Re: [Fis] THE NEW YEAR ESSAY Fis Digest, Vol 10, Issue 11 Mechanism and Model
Cari Tutti, nei miei numerosi libri, a partire da "Economia del patrimonio architettonico-ambientale" (1983), ho sostenuto che la triade semiotica significazione, informazione e comunicazione attraversa il mondo biologico, fisico e sociale e viceversa il mondo biologico, fisico e sociale attraversa la triade semiotica significazione, informazione e comunicazione. Quindi non ha senso pensare che il mondo informativo sia separato dagli altri mondi (cfr. Maturana e Varela). A seconda i processi o modelli dei suddetti mondi che si considerano si possono usare alcune o tutte le categorie di informazioni possibili: genetica (genealogica), termodinamica o naturale (entropico/neg-entropica), matematica (entropico-cibernetica) e semantica (storico-culturale o significato-significante). Le stesse unità autopoietiche possono allentare o ridurre la loro auto-referenzialità mediante l'informazione-comunicazione che supera la rigidezza o la chiusura dei loro codici. Per questo in "Valore e valutazioni" (1999) mi sono posto in una situazione intermedia tra Maturana-Varela e Niklas Luhmann. Per comprendere meglio il mio approccio è necessario: assegnare all'economia il ruolo di "scienza delle scienze" che le conferiva anche Ernst Mach; considerare l'informazione la "legge delle leggi" di tutte le scienze dell'uomo e della natura.l Un abbraccio affettuoso a Tutti, da un poverino esponenziale, quale sono. Francesco Rizzo. 2015-01-19 20:37 GMT+01:00 Joshua Augustus Bacigalupi < bacigalupiwo...@gmail.com>: > Josh Bacigalupi here, fellow pirate. Thank you all for this thoughtful > discussion. > > Work is a fundamental focus of Terry's project. We can all agree that the > creation of entropy is necessary to do work; such degradation of a gradient > is a necessary precondition of work potential, but not just any work. The > specific kind of work that some self-entailed proto-cell does in its > environment must be such that it increases the chances that such nascent > agency will have increased the chances of its own propagation in that open > system. Terry calls this teleodynamic work. > > But this isn't even the most stringent requirement we place on ourselves. > Not only must this work be relevant to its own persistence, *the > constraints necessary to enact this specific dynamic must be able to > persist for some finite time in the absence of any gradient what-so-ever.* > In other words, Terry's hypothesized "autogen" is specifically conceived to > retain the capacity to do self-efficacious work even after local chemical > equalibrium has been attained. > > Once a gradient is again available, any viable autogen must be able to > restart the very specific co-constraints of auto-catalysis and > self-organized containment, a process that we suggest must be able to both > self-repair and create new sets of co-constraint in wholly novel > substrates. This, in effect, spans the ontological gap from the vast > majority of physico-chemical dynamics to the first distinct dynamic of a > measurable medium of informational significance, whose benchmark of > significance is the persistence of autogenic constraints. > > Although intriguing, we are skeptical when speculating about vastly more > complex and likely intentional agents, like bacterium, or clearly > intentional agents, like humans. We suggest that focus on a priori > intentional agency skips the distinct logical step from ubiquitous > self-organizing dynamics, where rate of entropy production is increased > (dissipating not only the external gradient but the internal organization > itself), to the relatively rare "teleodynamics", where rate of entropy and > work production are mitigated by the autogen's normative relation to its > surroundings. > > Cheers, > Josh > > > > On Sun, Jan 18, 2015 at 8:51 PM, Terrence W. DEACON > wrote: > >> Hi Joseph, >> >> Glad to have you join in. My goal is (paraphrasing Einstein) to >> develop a model system that is as simple as possible but not too >> simple to provide a foundation for formalizing the concepts of >> reference and significance. If too simple, it would be helpful to know >> what is specifically missing. >> >> In considering more complex model systems the critical constraint is >> to avoid cryptically assuming a homuncular perspective that sneaks in >> some undescribed mentality (often an external observational >> perspective) to do the interpretive work and to define what >> constitutes reference and significance. I am unwilling to use a >> bacterium as my model, because we implicitly assume their end-directed >> and sensing capacities without explaining them. Nor am I willing to >> assume that nucleic acids are intrinsically informational or that >> information is just pattern replication, as has become a common >> assumption in many evolutionary theories. >> >> As I have said a number of times, my goal is not to deal with all >> aspects of the information concept, and certainly not at the level of >> human thought. I
Re: [Fis] THE NEW YEAR ESSAY Fis Digest, Vol 10, Issue 11 Mechanism and Model
