[From Bruce Abbott (951210.1715 EST)]
[From Dag Forssell (951209 2110)]
[Bruce Abbott (951203.1645 EST)]
You chose to not answer my post [Dag Forssell (951128 1420)].
Perhaps there was something wrong with my post from your
perspective. I have noted that skilled debaters simply change the
subject rather than concede the point when they find themselves
cornered.
So THAT's why so many of the key points I have made on CSG-L were never
replied to or commented on!
I can reply to that post paragraph by paragraph, if that is your wish.
However, in considering the argument you presented therein as a whole, I
realized that the reason for our difference of opinion concerning what label
to apply to psychological research might hinge on how one defines science,
and that the resolution of the issue might be achieved more efficiently by
presenting what seemed to me to be our different critera, and then
evaluating psychology's status by those criteria. It's not that there was
something wrong with your post; it's just that the argument it presents is
pretty much irrelevant if our dispute comes down to definitional differences.
Comments on your comments:
I think this makes sense. You offer two sets of criteria, your own
and mine (which involve structural models). I have no quarrel with
your portrayal of the criteria.
Good! Do you accept my criteria for scientific work, or do you hold
strictly to yours? I looked through your whole post but did not find an
answer. Without an _agreement_ on criteria, we cannot resolve our
differences through any argument.
Let us examine Reinforcement
Theory against your own criteria to see if it qualifies as science
on your own terms. I shall add emphasis by capitalizing some of
your words.
Wait a minute! You are going to judge all psychological research by your
assessment of reinforcement theory? So this is to be a critique of
reinforcement theory and a discussion concerning whether psychological
research in general qualifies as scientific work. Is this justified?
Bruce, I think you leave out as a criterion that the "scientific
method" must be appropriate to the nature of the subject matter
under study. To me, this is major.
And by what crystal ball do you determine, in advance of studying the
subject matter, what methods will be appropriate? How do you know that
certain methods will be "inappropriate" for studying the behavior of living
systems? You are viewing the situation with 20-20 hindsight, after a model
has been developed which implies that certain standard methods of scientific
research will yield misleading results.
If the "scientific method"
were appropriate to the subject under study, the rest would follow
naturally and yield useful results.
You mean like programmed learning, behavior modification, and systematic
desensitization? Or do you mean systemmatic, replicable functional
relationships such as those obtained in many operant studies?
I have heard before that psychologists are proud of and put great
stock in the "scientific method". This is a major justification
for considering psychology science, I agree. However, PCT shows
and I think it has been amply demonstrated to you on CSG-L that the
"scientific method" as commonly applied to inert matter is applied
in an inappropriate way when used by life scientists to study
individual living organisms. By mis-applying the scientific method
^^^^^^^^^^^^
in the customary way, psychologists learn everything about the
properties of the environment around the organism (an environment
the scientists have themselves created and carefully control), but
nothing about the properties of the organism. Already, psychology
fails to qualify as science.
Note how everything you say depends on HAVING the theory which research
efforts are aimed at discovering. You use the prejudicial term
"mis-applying" (marked in you paragraph above) when you mean "applying"; it
is mis-applying only in light of a developed theory. I'm hearing an
engineer talking, for whom other people's discoveries and insights,
accumulated over a very long process involving a lot of blind alleys pursued
and finally abandoned in favor of more fruitful formulations, makes possible
the strategy of building a precise structural model employing established
physical laws. An area of investigation that is still in the process of
discovering them is just not science to you.
Only after such conceptions have been developed can the sort of work you
hold as the only truly "scientific" work be conducted.
This is tragic for mankind. Never mind the status of psychologists
as scientists or not. Educators, parents, leaders, politicians,
counselors and anyone who tries to serve his fellow man has been
sold a bill of goods: misleading falsehoods. The world is WORSE
off with contemporary psychological "science" than it would be if
these "scientists" had the insight (which many do) and the
intellectual honesty as well as intestinal fortitude (which they
don't) to simply say: "We don't know".
This is pure polemics: a series of unsupported assertions which reveal the
depth of your own prejudice concerning psychologists and their work. Dag,
do you know how this sounds?
Your own (like Thorndike's) descriptions of various observations
are not merely observations. You habitually include a vast number
of assumptions and interpretations contained in your language and
descriptions. When the unstated assumptions prove false, as most
do when pointed out and examined, the overwhelming majority of
psychological observations by yourself and others fall.
