ambiguity, uncertainty, fuzziness, doubt

greenspun.com : LUSENET : History & Theory of Psychology : One Thread

I am currently writing a dissertation on theory development and am approaching it from what I believe is a slightly different perspective than is traditional. I do not contest the need for precision and accuracy in scientifically rigourous development of theory, but I also believe, for numerous reasons that I won't outline here, that we not only need to tolerate ambiguity, uncertainty (the principle upon which physics balances), doubt (Descartes 'acid' to be poured over uncertainty to produce 'cleaner' ideas) and fuzziness (nonlinear/non-dichotomous variables), but we need to embrace them. A celebration of these concepts rather than a denial of their usefulness, under the 'illusion of precision' is surely called for? Or am I totally wrong? Any suggestions?

-- annette (net_davey@hotmail.com), May 27, 2002

Answers

You may find some earlier discussion here to be (perhaps, only slightly) related to your query. Look below under Older Messages, scroll to History and Philosophy of Science, and link to "Is "emergentism" to psychology what vitalism was to biology?"

-- Roger K. Thomas (rkthomas@uga.edu), May 28, 2002.

Hi Annette, so let me make sure we are on the same page. You're talking about a theory of science here? And the basic elements are public data and propositions, that is statements which are either true or false - right? I mean, what good would a paradoxical statement do you? Take the proposition, "This sentence is false." Now, saying it implies it is true, but is it? It could be true in which case it's false. It could be either true or false or maybe neither. Is that enough ambiguity for you? Usually theory is doing everything it can to avoid such a proposition. How would you handel the mathematical expression 1=0? I mean, in your theory is 2+2=4 on Mondays and 6 on fuzzy Tues? Although problems in language and symbolic expressions of logic might prohibit absolute certainity, I think certainty is pretty much the goal of theory building. I doubt if you are going to find many mathematicians or theoritical psychologists for that matter running to celebrate a theory that advocates an ambiguous/paradoxical proposition. Best, David

-- david clark (doclark@yorku.ca), May 28, 2002.

Thanks David, But no. I think we are actually in a different book, let alone page! You are talking about statements like 'Only one thing is certain-- that is, nothing is certain. If this statement is true, it is also false.' All great fun, but totally useless as a scientific pursuit.

I am talking more about psychology/science operating under an 'illusion of precision'. It is more a reminder that under the theory of probability the best we can afford is an approximation of truth, but often we act as if theory is fact. And so does the general public. What I propose is an approach to theory development, rather than a theory of science per se. In 1956 (or thereabouts) Oppenheimer addressed the APA with a warning that psychology was still emulating a science of physics that no longer existed. This is my starting point. He suggested that we not only understand the principle of 'uncertainty', but embrace it as a means to progress. I do not question the need for accuracy of measurement and such.

The fuzziness I refer to is 'fuzzy set theory' as a means to eliminate some of the problems of linear and dichotomous measurements. Do I exhibit high self-esteem or low self-esteem? True or false? Black or white? The ancient Sorites dilemmas... The world does operate in shades of grey, and so does our subject matter. Acceptance of 'fuzzy' boundaries rather than 'crisp', and development of a means to measure them, could lead to greater precision in the long run.

As for ambiguity. I draw an analogy from the concept of 'figure- ground'. As humans we exhibit an extraordinary amount of laterality, both in perception and in language. Think about the issues so hotly debated within psychology. NOTHING is clear. In hearing a theory (or reading about it), if the listener/reader/psychologist is open to ambiguities then sometimes the main contentions of that theory can become ground and the backround 'noise' can become figure. New theories are born through this ( and a large dollop of scientific method and intelligence). I contend that in teaching our newer generation of psychologists we need to be a little less fearful of these concepts and a little more sceptical about the illusory precision that is so cherished.

I could go on, but I have too many words already. Are you reading a little of my page yet?

:) cheers, A

-- annette (net_davey@hotmail.com), May 29, 2002.


Actually David,

I just read your answer again, you were just confirming what I was NOT saying am I right?

A

-- annette (net_davey@hotmail.com), May 29, 2002.


Hi Annette, I just have one comment to what you said: "I contend that in teaching our newer generation of psychologists we need to be a little less fearful of these concepts and a little more sceptical about the illusory precision that is so cherished."

My only comment on this is that I believe psychologists, if anything, are TOO accepting of the lack of precision in research. If anything, psychologists need to be aspiring to point-estimate predictions, instead of broad results that may or may not be replicable. So, I would say that we tolerate ambiguity and uncerntainty in our science well enough, and this is before proposing a theory to actually accomplish this. Using probabilistic statements helps to quantify the degree of uncertainty, and in this sense, we could be said to embrace, or at least be tolerant of, uncertainty.

-- Daniel J. Denis (dand@yorku.ca), May 29, 2002.