Josh Bacigalupi here, fellow pirate. Thank you all for this thoughtful discussion. Work is a fundamental focus of Terry's project. We can all agree that the creation of entropy is necessary to do work; such degradation of a gradient is a necessary precondition of work potential, but not just any work. The specific kind of work that some self-entailed proto-cell does in its environment must be such that it increases the chances that such nascent agency will have increased the chances of its own propagation in that open system. Terry calls this teleodynamic work. But this isn't even the most stringent requirement we place on ourselves. Not only must this work be relevant to its own persistence, *the constraints necessary to enact this specific dynamic must be able to persist for some finite time in the absence of any gradient what-so-ever.* In other words, Terry's hypothesized "autogen" is specifically conceived to retain the capacity to do self-efficacious work even after local chemical equalibrium has been attained. Once a gradient is again available, any viable autogen must be able to restart the very specific co-constraints of auto-catalysis and self-organized containment, a process that we suggest must be able to both self-repair and create new sets of co-constraint in wholly novel substrates. This, in effect, spans the ontological gap from the vast majority of physico-chemical dynamics to the first distinct dynamic of a measurable medium of informational significance, whose benchmark of significance is the persistence of autogenic constraints. Although intriguing, we are skeptical when speculating about vastly more complex and likely intentional agents, like bacterium, or clearly intentional agents, like humans. We suggest that focus on a priori intentional agency skips the distinct logical step from ubiquitous self-organizing dynamics, where rate of entropy production is increased (dissipating not only the external gradient but the internal organization itself), to the relatively rare "teleodynamics", where rate of entropy and work production are mitigated by the autogen's normative relation to its surroundings. Cheers, Josh On Sun, Jan 18, 2015 at 8:51 PM, Terrence W. DEACON wrote: > Hi Joseph, > > Glad to have you join in. My goal is (paraphrasing Einstein) to > develop a model system that is as simple as possible but not too > simple to provide a foundation for formalizing the concepts of > reference and significance. If too simple, it would be helpful to know > what is specifically missing. > > In considering more complex model systems the critical constraint is > to avoid cryptically assuming a homuncular perspective that sneaks in > some undescribed mentality (often an external observational > perspective) to do the interpretive work and to define what > constitutes reference and significance. I am unwilling to use a > bacterium as my model, because we implicitly assume their end-directed > and sensing capacities without explaining them. Nor am I willing to > assume that nucleic acids are intrinsically informational or that > information is just pattern replication, as has become a common > assumption in many evolutionary theories. > > As I have said a number of times, my goal is not to deal with all > aspects of the information concept, and certainly not at the level of > human thought. I merely propose to dissolve the implicit dualism in > our current concepts at the most basic level, so that for example it > will be possible to develop a scientifically grounded theory of > molecular biosemiotics. > > As to the point that we need to consider quantum effects, I worry that > it also allows another black box to stand in for an explanation. > Quantum effects are definitely real, and though well described, their > interpretation is even less approachable than the concepts of > reference and significance in information. I worry that we risk trying > to explain one mystery by invoking an even greater mystery. I suspect > that there are aspects of quantum theory that are problematic > precisely because we lack a clear understanding of the referential > aspect of information. So the reanalysis of information that I am > suggesting may actually contribute to a better understanding of the > information provided by quantum experiments, rather than the other way > around. The key link is to the concept of physical work (which I argue > is essential for defining reference and significance). In this > submicroscopic domain where the concept of physical work requires a > different framing (though what this is is not obvious), the very > nature of reference must also be reframed. This is an implication of > this analysis that I would love to see developed. > > — Terry > > On 1/18/15, joe.bren...@bluewin.ch wrote: > > Dear Pedro, Terry, Bob L., Bob U., Loet, Gordana and All, > > I have been in transit from Switzerland to California and only now have a > > moment to even start to comment on what
Re: [Fis] THE NEW YEAR ESSAY Fis Digest, Vol 10, Issue 11 Mechanism and Model