My descriptions were intended to illustrate how the observations would be
interpreted theoretically. This is why they included assumptions and
interpretations (and not a "vast number," by the way). What unstated
assumptions proved false? You must have been reading something I didn't
see? Where was any proof of such offered? Where were the assumptions
pointed out and examined? These vague charges are inventions designed to
support your case, methinks!
You have had an incredibly difficult time to explain just exactly
what the theory of reinforcement is. You have demonstrated that
EAB types don't know themselves. Recent contributions by Samuel
Saunders and Chris Cherpas make it clear that there is a great
multitude of such "theories" and that they are made up on the fly,
ex post facto, as needed to "explain".
I've been working on a post entitled "On Reinforcement Theory" that will
deal with this issue head-on. Synopsis: What we have been calling
"reinforcement theory" is not a finished theory ready to be precisely
applied to specific cases; it is a theory-in-development. Processes have
been inferred from empirical evidence; these basic processes are assumed to
underly any observed behavior (and these processes include more than
reinforcement). However, the quantitative rules through which those
processes act to influence observed behavior remain to be worked out; this
is the focus of basic research in the area. Different proposals have been
made and submitted to empirical test; these tests have ruled out some
proposals while producing data consistent with others. This is why there
appear to be a "great multitude" of theories. But they are not made up "on
the fly, ex post facto, as needed to explain," as you put it. New models
are developed to account for extant data, and are then used to generate
hypotheses that can be submitted to empirical test. This is how science is
supposed to work.
Rick is challenging you to come up with a test at long last.
Apparently there have not been any attempts to falsify
Reinforcement Theory to date. It appears that this is because
there really is no such thing as Reinforcement Theory. This again
disqualifies EAB as science.
What you have not understood is that reinforcement theory is a theory under
development, not a finished product that can be used to make quantitative
predictions over a wide range of situations. The basic processes assumed to
be at work are based on, and supported by, empirical observation. Work in
this area has taken the basic processes as established, and has concentrated
on developing the quantitative rules by which they function. This is the
kind of work Kuhn referred to as the "pick and shovel work of science,"
aimed at developing and extending understanding within the framework
provided by the current paradigm. If this kind of work disqualifies EAB as
science, then it disqualifies all work, past and present, devoted to
elaborating the current theoretical structures in any field whatsoever.
So far, reading and skimming the great number of posts on
Reinforcement Theory, Ex-EABers have made it very clear that
discarding Reinforcement Theory is not an option. All you need to
do is to ignore any problems and reword the verbal statements. It
emerges that Reinforcement Theory is not capable of disproof. If
an apparent disproof shows up, it is set aside, as you do in the
following:
This is a completely unfair and false characterization of what we have been
saying and doing. Speaking for myself, I have already discarded
reinforcement theory as you understand it. What I am seeking at present is
an account of how the processes and observations of EAB can be understood
and explained by PCT. I don't care if every one of them turns out to be a
"side-effect" of control (irrelevant or relevant) or a mere "description of
the environment" (although I doubt that will happen). However, the several
processes identified in EAB research are not all represented in PCT, and a
number of phenomena identified in EAB research may prove valuable in guiding
the development of HPCT. I want to see how PCT and these observations relate.
Your statement above -- "Well, these guys aren't reinforcement
theorists, they're just humble experimentalists reporting their
data." -- offered with an apparently straight face is a sad comment
indeed. Surely these are psychologists with PhD's??? Was this
research report reviewed by a committee of expert peers before
publication? If so, did they not understand either?? Is
Reinforcement Theory so obscure and convoluted that a division of
labor as you indicate is called for? It comes across as a giant
cop-out in a way that I think unthinkable in the physical sciences.
My statements may make more sense to you if you recognize that the theory as
developed thus far is known to have a number of problems, which have led to
a great deal of activity in an effort to identify the sources of those
problems and, if necessary, modify the theory accordingly. But the
particular researchers we were talking about had other fish to fry, and did
not wish to do more at that time than note the problem.
Bruce, you never commented on this. Here, clearly, is an instance
of what looks like disproof. You duck, look the other way and
continue the discussion with the next subject. There is something
intellectually dishonest about this. The name of the game is
denial. This time you disqualified your peers as scientists by
your own criteria.
I am sorry if I overlooked commenting on this; I am not trying to "duck"
anything.
