy.  However, I expect the
     Committee of Investigation to examine a full range of issues
     relating to the article and to the research preceding it.


     Until the Committee of Investigation has completed its work
     to determine which, if any, allegations are valid, it would
     be inappropriate for me to comment further on this matter.
     Indeed, all those who believe in fairness and in due process
     should take special care not to prejudge the conduct of
     persons who have engaged in this or any other research.
     While the well being of human participants, as well as the
     search for truth, must always be essential guiding
     principles, so also must be respect for the reputation and
     academic freedom of researchers and for due process.
     Carnegie Mellon University will continue to adhere to those
     principles.


The committee will presumably ask a number of questions to address
public concerns about the study. Among the troublesome questions
include the following:


                         QUESTIONS FOR RIMM


Although the personal biography and life of a researcher can be and
often is of relevance to especially qualitative research, private
lives rarely are of significant relevance. This case is is an
exception. If, for example, a scholar has a history of deceit,
fabrication, or other behavior that raises questions about the
veracity of research, the private history related to such acts
becomes relevant.


In Rimm's case, there seems to be disturbing history of behavior
that would be of special concern to professionals assessing his
credibility.


A few credibility-challenging examples include:  His authorship of a
book listed in the Library of Congress records as:  "How to Exploit
Women, Dupe Men, and Make lots of Money." There appears to be no
public copy available in the LoC, so only excerpts alledged to be
from the book are available. In CyberWire Dispatch, Brock Meeks cites
an excerpt:


     Into my mailbox flow excerpts of Marty's "how to" manual.
     Here is a sample of his turgid prose, taken from the Usenet
     posting, from a chapter on Anal Sex:  "When searching for
     the best anal sex images, you must take especial care to
     always portray the woman as smiling, as deriving pleasure
     from being  penetrated by a fat penis into her most tender
     crevice. The male, before ejaculation, is remarkably attuned
     to the slightest discrepancy; he is as much focused on her
     lips as on her anus. The slightest indication of pain can
     make some men  limp."


*QUESTION: Was this book used to entice the research subjects
(the sysops) to participate in the study?


"Books in Print" information reveals the following information:




      AUTHORS:        Rimm, Martin
                      Speranza, Carolyn; Illustrator


      TITLE:          The Pornographer's Handbook; How to Exploit
                      Women, Dupe Men, & Make Lots of Money


      PUBLISHER:      Carnegie; 03/1995


      EDITION:        Orig. Ed.


      PAGINATION:     67p.


      ISBN/PRICE/BINDING:
                      0962547654;$5.95 pap.


It has been reported on The Well that "Carnegie Press" and Rimm
share the same address and phone, among other things.  One close
observer of the study raised the following questions:


*QUESTION: In what states did Martin Rimm register "Carnegie Press"
as a corporation?  Did he do so in Pennsylvania? In New Jersey? In
Delaware? Elsewhere?


*QUESTION: If Rimm registered "Carnegie Press" as a corporation in
April 1994, or before that date, is there an ethical issue raised by
his intention to profit from grant-funded research on nonconsenting
subjects?


*QUESTION: Is there an ethical issue raised by the choice of the
name "Carnegie Press"?


Was the book satire, was it a methodological access key, or was
it intended as a serious marketing guide? It's hard to tell,
but the following post from Rimm to an anonymous correspondent
raises further questions:


      Date: Sun, 18 Dec 1994 22:02:01 -0500 (EST)
      From: Martin Rimm <mr6e+ () andrew cmu edu>


      To some extent, but the truth is I am ahead of the
      pornographers.  However, with mainstream pornographers
      moving on-line, some of the best software engineers in the
      country are now working for the pornography industry, and I
      expect within a year or two they will leap ahead of me.
      Recently, Kenneth Guarino, of Southe Point Enterprises, the
      largest adult video distributor in the country, hired a
      team from Microsoft.


                        .....................


      Once my study is published, it will be obvious to them why
      such research is useful. In a two hour video, or magazine
      with 100 pictures, pornographers never knew what the
      customers really wanted. Now they can find out. Personally,
      I'm getting out of the pornography business, as I want to
      move on to other subjects on the net.


*QUESTION: Which side of the fence was Rimm on?


