Howard Pattee's 2008 paper "Physical and functional conditions for symbols, codes, and languages" is available for download here. I recently re-read it in detail and engaged in what was for me an interesting and rewarding email dialog with Howard (who is still paying the price for being Chairman of my PhD committee nearly two decades ago :^). He has given me his permission to share his comments on my email critiques of his 2008 paper, and I plan to do so in subsequent postings in this "Dialog with Howard Pattee" series.This is the first in a multi-part posting that includes portions of our email dialog.
Click for Part 2 - Determinism vs Probability
Click for Part 3 - QM and Chess Analogy
Click for Part 4 - Property Dualism
Click for Part 5 - Flatland and Higher Dimensions
HOWARD A FAVORITE ON THIS BLOG
Howard has been an Author, Commenter and/or Subject on this Blog almost since its inception in 2007. Although he has not actively posted here for some time, I've noticed that postings by or about him, as well as postings where he has made comments, consistently garner high numbers of page views, year after year. Clearly, his name is a popular Google search term!
Although I have known and highly respected Howard since I first met him at Binghamton University as a System Science professor in the early 1990's, I still have difficulty getting my anal engineer's brain around his important contributions to the field now called "biosemiotics" (from the Greek bios meaning "life" and semeion meaning "sign"). At my request, he summarized the field of biosemiotics on this blog, here.
HOWARD'S RESEARCH INTERESTS
He began his graduate study of physics in the 1940's and was attracted to the classic problem of the physical basis of life. Rather than concentrate on the chemical basis of how living organisms are created from non-living matter, he approached it from the conceptual problem of where symbolic function emerges in the context of physical laws. Howard's research interests include: Physics of Symbols, Origin of Life, Epistemology, the Symbol-Matter Problem, the Quantum Measurement Problem, Biosemiotics, and the Epistemic Cut.
The very idea of the "physics of symbols" seems like a contradiction in terms to me. Howard and I have gone round and round on the "reality" of how a physical thing becomes a "sign" and a "sign" becomes a "symbol". Although Howard is not a "dualist" in the way René Descartes considered "mental" and "material" to be two different substances, he calls himself a "property dualist" in our dialog, believing that the world consists of one type of physical substance that has both "physical" and "mental" properties.
His latest book, co-authored with Joanna Rączaszek-Leonardi, was published in 2012 as Biosemiotics, Volume 7 2012, LAWS, LANGUAGE and LIFE, Howard Pattee’s classic papers on the physics of symbols with contemporary commentary.
Links to many of his papers are available at: http://binghamton.academia.edu/HowardPattee.
I hope Howard's many fans will post comments on this Blog.
advTHANKSance!
Ira Glickstein
4 comments:
"With Pattee, the answers elude, but the questions improve."
I found this pithy quote (from Dennis Waters) on a Scientific Ameican blog posting that is well worth reading, see http://blogs.scientificamerican.com/culturing-science/2013/05/16/howard-pattee/
Ira
Dr. Pattee argues that 'symbols' are necessary for evolution. But are they sufficient? In 1940, anthropologist Leslie White published a paper in Philosophy of Science entitled 'The Symbol: Origin and Basis of Human Behavior.' White maintained that human language was qualitatively different from other primate communication systems (and all other animal communication systems) as it - and only it - supported an autonomous stream of 'super-organic' (i.e., super-genetic) information capable of increasing complexity through time. Overwhelmingly, linguists today accept this distinction. Other primates can certainly communicate with arbitrary 'symbols' that are learned. But such primate communication systems are 'stuck' at a level of complexity far below what Von Neumann believed was required for emergence of a universal automaton, and hence for evolution. Hence, other primates show complex communication, but they have NOT developed 'syntax,' a complex system of 'grammar' that is above the Von Neumann threshold, and CULTURAL EVOLUTION. Some behavior in mammals is under direct control of biological structures, some is encoded IN biological structures (i.e., neurological tissues) in the form of learned 'frozen accidents.' But for all mammals except man (so far), communication systems can encode only limited levels of information, i. e. there is no cultural evolution (although there is learning and 'symbols'). True language is both learned AND capable of non-genetic evolution. Biologists speculate that complex and repetitive hand movements involved in tool making amongst Australopithecines led to pre-adaptations in the brain (e.g., Broca's area) which, later, allowed vocal communication to link into these brain systems that facilitate complex, recursive, and embedded rules, thus producing a brain area in which a true syntax could be encoded (Chomsky's 'language acquisition device'). So, my question for Dr. Pattee - do we not need to distinguish between 'message units' in a chimp communication system, that lacks syntax and is incapable of ever-more-complex CULTURAL evolution, and the layered units of meaning in human language, which allow complexity above the Von Neumann threshold and, hence, cultural evolution? It would seem Pattee would label both communication units used by chimps, and true morphemes, as 'symbols.' Both are 'frozen accidents' that are not determined by physical/chemical laws. But, as White said, and as most linguists and many biologists now agree, cognitive 'symbols' of chimps, do not support cultural evolution, while human symbolic behavior does. Don't we need different terms for communication units in a non-culturally-evolving chimp communication system, and for units in human languages that do support evolution? (White suggested symbol v. sign).
Ira & Howard . . so glad Howard is back.
Would like to comment on his 2008 paper, but I get the error message that comment must not exceed 4096 characters -- even though it is fewer than that.
Maybe I should send it to Ira as a word attachment?
Thanks.
Eric
. . . BTW, pursuant to Howard's apparent distaste for the many-worlds interpretation of QM, there are now several working prototypes for quantum computers.
Do you know what the first quantum-computer program printed as output?
"Hello worlds."
Or not.
My Answer to Howard Pattee
(posted in two parts, because the whole thing was too big)
Part I
Though I studied neural nets in grad school and not so much the theoretical biology that Howard was doing, I have since developed an interest in the latter which converges with my understanding of information and computation.
As regards the infamous subjectivity of information, Bateson, 1972, has defined information as, "A difference that makes a difference," and Searle, 2002, has said the information and computation are in us ("observer- relative.")
So we can say (echoing Ira's diagram),
OBSERVER
(arbitrarily interprets)
AS INFORMATION
(the)
MATERIAL.OBJECT
(matter).
Thus, a highly simplified model of interplay between MIND, MATTER, and INFORMATION is suggested.
If this relates to Howard's questions re theoretical biology, it may be that the epistemic cut occurs not merely upon the advent of life, but of conscious life.
Now, the plot thickens when we ask how computation enters the picture. If information (or DATA) depends on an observer, then (algorithmic) changes to that data should likewise be observer-relative.
Alternativly, one can view the computation as
RELATIVE TO A CONSCIOUS AGENT, obtaining the following diagram:
CONSCIOUS.AGENT
(generates/causes)
ALGORITHM/COMPUTATION
(by changing)
MATTER.
Changes to matter (velocity, temperature, &c) are effected by expending energy, yielding,
AGENT
(effects via conscious intention)
COMPUTATION
(by expending)
ENERGY.
Now, taking observer to be passive & agent to be active, we obtain,
PASSIVE.MIND
(interprets)
DATA
(from)
MATTER.
-- and --
ACTIVE.MIND
(via intent)
effects ALGORITHM
(by expending)
ENERGY.
Without belaboring this further, this model suggests a relationship between mind, physics, and information, accounting equally for data and algorithm.
End Part I
Post a Comment