PS typo correction line 5 from bottom: ... To specify information *that* a given constraint-state of a On 1/19/15, Terrence W. DEACON wrote: > Hi Loet, > > I do indeed consider this relationship to be measurable and thus > expressible mathematically. This in itself doesn't mean that it > ignores content. Indeed, a specific content and a specific target > function-state are prerequisites, and so must be assumed in the > analysis. In my opinion, as necessary assumptions, this makes the part > of the background physics. So there must be both universality and > physical specificity to this analysis— the specificity of referent and > significant end-state treated as givens in the equation. > > The term "expected" plays a crucial role here. It introduces a > Bayesian implication behind Shannon's analysis. But it also is what > necessitates the self-repairing, self-reproducing features of > autogenesis. To specify information what a given constraint-state of a > medium represents there must be a reference state. However, it cannot > be MEP or even maximum thermodynamic entropy (analogous to Shannon's > entropy) but instead the work differential between current state of > degraded autogenesis and a reconstituted or reproduced autogen. > > — Terry > > On 1/18/15, Loet Leydesdorff wrote: >> Dear Terry and colleagues, >> >> >> >> “As I have said a number of times, my goal is not to deal with all >> aspects >> of the information concept, and certainly not at the level of human >> thought. >> I merely propose to dissolve the implicit dualism in our current concepts >> at >> the most basic level, so that for example it will be possible to develop >> a >> scientifically grounded theory of molecular biosemiotics.” >> >> >> >> Is the crucial point that an expected information content is always >> referential to a maximum entropy and therefore a relational concept? The >> significance/meaning is thus provided by the redundancy? >> >> >> >> I doubt whether this is part of the physics (as you seem to claim). It >> follows from the math and is yet content free; in other words, it can be >> provided with meaning given any system of reference or, in other words, >> discourse. The universality of the claim would thus be based on the >> mathematical (dimensionless) character of it. >> >> >> >> Best, >> >> Loet >> >> >> >> > > > -- > Professor Terrence W. Deacon > University of California, Berkeley > -- Professor Terrence W. Deacon University of California, Berkeley ___ Fis mailing list Fis@listas.unizar.es http://listas.unizar.es/cgi-bin/mailman/listinfo/fis
Re: [Fis] THE NEW YEAR ESSAY Fis Digest, Vol 10, Issue 11 Mechanism and Model
Hi Loet, I do indeed consider this relationship to be measurable and thus expressible mathematically. This in itself doesn't mean that it ignores content. Indeed, a specific content and a specific target function-state are prerequisites, and so must be assumed in the analysis. In my opinion, as necessary assumptions, this makes the part of the background physics. So there must be both universality and physical specificity to this analysis— the specificity of referent and significant end-state treated as givens in the equation. The term "expected" plays a crucial role here. It introduces a Bayesian implication behind Shannon's analysis. But it also is what necessitates the self-repairing, self-reproducing features of autogenesis. To specify information what a given constraint-state of a medium represents there must be a reference state. However, it cannot be MEP or even maximum thermodynamic entropy (analogous to Shannon's entropy) but instead the work differential between current state of degraded autogenesis and a reconstituted or reproduced autogen. — Terry On 1/18/15, Loet Leydesdorff wrote: > Dear Terry and colleagues, > > > > “As I have said a number of times, my goal is not to deal with all aspects > of the information concept, and certainly not at the level of human thought. > I merely propose to dissolve the implicit dualism in our current concepts at > the most basic level, so that for example it will be possible to develop a > scientifically grounded theory of molecular biosemiotics.” > > > > Is the crucial point that an expected information content is always > referential to a maximum entropy and therefore a relational concept? The > significance/meaning is thus provided by the redundancy? > > > > I doubt whether this is part of the physics (as you seem to claim). It > follows from the math and is yet content free; in other words, it can be > provided with meaning given any system of reference or, in other words, > discourse. The universality of the claim would thus be based on the > mathematical (dimensionless) character of it. > > > > Best, > > Loet > > > > -- Professor Terrence W. Deacon University of California, Berkeley ___ Fis mailing list Fis@listas.unizar.es http://listas.unizar.es/cgi-bin/mailman/listinfo/fis
Re: [Fis] THE NEW YEAR ESSAY Fis Digest, Vol 10, Issue 11 Mechanism and Model
Dear Terry and colleagues, “As I have said a number of times, my goal is not to deal with all aspects of the information concept, and certainly not at the level of human thought. I merely propose to dissolve the implicit dualism in our current concepts at the most basic level, so that for example it will be possible to develop a scientifically grounded theory of molecular biosemiotics.” Is the crucial point that an expected information content is always referential to a maximum entropy and therefore a relational concept? The significance/meaning is thus provided by the redundancy? I doubt whether this is part of the physics (as you seem to claim). It follows from the math and is yet content free; in other words, it can be provided with meaning given any system of reference or, in other words, discourse. The universality of the claim would thus be based on the mathematical (dimensionless) character of it. Best, Loet ___ Fis mailing list Fis@listas.unizar.es http://listas.unizar.es/cgi-bin/mailman/listinfo/fis