Rimm appears to be no stranger to controversy involving deception.
Press reports indicate that, at age 16, Rimm posed as an
Arab sheik to "infiltrate" an Atlantic City casino to "expose"
teenage gambling. A New Jersey news story raises further questions.
An excerpt:


     From the Atlantic City Press, Aug. 30:


                       CYBERPORN RESEARCHER LINKED TO A.C. PRANKS


          * Marty Rimm, author of a controversial study of pornography
            on the Internet, was investigated by gaming officials for
            an alleged publicity stunt gone haywire and other hoaxes
            involving the Taj Mahal casino.
              ____________________________________________


         By RAY ROBINSON


         The Press of Atlantic City Online


        Marty Rimm, author of a widely publicized study of pornography
        on computer networks, was suspected by state investigators of
        pulling two creative -- and potentially expensive -- pranks on
        the Trump Taj Mahal Casino Resort in 1990, according to
        documents reviewed by The Press.


Although such behavior of itself does not necessarily subvert
credibility, it does raise a subsequent question of whether similar
methods were used to gain access to a research setting or produce a
written product that is embargoed prior to publication in a non-peer
reviewed outlet.


*QUESTION: Was the access to confidential information and other data
from sysops or other sources gathered under pretense?


Rimm states explicitly in his study that his "research team" did not
generally reveal to research subjects, the sysops, that they were
being studied (GLJ, 1995: 1878).  CyberWire's Brock Meeks wrote:


     How did Marty pull this off?  Adult BBS operators aren't
     known for their openness and trusting attitudes, in general.
     When I asked Marty how he was able to do what had taken me
     years to do -- develop sources inside this network of adult
     BBS operators --he said:  "[Y]ou didn't have powerful
     software which you could use to convince them that you
     indeed had something to offer. What took you years I could
     do in anywhere from five minutes to two months.  You'll have
     to figure the rest out."


*QUESTION: Did Rimm lie to gain access to sysops and their data?


In his introduction, Rimm lists more than a score of professors,
administrators, and staff as part of the "research team." Some have
disassociated from the study or indicated that they had little, if
any, significant role in it.  What role did the listed members of the
research team play?  Where they full collaborating participants in a
"research team" as he implies?  Rimm's primary advisor, CMU Professor
Marvin Sirbu, in a letter to EFF staff counsel Mike Godwin, even
alluded to several of the professors, presumably two of whom are at
the University of Oregon and one of whom is a Dean, as collaborators
as a means to justify the legitimacy of the study.


Consider the following:


     From: Martin Rimm <mr6e+ () andrew cmu edu>
     Newsgroups: cmu.cs.discussion
     Subject: Re: More Censorship
     Date: Fri,  4 Nov 1994 00:31:15 -0500


     The team of researchers consists of seven professors, three
     deans, four lawyers, two lobbyist groups, six undergraduate
     research assistants, three doctoral students, three
     programmers, and an art instructor.


     Martin


     \enddata{text, 547925302}


     Date: Tue,  4 Apr 1995 21:25:59 -0400 (EDT)
     To: mnemonic () eff org
     Subject: Fwd: INTERNET ADULT BBS STUDY


     We appreciate your interest. We are making every effort to
     get you a complete copy of the study before publication. In
     the meantime, we would greatly appreciate an independent
     check of our legal notes, which the journal helped us with.
     (No one on our team is a lawyer). We need to return our
     final edits to the journal on Thursday, April 14. If before
     then you have a chance to review the attached, your comments
     would be most appreciative.
     Thanks again,
     Martin Rimm
     Principal Investigator


It is interesting that in the first public post, Rimm listed four
lawyers as part of the research team. In the April passage, Rimm
indicates that no lawyers are members of the team.  Of itself, this
may seem unimportant until one asks whether this was, in fact, an
authentic research TEAM.


*QUESTION: Who, precisely, was on the research team, and what was
their role?


*QUESTION: Was this a legitimate research team, or was it simply
window-dressing used to enhance the study's credibility?


*QUESTION: Was, as Rimm claims, the data actually collected by a team,
or was he, himself, the primary data-gatherer, using the language of
research inappropriately?


*QUESTION: If this was, in fact, a legitimate research team
comprised of administrators and faculty, then what was their role in
the demonstrable deception? If this research was in fact a true
collaborative effort, would that not then also mean that that a
score of CMU personnel are complicit in demonstrably unethical
research?






In the study, Rimm claims to have talked to a number of sysops, both