Disproof is in the eye of the beholder. If the theory as it currently
stands is still very "loose," there could be many reasons why the data do
not appear to conform to expectations, which would require an extensive
investigation to support or disconfirm. There is nothing intellectually
dishonest about this at all, and no one was denying anything. When data do
not appear to fit theory, this could mean that (a) the theory is wrong, (b)
there is something wrong with the data, (c) one or more assumptions of the
theory have not been met under the conditions of the test, or (d) it is
unclear how to apply the assumptions to the current situation and thus make
a clear prediction. The fact that an apparent discrepancy has arisen should
be a cause for concern (as was expressed by the researchers), but until
confirmed through further experimental work it is not disproof. Scientists
of all kinds behave this way when the application of a theory to a given
area is uncertain.
So, the definition of reinforcement disqualifies reinforcement
theory as science.
Reinforcement theory, EAB-style, BEGINS by assuming that some consequence of
behavior acts as a reinforcer under given circumstances. Any predictions
(usually qualitative but not always) are then based on the assumption that
the reinforcer is functioning as such. Similarly, application of control
theory BEGINS by assuming that a perception is being controlled. The way to
find out is to conduct tests. The way to find out if a given consequence of
behavior is acting as a reinforcer is to conduct tests. The definition of
controlled perception disqualifies control theory as science, by your logic.
Dag, it seems to me that our dispute is a dispute over
criteria. I would hold that anyone engaged in any part of the
activities I have listed above is doing science. . . . I
hold that most of what goes on in conventional psychological
research falls into one or more of these activities and
therefore qualifies as scientific work.By your own criteria, I disagree with your conclusion.
But you have not applied my criteria to anything but testing reinforcement
theory, under the false assumption that it is a well-developed quantitative
theory rather than a paradigm, within which a final theory with all the
necessary bits and pieces has yet to be achieved.
Your statement of my criteria is good enough. These apply to
modern physical science.
No, these apply to a _mature_ science, one in which the major theoretical
questions have been answered, so that a large body of established principles
can be applied with strict mathematical logic to make predictions applicable
to novel circumstances. Here, the basic rules themselves are no longer much
in doubt; the applications usually can be made to yield precise predictions
(all the important variables and their rules of action are known), and
therefore, when even a small discrepancy between data and theory appears, it
is a cause for grave concern about the foundations of the science (e.g.,
Newton vs. Einstein).
Yet even as we speak, there are physicists trying to understand how to make
relativity theory and quantum theory work together. The fundamental nature
of the universe is in question and there are alternate models: superstrings,
10-dimensional particles, tachions, magnetic monopoles, "dark matter,"
gravitons, singularities: a veritable zoo of proposed entities designed to
account for the appearances in a way that provides an internally consistent
theoretical picture. They are as confused as reinforcement theorists, and
just as prone to invent novel principles to explain the discrepancies in
their data.
Now I still think we were doing science, there in Glass
Science, back in 1973. By your criteria we were doing mush.
But if this is mush, where does the knowledge come from that
eventually allows one to construct the precise mathematical
models that are the sine qua non of an advanced science? It
would seem to me that there have to be quite a few "mushists"
at work providing the foundation so that the "real" scientists
can come along and do the real science.
As I see it, you were doing Alchemy. As I have said before,
Alchemy works. You work with physical ingredients, following
recipes that are carefully recorded and with experiments that are
replicated -- in mass production, no less.
The engineer speaking again. If you can't work it out with a calculator
(actually, a slide-rule came to mind, but I know that's outdated), it ain't
science. What nonsense. That disqualifies most of modern science.
The results are subject
to description in the form of physical measurements, not just free-
wheeling subjective interpretation in words alone, as when you
"describe" the struggles of cats or geese.
Again, you insist that quantitative observation is the only type that would
qualify as truly scientific. That leaves out most of the life sciences, and
much of geology and palentology as well. As for "free-wheeling subjective
interpretation in words alone," one tries to be as objective as possible.
Where do you get the idea that scientific observations in this area are
typically "wild, free-wheeling" and riddled with subjective interpretation?
Have you ever actually READ any of this research, or is this based on my
loose discriptions of it? Even granting that description should be as
neutral as possible, no observation is free of interpretation -- simply
choosing what to observe or measure involves an assessment of what is
important to observe or measure. Furthermore, most of what goes on in EAB
involves quantitative measurement, so your attempt to characterize EAB
research as involving this sort of practice misses the mark.
Most significantly, the basic scientific method, the same one used
by behavioral scientists and laid out in your book (as I understand
it from Rick's review) IS APPROPRIATE for the study of inert
matter, such as glass. Therefore, I have no quarrel with your
Glass Science.
When the basic scientific method is appropriate, as in Glass
Science, you make headway.