Re: [Fis] THE NEW YEAR ESSAY Fis Digest, Vol 10, Issue 11 Mechanism and Model
Hi Joseph, Glad to have you join in. My goal is (paraphrasing Einstein) to develop a model system that is as simple as possible but not too simple to provide a foundation for formalizing the concepts of reference and significance. If too simple, it would be helpful to know what is specifically missing. In considering more complex model systems the critical constraint is to avoid cryptically assuming a homuncular perspective that sneaks in some undescribed mentality (often an external observational perspective) to do the interpretive work and to define what constitutes reference and significance. I am unwilling to use a bacterium as my model, because we implicitly assume their end-directed and sensing capacities without explaining them. Nor am I willing to assume that nucleic acids are intrinsically informational or that information is just pattern replication, as has become a common assumption in many evolutionary theories. As I have said a number of times, my goal is not to deal with all aspects of the information concept, and certainly not at the level of human thought. I merely propose to dissolve the implicit dualism in our current concepts at the most basic level, so that for example it will be possible to develop a scientifically grounded theory of molecular biosemiotics. As to the point that we need to consider quantum effects, I worry that it also allows another black box to stand in for an explanation. Quantum effects are definitely real, and though well described, their interpretation is even less approachable than the concepts of reference and significance in information. I worry that we risk trying to explain one mystery by invoking an even greater mystery. I suspect that there are aspects of quantum theory that are problematic precisely because we lack a clear understanding of the referential aspect of information. So the reanalysis of information that I am suggesting may actually contribute to a better understanding of the information provided by quantum experiments, rather than the other way around. The key link is to the concept of physical work (which I argue is essential for defining reference and significance). In this submicroscopic domain where the concept of physical work requires a different framing (though what this is is not obvious), the very nature of reference must also be reframed. This is an implication of this analysis that I would love to see developed. — Terry On 1/18/15, joe.bren...@bluewin.ch wrote: > Dear Pedro, Terry, Bob L., Bob U., Loet, Gordana and All, > I have been in transit from Switzerland to California and only now have a > moment to even start to comment on what has become for me one of the most > interesting and useful exchanges on the list. > I would like to try to summarize my position as follows: I agree with > Terry's mechanism and I disagree with his model. As others have said much > better than I, Terry has made a major contribution to information science > (and philosophy) with his description of orthodynamic, morphodynamic and > teleodynamic processes. It can and should form the basis of all claims that > informational processes can have reference and display significance. > On the other hand, for reasons that I cannot completely express, the > autogenic model system does not, as they say here in the U.S. "work" for me. > It is stated to be simple, the simplest model, and the possibility exists > that it is too simple. We are by now all in agreement about the shortcomings > of autopoesis. But in a similar vein, in my paper in Information on Terry's > book, Incomplete Nature, I suggested that the discussion of 'information as > absence', a profound concept introduced by Terry, needed to be supplemented > by maintenance, in part, of 'information as presence'. Reciprocal > autocatalysis, also, is by no means a concept that does not still retain > many assumptions, for example, how is reciprocity achieved, what are its > implications and what is the meaning of 'auto'. Essential aspects of the > evolutionary dynamics of information may be inexpressible in the current > autogenic model due to what appear to me to be a lack of sufficiently deep > roots in quantum mechanics. > I thus see a very valuable ontological (pace Jeremy) critique of Terry's > work in progress from at least five or six perspectives, including my own. > Thank you and best wishes for 2015, > Joseph > Message d'origine > De : dea...@berkeley.edu > Date : 18/01/2015 - 13:22 (PST) > À : gordana.dodig-crnko...@mdh.se > Cc : fis@listas.unizar.es > Objet : Re: [Fis] THE NEW YEAR ESSAY AND FOUR GREAT SCIENTIFIC DOMAINS Fis > Digest, Vol 10, Issue 11 > Typo in line 7 (correction) > experimentally determine whether or not it "works" as proposed. > On Sun, Jan 18, 2015 at 1:20 PM, Terrence W. DEACON > wrote: > Gordana's response provides a wonderful opening for digging into some of the > most challenging and subtle issues lurking behind this essay. > For now I will respond to the c
