Path: utzoo!yunexus!unicus!craig
From: craig@unicus.UUCP (Craig D. Hubley)
Newsgroups: sci.bio
Subject: Re: Intelligent Parrots, or Self-deception and Gullibility.
Message-ID: <2317@unicus.UUCP>
Date: 10 Mar 88 00:21:26 GMT
Article-I.D.: unicus.2317
Posted: Wed Mar  9 19:21:26 1988
References: <1988Mar4.162334.18184@utzoo.uucp> <4299@blia.BLI.COM>
Reply-To: craig@Unicus.COM (Craig D. Hubley)
Organization: Unicus Software Inc., Toronto, Ont.
Lines: 80

In article <4299@blia.BLI.COM> heather@blia.BLI.COM (Heather Mackinnon) writes:
>One of the key points that has returned to me over and over in the study
>of biology is that all living things are more similar to all other living
>things than they are different.  We are more similar, genetically and
>at the cell level, than we are different from garbanzo beans.
>
>...
>
>Should we be surprised to discover that we are very similar to all of the
>other animals who have evolved along with us; that we share their "bestial"
>traits and that they share our "human" ones?

No, we shouldn't be surprised.  However, *which* of the myriad similarities
between organisms is responsible for a particular behaviour is a matter
for sustained and rigorous research.  To say that Clever Hans can add or
Koko can talk is easy.  Adding and talking are easy things for humans to
appreciate.  To say that they are responding to subtle cues is difficult:
Which subtle cues ?  Why can't the researchers control their cueing ?
Do you mean to say that *humans*, with their free will, respond that way too?
It moves the assumptions from inside the animal's head, where they are hidden,
to the relationship between the animal and researcher, which is more easily
scrutinized and less easily characterized.  In short, it opens a can of worms
that were neatly hidden, though still there, before.

That is, ascribing Koko's `eye hat' to a sense of humour or Nim's long
nonsentence to a desire to communicate urgency is cheating, in the scientific
sense.  There is no proof of that.  The fact that human children, at any age,
do not utter such sentences as "me give me me orange eat give me orange",
but rather quite different constructions, suggests strongly that the 
fundamental processes involved may be quite different.  I will give apes more
`credit' (if being like humans is credit!) than that:  I think that Koko knew
what an eye was, and what a hat was.  I think that Nim knew what `me',
`give' and `orange' meant.  But my dog knows `Heili', which is her name,
`walk', and `cookie'.  If I say "Heili cookie walk", then she will most
definitely understand me fully, though in fact I have said nothing in English.
I suspect that a similar fault underlies animal language constructions.
There may simply not be enough analytical/synthetic `firepower' to create
abstract linkages, though there is strong evidence that linkages to concrete
objects and even events are made and understood just as in humans.

I furthermore don't doubt that animals can deliberately lie.  Playing dumb,
however, is a phenomenon that many of us have observed in our pets.  For an
animal to have some rudimentary sense of responsibility and blame does not
to me seem necessarily indicative of a great deal of mental firepower.
If it did, then I doubt it would be ingrained as deeply in human culture
as other rudimentary behaviour such as sex.  To give an animal the ability
to finger-point and *directionally* shift the blame indicates more, but
`Kate break sink' suggested that there was little if any understanding of
the plausibility of the lie.  To stack up boxes to reach a hanging banana
seems to be slightly more intelligent behaviour than a rather dumb lie,
and we already knew that apes could do that.

I think what many anthropologists and some biologists tend to miss is that,
for a given phenomenon, there are dozens of possible explanations, not all
of which anthropologists or biologists are qualified to enumerate.  The
statistical significance of the utterance, that is, was `eye hat' the only
one of a dozen two-word utterances that made any sense, is also often not
closely enough considered.  Some such statistical mistakes are deeply
ingrained in human thought, even in the `estimates' of trained statisticians.
Suffice it to say that complete charts of all of the ape utterances over
an extended period of time, complete with a ratio of the total to the number
that make sense, are usually not included with such a study.

Furthermore, many of the researchers are *not* trained to avoid cueing:
quite the opposite, many have worked with young children, where positive
cueing is usually desirable.  No doubt it encourages a secure atmosphere
in which to learn; but is it learning ?

Marvin (Martin?) Gardner, the Scientific American columnist, has written
a book on ``Science: Good, Bad and Bogus'' with a chapter where he outlines
the objections to this research in some detail.  Suffice it to say that all
of the points I've seen on this issue in this group are covered there,
along with guides to the primary and review literature.

I would love to see some real proof.  But the key word is `proof' not `love'.

	Craig Hubley, Unicus Corporation, Toronto, Ont.
	craig@Unicus.COM				(Internet)
	{uunet!mnetor, utzoo!utcsri}!unicus!craig	(dumb uucp)
	mnetor!unicus!craig@uunet.uu.net		(dumb arpa)