Dear BEE-L subscribers,

On 19 February I provided a note about “confirmation bias” and the 
problems it provides in scientific research (reason vs belief).  On the 
21st, Bob Harrison picked up on that point and requested input about 
statements he had read in a book about honey bee communication.

On the 23rd I responded to his query, with examples about various 
treatments attempted against varroa mites (usually a wish to believe, 
again).  Beekeepers may embrace a treatment that seemed to work for 
someone else but remain silent when their own colonies crash with that 
treatment.  Then I gave some examples of how confirmation bias has 
confounded events in the bee language controversy.

As expected, my response to Bob’s query in terms of confirmation bias 
elicited a spirited exchange of opinion about the reality of bee 
language — in some cases, striking examples of confirmation bias — for 
which I thank.  Those comments sort of proved the point I made on 19 
February.

Belief in bee language is really an emotional issue.  The controversy 
revolves not about evidence but about one’s attitude toward evidence 
and how the evidence was gathered.  If those who believe in bee 
language were firm in their belief system, they would merely shrug off 
counter arguments.

Innumerable times I have been drawn into a debate about bee language 
and eventually learned to ask a simple question, “Do you think it 
conceivable that honey bees DO NOT have a language?”  How rapidly and 
what answer emerges lets me know whether that person is open to 
reasonable debate or locked into a belief system.

During the past several decades we have been treated to one “proof” of 
bee language after another.  One can only ask why sequential proofs are 
necessary if bee language is already “fact”.  (One does not prove a 
hypothesis true; it always remains a hypothesis.)

In the exchanges of this past week a few points deserve clarification.

Accusations were tossed back and forth about whether BEE-L contributors 
had read the original research papers on the subject, but no one came 
up with what rules should apply while reading those papers — such as:

1)  Did the researchers WANT to obtain “positive” results?

2)  Were the experiments a true test of the hypothesis — or could the 
results be interpreted ambiguously?

3)  Did the researchers include results from earlier experiments that 
did not conform to their expectations?

4)  Did the reader of such publications WANT “positive” results — that 
is, did they somehow have immunity from confirmation bias?  (As with 
Chamberlin’s comment: “The mind lingers with pleasure upon the facts 
that fall happily into the embrace of the theory, and feels a natural 
coldness toward those that seem [not compatible with the theory]. 
Instinctively there is a special searching-out of phenomena that 
support it, for the mind is led by its desires.”)

5)  Did the reader carefully analyze the publications to determine 
whether the researchers had sought only confirmation, or used blind 
design, employed adequate controls, and conducted double controlled 
experiments?

6)  Did the researchers provide all the data they had gathered, or had 
they published selected results from bees that performed as 
expected/wanted?

    In that light, a couple of the respondents singled out the radar 
tracking experiments as “final proof” of bee flight paths.  We can 
rightly ask, “Were the above rules adhered to in those experiments?”  
That’s easy to address.  Would they have spent about half a million 
dollars to conduct an experiment that could negate their belief system?

As iconoclast in these matters, I direct BEE-L to my analysis of those 
supposed “conclusive” radar tracking experiments:

beesource.com/pov/wenner/radar.htm

Will bee language advocates open that URL and study the case I made?  I 
would hope so, because it illustrates how well the von Frisch 
odor-search hypothesis of the 1930s can explain honey bee search 
behavior — as included in:

beesource.com/pov/wenner/bw1993.htm ("The language of bees").

Finally, one respondent resorted to demanding answers to a set of WHY 
questions.  Sorry, but those are teleological, not scientific 
questions.  We covered that topic in Excursus TEL in our 1990 book, 
ANATOMY OF A CONTROVERSY.  Quite frankly we may never know why some 
behavior or another exists, as clearly pointed out by Dick Allen in one 
of his responses.

As Bill Trusdale commented, “The problem is, most of the rest of us do 
not really run actual experiments but operate with no control of 
variables, no control group, and no idea just what we have proved at 
the end other than ‘it works.’ The leaps we make from that point are 
astounding. And fun to discuss.”

And now, what about the significance of the results (that no one 
mentioned) of the honey bee genome analysis (170 odor receptor sites 
but none for bee language — despite it being touted as an "instinctual 
signaling system")?

But now I have to get ready to sell at the Farmers Market on Saturday 
(avocados and cherimoya fruit).

Best wishes to you all.

											Adrian

Adrian M. Wenner		(805) 963-8508 (home office phone)
967 Garcia Road			[log in to unmask]
Santa Barbara, CA  93103	www.beesource.com/pov/wenner/index.htm

"Having one view prevail is harmful; it becomes a belief system, not 
science."
					Zaven Khachaturian — 2006

Search the archives often at http://listserv.albany.edu:8080/cgi-bin/wa?S1=bee-l