The danger of "Danger"

Mike Clark mrc7 at cam.ac.uk
Tue Jun 8 13:21:04 EST 1999

In article <7j9h3i$kup$1 at panix3.panix.com>, Ian A. York
> As I've said before, I think Matzinger and Fuchs were right to reject
> "self/non-self discrimination" as THE explanation for the immune response.
> The immune system is not an advertising jingle, to be summed up in a
> catchy phrase and a 30-second sound bite.  They were wrong, however, to
> try and replace self/non-self with their own little advertising jingle.
> There is no reason whatsoever for the immune response to tidily follow a
> single set of rules, and the "danger" theory is trying to squeeze the
> whole thing into a tidy set of rules.  The immune system is an
> evolutionary gamisch; whatever worked has been preserved, never mind
> whether it fit into a paradigm or could be explained in less than 25
> words.  Trying to put the immune system into a paradigm basket is
> oversimplifying to the point of absurdity.  "Sometimes the immune system
> sees self vs. non-self; sometimes 'danger' signals are involved; sometimes
> other things are more important" may not get me interviewed as a radical
> new thinker, but it's a more useful way of thinking about the immune
> system than any rigid approach.

I must say that I agree very strongly with the points you make in the above
paragraphs. I guess the problem, as so often in biology, is that when you
try to explain in as few words as possible a complex system which has
evolved over a long period of time you end up with an absurd
over simplification of the reality. 

> As I've also said, it looks to me as if the "danger" hypothesis is
> becoming unfalsifiable.  Some of the original predictions have been
> falsified (apoptosis as a a non-danger signal; yet apoptotic cells are
> apparently good triggers for an immune response), and far from falsifying
> the paradigm, they've sometimes been taken as *evidence* for it.  I'm not
> at all comfortable with that.

Again to some extent I agree with you on this point. The importance of a
hypothesis is that it should make definite predictions which can be tested
experimentally. If the experiments show that the predictions are incorrect
then this should be taken as falsification of the hypothesis and it should
be rejected. The problem arises that many of the predictions made by
self/non-self and also danger/non-danger have been shown to be falsifiable
and thus both should be rejected.

Of course many scientists play the trick of modifying the hypothesis and
thus generating new predictions but at the same time keeping the same name.
I guess there is no real harm in this except that it does cause so much
confusion when everyone is actually arguing about quite different ideas
without realising it.

So are we talking about 'self/non-self' versions 1,2,3,4, or 5 etc as
opposed to danger/non-danger versions 1,2,3,4, or 5 etc?

> Enough rambling.
> Ian 

Perhaps we should merely use the global and simplified concepts of self
non-self and danger non-danger as useful introductory tools to be used to
teach basic immunology but should then point out that they both must be
rejected as unsatisfactory descriptions at a more advanced level of

Mike Clark,                        <URL:http://www.path.cam.ac.uk/~mrc7/>
 o/ \\    //            ||  ,_ o   M.R. Clark, PhD. Division of Immunology
<\__,\\  //   __o       || /  /\,  Cambridge University, Dept. Pathology
 ">    ||   _`\<,_    //  \\ \> |  Tennis Court Rd., Cambridge CB2 1QP
  `    ||  (_)/ (_)  //    \\ \_   Tel.+44 1223 333705  Fax.+44 1223 333875

More information about the Immuno mailing list

Send comments to us at biosci-help [At] net.bio.net