[FRIAM] Downward causation

Wed Nov 22 23:55:05 EST 2017

Nick,

Whenever I say this it doesn't seem to register.  Pearl, Glymour, Spirtes,
et al have put statistical causal reasoning on a firm foundation.  This
involves learning causal models from observational rather that experimental
data, including data from the past.  Also remember the distinction between
"actual" causation (hitting this jar with a hammer causes it to break) and
statistical causation (smoking causes cancer).

There is an extensive and growing literature on these topics.

Frank

Frank Wimberly
Phone (505) 670-9918

On Nov 22, 2017 9:43 PM, "Nick Thompson" <nickthompson at earthlink.net> wrote:

> Hi, Eric,
>
>
>
> Well, I would like to say that my personal version of the Pragmatic Maxim:
>
>
>
> *Consider what possible experimental effects the invocation of your
> conception has; those effects are the entire meaning of your conception.*
>
>
>
> … means that the causality makes reference to experiments or to nothing
> whatsoever.   The problem is, of course, that strictly speaking that means
> we cannot apply causality to past events, including evolutionary ones.
> That would seem to be overkill.   There is, of course, the comparative
> method and, of course, “thought experiments.”   Nothing in the maxim, I
> suppose, requires me to actually perform the experiment; only to
> conceptualize it.  Seems like mushy ground.
>
>
>
> Nick
>
>
>
>
>
> Nicholas S. Thompson
>
> Emeritus Professor of Psychology and Biology
>
> Clark University
>
> http://home.earthlink.net/~nickthompson/naturaldesigns/
>
>
>
> *From:* Friam [mailto:friam-bounces at redfish.com] *On Behalf Of *Carl
> Tollander
> *Sent:* Wednesday, November 22, 2017 7:10 PM
> *To:* The Friday Morning Applied Complexity Coffee Group <
> friam at redfish.com>
> *Subject:* Re: [FRIAM] Downward causation
>
>
>
> One of the recurring conundrums of teaching.  Finger pointing at the
> moon....
>
>
>
>
>
> On Nov 22, 2017 14:32, "Eric Charles" <eric.phillip.charles at gmail.com>
> wrote:
>
> "Is there any way to put those two things together:  the abduction thing
> and the misattribution thing? "
>
>
> I would head in a different direction. The question is about, how one does
> the attribution; the answer is, most people do it poorly. In a large part,
> the history of scientific *method *is a history of determining the
> conditions under which we allow causal attributions. When I used to teach,
> I illustrated this most directly in my intro-to-behaviorism class.
>
>
>
> That class included a lot of discussion of applied behavior analysis
> (altering the environment of a person in an effort to improve their
> behavioral functioning within that environment). The central challenge is
> that the ABA practitioner typically only has access to the (usually a)
> child for a very limited time, and you don't want to jump to the conclusion
> that your efforts are working when external factors might equally explain
> the change in the child's behavior. We work up from very basic methods of
> increasing confidence. We eventually build up to an ABAB design, in which
> the prospective solution is applied, then removed, then applied, then
> removed. Every time the problem behavior goes away, comes back, goes away,
> and comes back, etc., our confidence increases that our intervention is *causing
> *the improvement in behavior, because it is increasingly unlikely that
> some other factor just so happens to be varying at exactly the same times.
>
>
>
> Part of the process of "becoming" "a scientist" is increasingly the
> sophistication of research needed before you draw such conclusions... or,
> perhaps more accurately, how well you match the tentativeness-vs-solidity
> of your beliefs to the type of empirical evidence in favor of them.
> Eventually one is drawing on a wealth of difficult-to-specify
> domain-specific knowledge in support of any conclusion, but likely
> justifies the conclusion on the basis of the latest bit of crucial evidence
> (the one which, for them, solidifies the pattern).
>
>
>
> Though... suddenly I might have a legitimate response to your inquiry: I
> would hypothesize that people often mistakenly point at the bit of
> information that was crucial to them, rather than the larger pattern that
> the crucial bit of information brought into focus.
>
>
>
> With Murder on the Orient Express on my mind.... Hercule Poirot would
> narrate such a thing explicitly, would he not? He would say "The crucial
> clue in helping me unravel my confusion was X" and then he would explain
> the larger pattern thus illuminated. A lesser detective would act as if the
> clue itself were crucial in its own right - "This is the key!" - even if it
> was a trivial thing on its own, thus committing a dramatic misattribution
> by virtue of not being self-aware of the abduction taking place.
>
>
>
> Did that get anywhere?
>
>
>
>
>
>
> -----------
> Eric P. Charles, Ph.D.
> Supervisory Survey Statistician
>
> U.S. Marine Corps
>
>
>
> On Wed, Nov 22, 2017 at 1:07 AM, Nick Thompson <nickthompson at earthlink.net>
> wrote:
>
> Hi, Eric,
>
>
>
> Thank you, Eric.   OF COURSE, that is what I should have said.  Thank you
> for saying it so excellently.  Peirce did in fact see causal attribution as
> a form of abduction.  I  would hope I would have thought to say it myself,
> if I wasn’t so distracted by the “counter-factual” thang.  But that way of
> speaking makes me CRAAAAAY-ZEEE.  How can something defined in terms of
> something that didn’t happen
