|
2007
Interview
Question. In a prior interview you stated the completion of the Red Tail Math publication requires an analysis of emergence, your focus for several years now. What is the current progress or status of this undertaking?
Answer. Emergence is a very difficult subject and I have the impression that we are not very far down the road toward a coherent representation – one that will allow us to make the phenomenon useful in an applied and practical way – certainly, that is an accurate description of my situation. Emergence is complex enough that I find myself reading, thinking, projecting my-as-yet-unclear notions about it onto experience for some time and then follows a period of unconscious processing – indeterminate length – than a return to the subject. I have no idea how long this process will continue and there is no light at the end of my tunnel as yet.
Question. In your paper Steps to an Ecology of Emergence with Carmen Bostic St. Clair and Thomas Malloy,
http://www.nlpwhisperinginthewind.com/art_EcologyofEmergence.htm you state “To begin to take steps to a mental ecology of emergence we first establish two fundamental assumptions from the methodology of transformational grammar - The centrality of human judgment based on direct experience and the proposition that the systematic nature of human behavior is algorithmically driven. Please comment on the importance of developing an operational definition of Emergence as it relates to human knowing.
Answer. Well, the most fundamental distinction is the question whether the emergence of a NEW pattern or phenomenon is a statement about the limitations of our ability to compute what will occur when two distinct processes are brought into contact or the present lack of appreciation about the elements and their characteristics going into contact from which something NEW might emerge or a failure to appreciate the contribution of the processing by the human neurology or that there are non-linear phenomena for which we have yet to develop adequate methods for investigation. This is perhaps the most basic question, although it may turn out to be one of the last answered.
If this is an accurate description of where we are in this investigation (and it is accurate for me), then we don’t even know whether the phenomenon is a primary investigation of our own neurology, the as-yet-uncharted processes in the world or some amazing interaction effect between the two. From a epistemological point of view, this is intolerable.
Question. Automata Theory is a specialty you used to teach for the math department at the University of California, Santa Cruz. You have commented previously “Ask anyone using that term to define it and you will be treated to what in German is called Schaum – meaning hand waving and vague responses.” How might you define automata theory for the novice? Why is an understanding or appreciation of automata theory important for NLP Practitioners – Trainers - Modelers?
Answer. The formal definition is the study of abstract machines (computational machines, more specifically). It is a theoretical field where different hierarchies of abstract machines are defined by classes of properties and then the implications of placing these different sets of constraints on them are explored. As with any abstract mathematical model, it has no inherent meaning or application and may be considered a complex and extended foray into formalisms.
Now, this particular case – Automata Theory – has a well-known and powerful application – it is the underlying theory for all computational embodiments – the computer being the most recognizable of these. In this sense, all theoretical work in computation is founded on Automata Theory. There are other applications – historically, Chomsky used a mapping between hierarchies of grammars and hierarchies of automata to demonstrate (actually PROVE – in the strong formal sense) that the grammars of the previous generation (Bloomfield) were unable IN PRINCIPLE to describe some basic syntactic structures that occur in every (the possible counter-example is a Brazilian tribe known as the Piraha) natural language. This was such an effective and compelling argument that his (Chomsky) position rapidly supplanted that of Bloomfield and his students.
As for your precise question:
Why is an understanding or appreciation of Automata Theory important for NLP types?
I am quite certain that it is possible to achieve mastery of the applications of NLP coded patterning with zero understanding or appreciation of Automata Theory. Speaking more personally, I find that having a completely formalized symbol systems (with no inherent meaning associated with it) – Automata Theory – is useful in the extreme. In some sense, one of the contributions of NLP is the extension of such formal thinking from language structures to the more general question of human behavior, including linguistic performance. One of the most interesting claims from Automata Theory is that if something can be described explicitly, it can be represented by a special type of automata: the Turing machine (named after a brilliant British mathematician Alan Turing). I find this simultaneously surprising and very useful. For interested readers, I recommend a very readable and clearly presented version of these notion in W. Ross Ashby’s Introduction to Cybernetics where there are many concrete and interesting applications of this class of thinking presented.
Perhaps it is useful to note that while mastery of the application patterns of NLP is perfectly possible without any understanding or appreciation of Automata Theory, it is also possible to drive a car without any understanding or appreciation of the internal combustion engine. And this works just fine until something goes awry. Certainly. It is non controversial to state that anyone without this understanding and appreciation will NOT innovate or discover more advanced or elegant forms for the phenomenon: neither the internal combustion engine or the patterning of NLP.
Question. A Batesonian epistemological approach specifies mental process as the transformation of differences across a richly connected network. Is this “richly connected network” similar to or a reference to the neurological transforms outlined in Whispering in the Wind? ( www.nlpwhisperinginthewind.com) What practical applications materialize from this approach?
Answer. For starters, it is the beginning of a refinement of the map-territory distinction with some precision. The set of transforms that occur between receptor and First Access (the point in our neurology where we first gain access to what is going on – for example, the occipital lobe for visual experience) guarantees that the map-territory point applies with full force.