Round 2, OK. Let?s approach this from a different direction.

1st Chaos Theory is about statements which rest on exponential variables and interactions, and uncertainty requires a probabilistic analysis. Even Turing?s proof for the problem that couldn?t be solved by an algorithm required a proof provided by the invention of a universal-computing machine. Just because a set of mathematical logicians decided for a round about approach to a satisfactory solution for the present, doesn?t mean (1) that they abolished the requirement for a rigorous proof, and (2) it doesn?t mean all the other mathematicians threw their hands up in defeat and gave up looking for the absolute answer to where an electron is in space or what the inciting cause of a thunder storm is either. So, fuzzy logic, non-linear systems, feed forward systems ? whatever; neither the uncertainty principle or chaos theory were proposed as a license to make claims without rigorous proof which ?really- is the method of science, rigorous proof that is.

If Oppenheimer argued for uncertainty, remember at least one man didn?t believe God played dice with the universe.

Back to your theory. I assume you have something to say, assert, posit. I mean, you do have something to say about something psychological don?t you? I guess you will have to make that statement in the form of a subject and a predicate, and since we are about science, I assume you are going to make some claim to the truth, the validity or correspondence of your statement. Do you have something different in mind that the form of proposition: x is y or 20% of the time x is y? Sooner or later, you will want to take a position if you are a research psychologist.

Next step. After you make your claim, seems to me you are going to have to back it up. Like my friend Terry is found of say, an experiment is nothing but a no-holds barred attempt to convince yourself and your peers that you are right. In mathematics and even in behavioral science this is a brutal business of logical proofs and scientific demonstrations. I mean, no ill reflection on your character, but you can?t seriously expect me to take your word as warrant for a proposition, to ?believe you,? ? do you? You do know you are going to have to prove to me in exacting detail that what you claim is valid. Or, like Terry also said, well sure you can go ahead and do that experiment and make that claim, but who would believe you?

OK. Principles of uncertainty and chaos theory and anything else we can steal, they are sometimes useful concepts even if only for analogies, but to celebrate them? to embrace them? I don?t think so. I think everyone in this game is only too well aware of the ambiguities, every time you walk into a lab you?ve got doubt stalking you. I personally haven?t met a researcher with any illusion of precision as you call it. But embracing ambiguities isn?t the mantra of this science ? it isn?t going to happen.

Finally, this whole bucket of concepts has been around for decades, and I can?t help but think, if there was all that much to it, it would have taken hold a long time ago.

Best, David

-- david clark (doclark@yorku.ca), May 29, 2002.


Hmmm, let me think, David, you seem to have quite a srong feeling about this particular issue. If I just reiterate for a moment though...in my first mail i said quite clearly that I 'do not contest the need for precision and accuracy in scientifically rigourous development of theory...'. This is not the issue here. I am simply saying that, in the presence of such scrutiny we need to understand that life is not linear, it is lateral. There is MORE. Chaos theory is one area i am trying to avoid. Two reasons, it is done to death and it is NOT where I am coming from.

Lets take an analogy from music. The great composers worked within the scope of the laws of music, and believe me, they are every bit as stringent as the laws of the universe, and yet there was great divserity of composition. Precision and accuracy were the springboard of genius. This is my contention here. As in the arts, so for science. Is the Mona Lisa, the only way to paint a face; is Dickensian literature the only way to describe man's impoverished state? No! There is a multitude of diversity in all we do and precision and accuracy and, if you insist, mathematical computation at the core. Without embracing and celebrating the diversities brought about by being able to see from many angles, or working with what we DO have, which IS fuzzy, then we end up with melodious ditties for theories rather than brilliant compositions.

I feel that you are reading what you THINK I am saying, according to previous conceptions you may have, rather than really seeing what it is I am saying. And in answer to that part of your message, YES I have a LOT to say for Psychology. And I do my best to make it unambiguous...

I'm still thinking...:-) A

-- annette (net_davey@hotmail.com), May 29, 2002.


Hi Daniel,

Yes I take your point, but I think, once again that perhaps I have not explained my position clearly enough. There is way too much ambiguity in research that is only accountable for as sloppiness. I have no time for it. This is not the type of ambiguity and uncertainty that I mean to embrace. It is the ability to see beyond what is said and done...when all is said and done with the greatest of care.

cheers, a

-- annette (net_davey@hotmail.com), May 29, 2002.


thanks for the argument by the way...it really helps, and is just what i needed to kick start my sluggish academic attitudes now that my deadline approaches!

-- annette (net_davey@hotmail.com), May 29, 2002.

But Annette, what you suggest in reply to David's answer, that we need to be "celebrating the diversities brought about by being able to see from many angles" is in large part done today, no? How many different fields of research do we have to investigate one thing and one thing only? Consider something like depression. Do we not have MANY different perspectives on the disorder, and isn't it true that most psychologists would recommend a multidisciplinary approach to the subject? I would argue that it is because "depression" is something quite "fuzzy" still, and to deal with this fuzziness, the research community closes no doors on possibilities. But wouldn't you agree that the goal in such a case is to MINIMIZE fuzziness rather than celebrate it?