Re: [Fis] THE NEW YEAR ESSAY Fis Digest, Vol 10, Issue 11 Mechanism and Model
Dear Pedro, Terry, Bob L., Bob U., Loet, Gordana and All, I have been in transit from Switzerland to California and only now have a moment to even start to comment on what has become for me one of the most interesting and useful exchanges on the list. I would like to try to summarize my position as follows: I agree with Terry's mechanism and I disagree with his model. As others have said much better than I, Terry has made a major contribution to information science (and philosophy) with his description of orthodynamic, morphodynamic and teleodynamic processes. It can and should form the basis of all claims that informational processes can have reference and display significance. On the other hand, for reasons that I cannot completely express, the autogenic model system does not, as they say here in the U.S. "work" for me. It is stated to be simple, the simplest model, and the possibility exists that it is too simple. We are by now all in agreement about the shortcomings of autopoesis. But in a similar vein, in my paper in Information on Terry's book, Incomplete Nature, I suggested that the discussion of 'information as absence', a profound concept introduced by Terry, needed to be supplemented by maintenance, in part, of 'information as presence'. Reciprocal autocatalysis, also, is by no means a concept that does not still retain many assumptions, for example, how is reciprocity achieved, what are its implications and what is the meaning of 'auto'. Essential aspects of the evolutionary dynamics of information may be inexpressible in the current autogenic model due to what appear to me to be a lack of sufficiently deep roots in quantum mechanics. I thus see a very valuable ontological (pace Jeremy) critique of Terry's work in progress from at least five or six perspectives, including my own. Thank you and best wishes for 2015, Joseph Message d'origine De : dea...@berkeley.edu Date : 18/01/2015 - 13:22 (PST) À : gordana.dodig-crnko...@mdh.se Cc : fis@listas.unizar.es Objet : Re: [Fis] THE NEW YEAR ESSAY AND FOUR GREAT SCIENTIFIC DOMAINS Fis Digest, Vol 10, Issue 11 Typo in line 7 (correction) experimentally determine whether or not it "works" as proposed. On Sun, Jan 18, 2015 at 1:20 PM, Terrence W. DEACON wrote: Gordana's response provides a wonderful opening for digging into some of the most challenging and subtle issues lurking behind this essay. For now I will respond to the comparison between autopoiesis and autogenesis and what can and cannot be learned from each. In many ways this comparison is at the center of the conceptual challenge I offer. First, autopoiesis is a philosophical concept. Though various model systems have been proposed that purport to embody its logic, it is not an empirically testable hypothesis that would allow one to experimentally determine whether or n to it "works" as proposed. Here is the definition that Maturana and Varela provided in 1980: [an autopoietic system as one that] "constitutes itself ... as a concrete unity ... by specifying the topological domain of its realization ..." "So an “autopoietic machine” is one that collectively produces its material components as well as the network of relations between them that constitutes their unity in a discrete physical location." These latter properties are not attributed to any separate and distinctive mechanism over and above the closed co-production of components and yet are essential defining attributes. Indeed, this organization is described as the "fundamental variable which it maintains constant" (p. 79). In simple terms, autopoiesis is a highly abstract account of what must be the case for something to be a living organism. In this respect I consider it to be an updated restatement of Kant's concept of the self-organization that constitutes an organism, with the added stipulation that it also somehow [how?] determines systemic unity and coherence. Here is Kant in 1790: “An organized being is then not a mere machine, for that has merely motive power, but it possesses in itself formative power of a self-propagating kind which it communicates to its materials though they have it not of themselves.” (p. 558) and “... every part ... is there for the sake of the other (reciprocally as end, and at the same time, means).” (p. 557) Kant concludes that this isn't sufficient to determine intrinsic teleology (and by implication insufficient to determine that the concepts of function and adaptation, much less information). And that these are not intrinsic attributes of organisms. I believe that that he is right to concludes that these attributes alone only provide justification for assuming that teleological attribute are descriptive glosses, not intrinsic to organisms—assigned from a sort of extrinsic transcendental perspective. In a striking parallel, the evolutionary biologist J. B. S. Haldane gave the following definition of life in 1929: “A simple organism must