We seem to me to be at the very beginnings of investigating the specifics of this map-territory and anyone who appreciates this distinction will be a lot more tolerant of differences as the world gets quite slippery at such points
The entire point of the various models of language specification (meta model, precision model and the verbal package) rests on the insight that while languages are impositions on experience, the investigation (using the tool of language itself) has extremely useful consequences as people shift to overcome their personal limitations and obstacles to their continuing learning and evolution – none of this is possible in any deep sense without this point. Once the specification patterns have succeeded in connecting language with experience, we move into the really fascinating issues surrounding the relationship between experience and what’s out there and its; relationship to the representations we create within ourselves.. You are entirely correct to identify Bateson’s work as the inspiration for the development by Carmen and me in Whispering of the epistemological distinctions we offered in our presentation of transforms.
Question. Emergence as Metaphor is addressed in Steps to an Ecology of Emergence,
http://www.nlpwhisperinginthewind.com/art_EcologyofEmergence.htm
“The importance of metaphor’s function in a mental ecology is both pervasive and useful…” Turing’s work led to a powerful metaphorical alternative to the metaphor of a Universe designer….thereby provoking an alternative framework to theories and discourses that presupposed life needed ultimately to be explained by a designer. Is this an example of the powerful influence of a paradigm shift via metaphor?
Answer. If your observation is accurate, it would be a compelling example for such a paradigm shift through the providing of an alternative metaphor, less demanding in its belief requirements for acceptance. Since I lack the historical background to evaluate this claim, I will offer an alternative example: Wolfram’s paradigm challenge A New Kind of Science provides a wonderful example of what you are referring to (independent of the irrelevant but quite hot issues surrounding theism). Wolfram’s points out that in the present paradigm, non-linear processes are typically represented by sets of differential equations. We have been trained (by presupposition more than explicit tutelage) that when confronting with a complex phenomenon, to anticipate that the formal representation of such a phenomenon will be correspondingly complex: thus the representation of observed complex non-linear processes by sets of differential equations (notoriously difficult to compute). Wolfram points out that there is an alternative – in the 20th century, the creation of recursive rule systems (strongly connected with Automata Theory, by the way) offers a most attractive alternative. With a simple recursive rule set, it is possible to generate extremely complex phenomena it is a presently open question whether such recursive rule systems can adequately and usefully serve as models for the sets of non-linear processes presently under investigation in, for example, physics. Literally, until the development of the new mathematical metaphor (recursive rule systems), it was not possible to reduce the complexity and escape the perceptual trap concerning complexity that we had generated historically. Remember non-Euclidean geometry were developed as abstract formal studies long before any application was apparent – consider, for example, the gap in time between Rienmann’s geometries (suspending the parallel lines axiom and its application to Eistein’s relativity theory.
This seems to be an excellent example of one of Kuhn’s principle points (in, for example, hisThe Structure of Scientific Revolutions) – no paradigm will be dislodged until two conditions (minimally) are satisfied:
1. there are numerous counter-examples to the present paradigm or alternatively, there are significant phenomena recognized by the professionals in the discipline as important patterning to be described and the present paradigm is demonstrably incapable of describing them.
2. there is a well-defined alternative to replace the present paradigm that does not suffer the limitations described in 1) above.
The example mentioned previously of Chomsky successfully challenging Bloomfield dominant paradigm in linguistics is are precise example of this as is Wolfram’s current proposal.
I will trust that the reader recognized that formal systems are simply a special species of metaphors – the point you make is most welcome.
Question. “The power of the emergence insight is that wholes self-organize themselves as a natural function of the interplay of the processes that make them up.” Is emergence an advanced alternative to reductionism that takes the nature of complex things reducing them to the sums of simpler or more fundamental things?
Answer. As Marx wrote
“If there were no difference between appearance and reality, there would be no need for science “
I will pass on Marx’s failure to distinguish between reality and experience and simple point out that all explanations are descriptions of phenomena in metaphoric terms – sometimes the metaphors are verbal, sometimes formal (Automata theory or recursive rule systems,..). Note that all explanations are descriptions of X in terms of Y - not a bad description of metaphor - and are therefore simultaneous enlightening and reducing. I use the term reducing in contrast to reductionistic. For me, the distinction is that a reductionism is the description of X in terms of some pre-determined Y: for example, all differences in human behavior can be accounted for by differences in the genetic code. It seems to me that all explanation is an attempt to reduce the complexity of the thing investigated – otherwise, what would the point be: a map scaled to a 1:1 ratio with the territory (say, the terrain between San Francisco and Los Angeles) would be a replica and by definition, would be as large as the territory it is supposed to represent. I don’t see any point to such an exercise. I conclude that all explanations will involve reducing the complexity of the thing explained. However, this is distinct from reductionism in which the vocabulary and terms of explanation flow from some source other than the phenomenon under investigation. In the present political climate, for example, any objection to the war in Iraq is typically reduced by the administration to “not supporting the troops” therefore rendering political discussions moot. Watch the Colbert Report (Comedy Central) for many additional examples.
2005
Interview
|
