mutual intelligibility--non-human

Mike Lieber (U28550@UICVM.BITNET)
Tue, 29 Nov 1994 13:47:19 CST

J. P. Scott, best known for his research with dogs, gave an astounding
presidential address to the Society for Behavioral Genetics in 1976, later
published in the journal, _Behavioral Genetics_ in 1977. Entitled "Social
Genetics," this little gem suggested that the whole idea of behavioral genetics
was wrong-headed. Amost Batesonian in its thrust, Scott asserted that what
researchers actually observed was not behavior (whatever that was supposed to
mean) but an animal's response to something in its environment that functioned
as a signal. That is, what people call behavior is not a property of the
animal but a property of a relationship between the animal and something else.
Thus, behavioral genetics is about relationships, most of what researchers
observe being social relatioships. Why not just call it "social genetics"?
[I can see Bob Graber going ballistic, but what the hell.] The suggestion was
so empirically based as to be commonsensical, but so radical in his field that
it has been ignored ever since. It changed forever the way I think about the
term, behavior, however.

Scott illustrated his points with experimental work that he and some colleagues
had done with dogs at Cold Harbor. Using four different breeds of dogs, he
separated some dogs from litters in a way that there were four groupings of
same-breed puppies and another four of mixed breeds raised together. When they
were old enough to train, he paired same breed and mixed breed dogs raised
together and gave each pair certain tasks to complete. Predictably, same breed
dogs did better at the tasks than the mixed breed pairs. Then he started
switching the pairs--a dog raised with same breed with a dog raised mixed breed
and a mixed breed with a mixed breed who were not raised together. The results
were different. The pair with a dog raised mixed breed and one raised with
same breed failed each task miserably. The pair raised in mixed breed (but
different dogs of different breeds) had a bit of trouble at first but then was
able to complete the task, doing better and better in subsequent trials of
different tasks. As far as I know, he didn't continue the trials with the
same breed-raised/mixed breed-raised dogs long enough to see if the pair could
eventually work together. He concluded from these experiments that dogs
raised with familiarity of differences were not fazed by the next expierience
of difference.* For our purposes, the experience of learning how to interpret
signals that are not inherently part of one's own repertoire is one of
developing an eye for pattern. Once an animal (including us) has this
experience, pattern perception becomes more habitual in a way such that
unfamiliar experience triggers a search for pattern in that experience that can
be used as a guide to potential response. The result is mutual intelligibility
and it is clearly not limited to Homo sapiens.

*It is hard to say whether continued experimentation with same breed-raised
dogs would have eventually resulted in all of them learning about mutual
intelligibility. A lot would depend on the particular breed. Spaniels, for
example, seem to go psychotic more easily than Bisenjis and hounds, so
experiments with them have to be broken off sooner to avoid damage to the dog.

Mike Lieber