Again I ask, how do you tell what is appropriate in advance of the theory
you seek to develop? The THEORY tells you what is appropriate. Until then,
you just try out the standard methods that have seemed to work well in the
past. If these generate systematic results that make sense, that justifies
continuing their use.
When the basic scientific method is inappropriate, as in
conventional psychological "science", you make mush.
I remind you that the inappropriateness of these methods to all
psychological research questions has not been established. Believing it to
be true does not make it so. Phil Runkel (_Casting Nets_) noted that there
are research questions in psychology for which traditional methods are
appropriate. Do you disagree? If so, please state the basis for your
conclusion.
This is getting too long and taking too much time. In this section
you continue to apologize for the failures of by saying:Who would want less, if this is possible? Yet even when there
is as yet hardly a clue as to how a model can be constructed
of a given system, it is still possible to go a fair way
toward reaching those scientific goals, although necessarily
the results will be much more limited.You refuse to draw the uncomfortable conclusion that the limited
results tell you nothing about what you purport to study, but only
reflect the experimental setup. This is what you get from using
the scientific method INappropriately.
Sorry, I missed the argument there. All I see is the conclusion, which so
far as I can see is based on ignoring the whole thrust of my argument here
and simply reasserting your position. This is what people do when they
can't deal with an argument, so I'm told.
Perhaps what we need is to recognize that there is a continuum
between what you call science and the pure mush of unconfirmed
speculation, untested (and unstestable) theory, and anecdotal
evidence.Yes, there is such a continuum in Alchemy, for example.
But Alchemy is not normally considered science, so you are not actually
accepting my argument.
Much of what goes on in normal, everyday scientific work is
not the testing and refinement of quantitative physical
models, but it is still science in that it employs the methods
of science to establish a body of knowledge.No, your statement applies to psychological science, and it does
not employ methods of science appropriately -- it cannot establish
a valid body of knowledge with which one can form a continuum.
Another reassertion of your invalid conclusion. You have not established
that psychology does not employ the methods of science appropriately, you
have only asserted it. I defined several "methods of science," all of which
I argued constitute appropriate scientific activity during different phases
of inquiry. You have ignored this argument intirely in order to focus on
the testing of well-developed, mathematically rigorous theory, which you
believe to be the only truly scientific activity.
Meanwhile, here are all these researchers in psychology,
facing a system of enormous complexity and lacking a means to
anything more than manipulate variables and observe the
result.It is of enormous complexity only because they are ignorant of 20th
century developments and refuse to accept new ideas which are at
variance with what they already believe -- a perfectly
understandable control phenomenon, but tragic in its consequences
for mankind.
Do you imagine that PCT has solved the whole problem? Not even close! Do
you imagine that all researchers in psychology study the reinforcement
process? Wrong! Reinforcement "theory" (such as it is) is about as central
to psychological research as, say, Boyle's law is to particle physics.
Furthermore, such theory as exists in EAB began its development long before
there was any hint that systems engineering principles might provide any
help in understanding the "laws" of mind and behavior, and so there was a
loss of contact between these two developing areas of knowledge. It is now
evident that principles discovered by observing the behavior of simple
mechanical systems can indeed be appied with powerful results to the
understanding of mind and behavior but, because its potential usefulness was
not understood (again, because of lack of contact between the fields during
the development of these ideas within engineering), research training in
experimental psychology did not include the required engineering background,
and so researchers in this area are poorly equipped to handle the required
analytic techniques and by and large do not understand the logic.
As I've said before, this is a sad state of affairs that needs to be
corrected, and soon.
Please note that I am not defending this state of affairs,
although I think I understand how it came about, nor am I
arguing that descriptive theories (like Kepler's) are on a par
with mechanistic ones (like Newton's).But you do nothing but defend it in post after post. You also
insist on respect for this state of affairs and suggest that the
results of EAB research should be respected and can serve as
starting points for proper (PCT) analysis.
What I am defending is not reinforcement theory, but method. You have
bought into the argument that nothing of value could come out of EAB
research because it is not based on control theory and therefore uses
inappropriate methods that can only yield nonsense. You are wrong. Very
little research in EAB is designed to investigate the reinforcement process
per se; reinforcement is used in most procedures only to provide the
motivation for responding, so that other questions can be investigated.
Because you have only heard us talking about reinforcement theory, you take
it to be focus of such research, but there are many other processes to be
investigated, and these have nothing directly to do with either
reinforcement or the dynamics of established control systems. Herrnstein's
experiment on concept learning in the pigeon is an example. There, what is
being investigated is what pigeons are capable of perceiving. Such
information will remain useful long after the control-theory revolution.