>
>
>
> Before you wrote, I was about to get on my “mystery” high horse.  A
> mystery, you remember, is a confusion arrived at when a bit of language is
> applied to a situation where it doesn’t really work.  Causal attributions
> are often falsely singular, in the sense that , we often speak as if  the
> motion of a billiard ball was caused by the motion of the cue ball, say.
> But what we really have to back those attributions up is a pattern of
> relations between impacts of cue balls and motions of object balls.  When
> we step up to the next level of organization, the confusion disappears,
> doesn’t it?  Events of Type A are said to cause events of type B when
> experiments with proper controls show that an increase in the occurrence of
> type B events is dependent upon the previous occurrence of Type A events.
> But to say that any particular Type A event causes a Type B event is an
> abuse of language, a mystery.
>
>
>
> Is there any way to put those two things together:  the abduction thing
> and the misattribution thing?
>
>
>
> Nick
>
>
>
> Nicholas S. Thompson
>
> Emeritus Professor of Psychology and Biology
>
> Clark University
>
> http://home.earthlink.net/~nickthompson/naturaldesigns/
>
>
>
> *From:* Friam [mailto:friam-bounces at redfish.com] *On Behalf Of *Eric
> Charles
> *Sent:* Tuesday, November 21, 2017 6:43 PM
> *To:* The Friday Morning Applied Complexity Coffee Group <
> friam at redfish.com>
> *Subject:* Re: [FRIAM] Downward Hicausation
>
>
>
> What great timing! One of the best philosophy comics on the web right now
> is "Existential Comics." This very week they took a swipe at "causation."
> Here is an adventure of Sherlock Hume: http://existentialcomics.com/
> comic/212
>
> I suspect that the best I can do to contribute beyond that is to try fall
> back on my role of scolding Nick.
>
> Nick *should *be asserting that "causation" is a metaphor. The billiard
> ball are the understood scenario. Billiard balls sitting on a still table,
> unmolested don't move. But if you knock one ball into another ball, the
> other ball move so. When I say something like "The approaching lion *caused
> *the gazelle to move", I am invoking the metaphor that the lion-gazelle
> relationship is like that of the billiard balls. Had the lion not been
> doing what it was doing, the gazelle would not have moved away. It isn't
> simply a "counterfactual." It is an assertion (an *abduction*) regarding
> broad patterns of gazelle behavior that can be readily observed under many
> other situations.** Some of those, I have presumably already seen. Those
> constitute the "basic implication" of the metaphor. Others I have not
> observed, and those constitute potential investigatory events - not
> ethereal thought experiments. As in true of any metaphor, there are also
> aspects of the billiard-ball scenario I do not intend to map perfectly onto
> the lion-gazelle scenario (e.g., the lion and gazelle are not spheres).
>
> So that is where Hume and those like him go wrong. They want to beat the
> billiard balls scenario itself to death. But that's not how metaphors work.
> There is something understood about the billiard balls, and it is
> that-understood-thing that is being generalized to another scenario. Any
> attempt to explain the billiard balls will involve evoking *different *metaphors,
> which would entail different assertions (abductions). There is no
> foundation (Peirce tells us, amongst others), Descartes was on a fool's
> errand: In the land of inference, it is turtles all the way down.
>
>
>
> ** The breadth of the patterns being referenced is, I believe, where
> Frank's point about probability slips in. One could certainly simplify the
> complexity of the assertion by making lumping similar scenarios together
> and speaking about the probability of a certain gazelle behavior within the
> cluster of similar situations.
>
>
>
>
>
> -----------
> Eric P. Charles, Ph.D.
> Supervisory Survey Statistician
>
> U.S. Marine Corps
>
>
>
> On Tue, Nov 21, 2017 at 5:08 PM, gⅼеɳ ☣ <gepropella at gmail.com> wrote:
>
> Also Known As: Beware equating experience with existence.
>
> On 11/21/2017 02:00 PM, Frank Wimberly wrote:
> > Beware the tendency to think that if you can't immediately measure
> something then it doesn't exist.
>
>
> --
> ☣ gⅼеɳ
>
>
> ============================================================
> FRIAM Applied Complexity Group listserv
> Meets Fridays 9a-11:30 at cafe at St. John's College
> to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
> FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove
>
>
>
>
> ============================================================
> FRIAM Applied Complexity Group listserv
> Meets Fridays 9a-11:30 at cafe at St. John's College
> to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
> FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove
>
>
>
>
> ============================================================
> FRIAM Applied Complexity Group listserv
> Meets Fridays 9a-11:30 at cafe at St. John's College
> to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
> FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove
>
>
> ============================================================
> FRIAM Applied Complexity Group listserv
> Meets Fridays 9a-11:30 at cafe at St. John's College
> to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
> FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://redfish.com/pipermail/friam_redfish.com/attachments/20171122/90e8f674/attachment.html>