[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Comment on Brochu's Repeatability (a bit long as Ralph would say)

----- Original Message -----
From: "chris brochu" <cbrochu@mail.fmnh.org>
To: <dinosaur@usc.edu>
Sent: Tuesday, November 28, 2000 1:04 PM
Subject: Re: Kirkaldy and SVP Ethics Statement

[Again] toward the end of the message Chris Brochu states:

> [A]s a
> working scientist, I am forced to recognize the importance of data
> repeatability.

Repeatability is desirable in science, but it is not always possible in the
way described in the earlier contributions to this discussion.  Here Chris
Brochu uses repeatability to refer to repeated observations of the same
specimen by different individuals at different times.

Sometimes within paleontology repeatability is limited by the nature of the
materials we study.  One example is a trackway that was described by
Gillette and Thomas in 1989 and reinterpreted by me in 1992.  Yes, I went
and examined the trackway, repeating the observation.  However, Martin
Lockley, who reviewed my manuscript suggested that the trackway had
weathered significantly in the 3 or 4 years between the first and second
observations, and if so, by now the trackway may be almost gone, preventing
further observations-ending repeatability.

Sometimes within paleontology repeatability is limited by unforeseen
circumstances.  As has been noted over and over, the type of Spinosaurus is
gone, and we cannot make further observations of the type specimen to
confirm or extend Stromer's description.

Biology is replete with examples of unrepeatable observations.  For example,
consider a herpetologist trudging through jungle thickets and happening upon
a pair of rare leaf-nosed yellow-bellied hump-backed lizards deeply immersed
in courtship and mating.  He observes the lizards, records his observations,
writes it up, and publishes the observations.  Are those observations
repeatable in the sense used above?  No, unless one could somehow find the
same pair of lizards involved in the same behavior.  Thus, they are
unrepeatable observations, yet volume after volume of Copeia, Condor, and
other such journals contain just such anecdotal observations of animal
behavior.  The medical literature, too, is full of case studies of the
course of a disease or other condition in a single individual, which again
are unrepeatable.  What about physics? just check into the literature on
rare and fleeting ball lightning--unrepeatable observations.  All the above
sorts of observations are regularly published in peer reviewed scientific
literature, yet they are unrepeatable in the sense that later scientists
cannot go back and observe the particular incident described.

There is, however, a second sort of repeatability in the multiple
observations of similar incidents or phenomena by independent observers.  If
27 different observers see 27 different examples of the courtship and mating
behavior of the rare lizards, then the accumulated evidence of the multiple
observations of similar behaviors in similar lizards serves the same
function as multiple observations of behavior in a single incident.
Mistakes and misinterpretations in some descriptions can be identified and
corrected.  The same is true of the accumulated evidence from similar
medical case studies or from multiple separate observations of ball

The same is also true in paleontology.  Although it is desirable to have a
specimen in a museum collection so that that specific specimen can be
reexamined by other researchers, it is not absolutely necessary to the
pursuit of our science.  Examinations of other similar specimens will allow
us to repeat the observations made by other describers and identify and
correct mistakes they may have made.  All we need to do is keep finding more
specimens, and keep studying all the specimens we can get our hands on.

A number of people have argued that specimens in private hands cannot be
restudied by later scientists, therefore precluding the first type of
repeatability.  However, I am not convinced that that is the case.  In my
experience, private individuals holding specimens have always been willing
to permit examination of their specimens.  Moreover, I know of no case other
than the one Archaeopteryx specimen in which a specimen in private hands has
apparently been permanently lost.  Perhaps people with first hand knowledge
of instances where they have been prevented from studying specimens in
private hands can report their experiences.

It is true that individuals holding scientifically important specimens may
sell them, but there should be no trouble tracking down the buyer and there
is no reason to think that the buyer would be less likely to permit
scientists access to the specimen than the first owner was.

It seems to me that the discussion is a half-full vs. half-empty glass
situation, and I think it is half-full.  Even one bird in a bush is better
than none!

I am aware that other individuals may disagree with my assertion that it is
not absolutely necessary to be able to repeat observations of a specific
specimen.  I think anyone who has been following the discussions will know
that certain [perhaps many] individuals feel it is essential that a specimen
be forever available in a museum collection for it to be considered by our
science.  Given that the various viewpoints should be clear to individuals
who have followed these threads, I encourage individuals to reply with
observations and/or constructive criticism that they feel might advance the
discussion rather than merely restating their position.