My point is that I see nothing novel in what you propose. As for musical compositions versus science in general, I don't think you can compare the two. I can play a beautiful piece of music, and if you can't replicate it, I doubt it would raise an uproar. But, if I suggest that treament X be used to treat a cancer, then you would probably want to know how I came to that conclusion so you could replicate it and possibly expand on it.

Also, I'm not sure I understand your argument about non-linear systems. Non-linear systems are a part of science in general, but I would think that most have some function and are not simply random. Just because something is non-linear doesn't mean it's unpredictable.

-- Daniel J. Denis (dand@yorku.ca), May 29, 2002.



You write: "A celebration of these concepts rather than a denial of their usefulness, under the 'illusion of precision' is surely called for?"

Surely? It seems to me that the aim of science is not celebration but knowledge. Ambiguity and vagueness may be inevitable, and certainly we should come to terms with working in their midst, but the problems they pose are real: (1) they can lead us to believe we have knowledge when in fact we don't, detering us from doing the work that yet needs to be done, (2) they lead us to believe that we (dis-)agree with each other when we don't, either detering from necessary further work, or leading us into further unnecessary work. These outcomes are not, in the main, to be celebrated but avoided. We should not labor under a "illusion of precision" as you say, but (surely?) we should aim for *real* precision where we can.

It may be that ambiguity and vagueness are virtues in certain literary forms (where the aim may indeed be to "celebrate" the topic at hand), but this is not the aim of science.

-- Christopher Green (cgreen@yorku.ca), June 02, 2002.


A lot of the replys make the assumption that there is something that is a "certainty". Philosophers have for thousnads of years postulated about that which is certainty. Descartes' theory come down to the belief that awareness of your own thinking is a'certainty' and that based upon such a certainty, other ideas can be resolved with an absolute certainty. Considering that logic itself is based upon postualtes that are not, in themselves. provable by logic, it seems to me that the is not so much certainty, but agreement to not to disagree over a given point. Logic is a tool and it works well were it is applicable, but it is not always applicable. there is no logic that proves that my being aware of my own thinking, necessarilly proves that I am thinking. Basically, Descartes made a huge leap of faith, in that after doubting all, he was certain that he was thinking, yet this is not proven, just accepted. Logic itself is used not because it is 'proven' to work, but rather that one has faith that it will work.

-- Paul Crossett (pjc55@kvalley.com), June 05, 2002.

Ok, back to it,

A big yes on the certainty response! Look I think I may not have been clear? I am talking about the 'mind-set' of the theoretician. I am talking about the intrinsic ambiguity found within EVERYTHING. And I am talking about the fact that univocality is only a part of equivocality. We can use it as a boundary, but if we are really honest within our profession then we have to accept that there are options. Feynman calls it 'science in a straight jacket'. How else do theories progress (other than very slowly). Maverick scientists, and I don't mean cranks, get tremendous results because they are able to look beyond legalism and, NOT DENYING the laws and boundaries, they see BEYOND them.

Look at the work of Feynman (Physicist) Feyerabend (a bit out there I admit!) or even Einstein. If you look closely at their outlook on life you will find a little of what I am saying here.

Reiteration of main point...Precision is not questioned here. I am not against the rigourous methods of science. I simply believe that truly valuable scientific ideas come from people willing to go beyond these bounds. First we must have those bounds firmly in place. We would be foolish to teach children only to have lateral thinking before they were taught to think straight. Laterality is the next step. Another analogy...'specially for those of you who think that they are not relevant...Kohlberg's moral dilemma. You know the one! Do we stop at legalism or is there more for the above average thinker? Morallistically YES!!!! A little thought!

So with theory creation. We can have a similar theory of theory development. Beyond legalism...

Still thinking!

Cheers

-- annette (net_davey@hotmail.com), June 05, 2002.


Hi Annette, I think the confusion here might be from what we consider as being within the bounds of theory building. Are you talking about the "ideas" of science in the sense of coming up with a hypothesis to test? I don't think that lies within theory building exactly. The hypothetical ideas seem to come from familiarity with the material and intuition. In brief, that part - the initial germ and inspiration on which theory construction takes off - is probably more of an art. As far as I can tell, the best way to study it is to read biographies of inventive and creative people.

Propositions and proofs, the currency of theory and subject matter of debates on theory construction, is another matter, and it is this field I think a lot of your respondents are addressing. This is the field that many people study as symbolic logic or mathematics.

Are you making a distinction between the above?

Best, David

-- david clark (doclark@yorku.ca), June 08, 2002.


Moderation questions? read the FAQ