LINGUIST List 2.769

Sun 10 Nov 1991

Disc: Human Subject Research, II

Editor for this issue: <>

Directory

1. Prof. Roly Sussex, Re: 2.763 Query: Human Research
2. Fan mail from some flounder?, Re: 2.763 Query: Human Research
3. "Bruce E. Nevin", Re: human subjects
4. , Human subjects
5. Stavros Macrakis, 2.767 Human Research

Message 1: Re: 2.763 Query: Human Research

Geoff Nathan's request prompted me to get more precise information
on the human experimentation issue.
The document you need is, I think, US Code of Federal Regulations
45 CFR 46, "Protection of Human Subjects". These, I understand,
include categories of exemptions for social sciences and humanities
work. For instance, in my institution they are discussing, on
the model of the CFR, exemptions for
	1. research in educational settings involving normal
	educational practices with adults and/or children as
	subjects
	2.	Research involving the use of educational tests
	with no persomal identifiers with adults and/or children
	as subjects
	3.	Research involving survey or interview procedures
	using adults (NOT children) as subjects
	4.	Research involving the observation of public
	behaviour
	5.	Research involving the collection or study of data,
	(...) that are in the public domain
	6.	Diagnostic and therapeutic procedures that are an
	accepted part of treatment and are recognized as a current
	practice by the appropriate professional body.
Institutions differ, since they have internal rules, as well as
an obligation to state and federal legal requirements, and
to funding bodies and institutions like the NSF.
In the past a lot of trouble has arisen from treating social science
and humanities research on a biomedical model (the area where a lot
of the regulations have been developed for monitoring ethical issues
in human-related research).
Sensible institutions, on the model of the US exemptions, are trying
to make research a little less onerous at the administrative end.
CAVEAT: I haven't seen 45 CFR 46, and am reporting it at second
hand; so check it yourselves before quoting!
General rule: when in doubt, refer all cases to the appropriate
body. It will take time, but the legal implications of not doing
so can be daunting. There is also a need for a precedent in the
kind of research which linguists do, and it will take a number of
cases going through the processes for there to be a workable
precedent that we can invoke for new research projects.
Roly Sussex
Director
Centre for Language Teaching and Research
University of Queensland
Queensland 4072
Australia

Message 2: Re: 2.763 Query: Human Research

We have to clear all research through our Human Subjects Committee, including
informant work and classroom research. I don't know about research for
term papers, but almost anything else has to go through the process. The
less invasive it is, of course, the easier (and more automatic) the
approval, but it's for the protection of the people involved. For example,
I videotape my ASL consultants, and I get several levels of consent from them:
permission to use their data in my written papers, permission to show their
tapes to students and other researchers, and permission to thank them
by name if they wish. If there is a chance that someone will recognize
them (which is of course easier with video than with audio), they must have
the option to have as much confidentiality as they request. A paper-and-
pencil test is less invasive, but even using preexisting data must be
approved by our Human Subjects committee here; I assume the policy is\
similar at other institutions, since it is federally mandated. If you recall
(I do), this was an outgrowth of abuses perpretrated by people like Timothy
Leary, who gave LSD to undergraduates without telling them. Granted, it
's a far cry from asking for grammatical judgments, but it is imperative
to protect subjects' rights.
Susan Fischer

This is clearly a pervasive issue of the sort that this forum will I
think prove excellent at disclosing for collective action.
I suggest we arrange for an appropriate organ of the LSA intermediate
for linguists with the government about this. (a) State the original
concerns and intent of the legislation (experimental drugs, invasive
therapies, mind-warping procedures), (b) contrast what linguists do
(tape recordings, grammaticality judgements, reading tricky lists that
induce slips of the tongue, mind-warping things like that) (c) state
general caracteristics of research that should be exempted, in the
sublanguage of American jurisprudence (legalese), so as to encompass
linguistic research without need to name linguists as a "special
interest group," and (d) ask that exemption be provided for research
that meets these requirements. All pretty standard bureaucratic "if
<paragraph> fill out form A, otherwise fill out forms B-Z" stuff. I
should be surprised of other professional organizations did not also
have an interest in this, and the presentation would be strengthened by
involving them, though the process would be lengthened. There is a lot
of interest in simplifying bureaucratic procedures. The Document Design
Center (in D.C.) is a good resource--they've been advocates for plain
English in goverment forms and documents for a good number of years, and
they know the ropes and the supportive players.
	Bruce Nevin
	bnbbn.com

I recently applied for an exemption from the Committee for Protection
of Human Subjects at UC Berkeley for work on my dissertation, and
got it. This was for work on my dissertation in which i will be
interviewing people and collecting live data. I never heard about
having to deal with this issue at any level other than for thesis
and dissertation work. This applies to UCSC, UCSD and UCB where
I have done linguistics work.
Jon Aske

In 2.767, Joyce Neu, Penn State University <JN0PSUVM.bitnet> says:
 I am amazed that you do not have subjects sign informed consent
 forms for your research. I am a very strong believer in ethical
 research methods because I think that much research would be
 improved were researchers called upon to provide rationales.... It
 also provides students ... with a good model for how research
 should be done...
The consent forms may well be legally or administratively required,
but could someone please make explicit what is or could be unethical
about asking for grammaticality judgements without them?
To broaden the question: What is the potential harm to the subject,
and who should be protecting them? I can imagine scenarios of offense
(e.g. asking about taboo words or eliciting responses perceived by the
subjects themselves as `substandard'), but are reviews needed to
prevent this? After all, no board reviews a linguist's op-ed calling
for (say) ``the eradication of substandard English spoken by inferior
races''. This would surely (and justifiably) cause much more offense
and even damage (not that I know any linguists who would propose such
a thing).
Is this another manifestation of a lawyer-driven society, where an
institution is forced to take responsibility for an individual
researcher (on the deep pockets theory)?
	-s