There has been a whole series of events in the history of
psychology which has prevented it from moving from descriptive
and quasi-mechanistic theory to physically-grounded
mechanistic theory, even though the information needed to move
from the former to the latter has been around for a long time.I don't know about a whole series of events. Lack of a proper
concept stands out as a key reason.
Yes. After the proper concept has (finally!) been discovered, everything
else begins to fall into place. The weight of your criticism of EAB is that
it failed to identify the right concept. If HAVING the right concept is
what makes a field a science, then particle physics has a long way to go.
The difficulty is that even the WRONG concept can SEEM to work, and
difficulties encountered along the way are then seen as puzzles to be solved
within the paradigm rather than as refutations of it. It is only when
something better appears that the old concept can be abandoned, and in
retrospect appears obviously wrong, perhaps even absurd. But where does the
RIGHT concept come from? In this case, it gets imported from another field,
where a secure foundation of established principles made development of the
concept at this point in time possible.
Another is control for
prestige, an emphasis on BEING RIGHT as befits famous scientists
and esteemed professors, as opposed to the much humbler, and much
more scientific, emphasis on UNDERSTANDING CORRECTLY.
Now we're making unsubstantiated charges. You don't really know what
motivates these guys to behave as they have; you're just making up plausible
stories that put them in a bad light.
But I am arguing that much of what scientists do in most
fields is not restricted to building and testing such models;
where you see this kind of activity most in evidence is within
the applied fields of well-developed sciences for which the
required foundation of basic theory already exists. In those
areas the question is not so much trying to understand new
phenomena for which existing theory fails as in seeing how to
apply existing theory to specific cases. For engineers used
to doing only the latter, anything less may seem like mush.
What you may not understand is the amount of effort by
scientists doing the kind of WORK YOU WOULD NOT DEFINE AS
SCIENTIFIC THAT HAD TO BE COMPLETED before the magnificent
theories you see as the only product to true science could be
conceived and tested.Again, an apology for misguided EAB research.
No apology necessary. How does one know that a given direction is misguided
until one has found a better guide that leads in a different direction.
You're saying that competent work is incompetent if it ends in a blind
alley. How do you know it's blind until you get there?
I am sure there has
been a lot of gruntwork in the physical sciences before today's
magnificent theories were created in the minds of clear thinkers.
And if that work employed systematic observation, experiment, hypothesis,
and testing properly, it was scientific work. Do you imagine that these
theories sprang up in the minds of clear thinkers without the groundwork
being laid first by all these efforts? There is a reason Newton came AFTER
Copernicus, Galileo and Kepler.
You can count all of mankind's experience since the dawn of time in
this category. But once the clear theories are developed, tested
and found to work, the old efforts are laid aside and forgotten.
Yes, and this will happen to reinforcement theory soon enough. Most
researchers in this area havn't even heard of control theory, and most who
have, have not understood it correctly. They're not going to lay aside the
current viewpoint until you can get them to them understand the new one and
demonstrate that it works as advertised.
Reinforcement Theory is not something that had to be completed
before PCT could be conceived and tested. Reinforcement Theory is
a misleading aberration. It will be forgotten, and the sooner the
better.
It could not be laid aside until a better conception appeared. And yes,
it's on its way out, although most in the field do not know that yet.
It sure seems to me that the overwhelming evidence shows that
Reinforcement Theory and its EAB application do not live up to
Bruce Abbott's criteria for science, and that the mis-application
of the scientific method itself further disqualifies them.
You didn't apply my criteria; you reapplied your own. Furthermore, your
misanalysis depends on a misconception of the current state of reinforcement
theory and of the research efforts underway to further develop it, as well
as an undue belief in the centrality of reinforcement theory in
psychological theory and research. No field of science is immune from
making theoretical mistakes, but this does not disqualify the whole field
from counting as a science. If it did, then chemistry would be out because
of phlogiston, polywater, and cold fusion. Furthermore, the rejection of
phlogiston did not render the phlogison-inspired research on heat and
temperature useless, nor will PCT's overthrow of the reinforcement concept
render most EAB research findings irrelevant, let alone most psycholgical
research findings.
The distinction you make is the basic difference, not between science and
mush, but between engineering (where the laws are known and only how to
apply them to a given problem remains to be worked out) and science as it is
generally practiced (where the laws so dear to the engineer must somehow be
discovered). Psychology uses the methods of science to discover basic
principles and develop an organized body of knowledge about its subject
matter, and thereby qualifies as a science, except perhaps by an engineer's
criteria.
Regards,
Bruce