The battle for the heart of The Philosophy of Information: Ontic and Cosmic Implications of the natu
I discovered a long time ago that many of the contemporary debates in the philosophy of science and in metaphysics end up echoing themes and arguments as old as Aristotle's Physics - including the metaphysical musings and debates that inevitably arise when physicists (who know more about nature than anyone other than perhaps molecular bioscientists and evolutionary biologists) find themselves at the limits of what the collider, accelerator, bubble chamber, telescopes, and mathematics can do for them. For just a few examples: consider the recent musings of Max Tegmark, Nick Bostrom, Hugh Everett, Stephen Hawking, and a slew of string theorists, black hole physicists, quantum theorists, and cosmologists besides. We have some of the best minds in the world telling us such things as every quantum event involves the selection of one out of many materially real possible universes.
It is hardly surprising that under such circumstances, the nature of information itself has become something of a fiercely argued proposition, to use the colloquial. This has become even more serious now that considerations about the metaphysics of information have been linked with scientific realism versus anti-realism, and with debates about structural realism, instrumentalism, and such adjunct and motivating concerns as pessimistic meta-induction. One of the primary battle grounds in the nature of information debates is that of subjectivism versus objectivism about the nature of information. Simply stated - this is the application of a distinction that applies to many topics in metaphysics and is about whether information exists - or can obtain - in the absence of some kind of interpreter or subjective perceiving agent (in some cases the agent may be a non-cognitive or mindless consumer, but in many cases the subjectivist about information regards that a sentient cognizance is required as subjective reciever and interpreter.) The subjectivist assertion - put broadly (there are many subtle re-interpretations of it) - is that there can be no signal and no information, and certainly no semantic information (another debate in its own right), without either or all/both of an observer, receiver, consumer, and what I will call an agentive encoder (and in many cases an agentive decoder/receiver). What I will call the strong information-subjectivist idea is that encoding of real information could not occur in the absence of some kind of cognitive agent, or conscious teleological guidance fulfilling the functional role of the agentive encoder. What I will call the weak information-subjectivist idea is that there must be one or more of a teleologically configured functional agentive encoder and reciever, or a similarly organised observer or interpreter, but that these do not have to be conscious nor even cognitive agents. This latter variety of agentive encoder (and corresponding decoder or receiver) is the favourite of most philosophers of biology and molecular biosceintists interested in the metaphysics of information with respect to biosynthetic processes and genetic replication (including gene transmission).
To preview some of the more pressing sub-themes: anti-physicalists, scientific anti-realists, and supernaturalists are to varying degrees at pains to emphasise that:
1. Information (and signs, signals, messages) cannot be or reduce to the physical or to physical structures
2. One cannot have any information in the absence of a designer/teleological progenitor/interpreter/reader/perceiver of some kind (what I have called the teleological-agentive encoder/decoder). My position in the debate is that I come down hard on the side of objectivism: no subjective receivers, observers, perceivers, or interpreters are required for information to obtain. Since before the publication of my paper on the nature of semantic information (which indirectly addresses the issue of the symbol grounding problem in an unorthodox manner) I have been a proponent of anti-subjectivism about information. Since at least 2012 I have emphasised that DNA and quasars alike carry information and have signal pathways - independently of any agent perceiver or observer whatsoever. One would be hard pressed to find a DNA scientist that would accede that, had no sentient perceiving agents including human beings ever existed to analyse DNA, then it would not contain any information, nor that the information would not have in some basic functional or else reductive sense - meaning (some time later in 2014 John Collier produced a similar set of suggestions - Collier, J. Signs without Minds (Eds.), pp. 183–197. Dordrecht: Springer Netherlands. http://doi.org/10.1007/978-94-007-7732-3_10. See also Griffiths and Stotz Chapter 7 of Genetics and Philosophy).
The philosophy of biology and molecular bioscience has been a long term stage for this debate. The teleosemantic theories of Ruth Millikan were imported by her into the debates about the nature and function of information in DNA science and the gene, and have been embellished and altered in the form of what has come to be known as infotel semantics (Nick Shea). Basically stated, these theories (with variability between them) ascribe some low level sense of meaning or semantic content to physically transmitted genetic information on the basis that in processes like DNA replication and protein synthesis, there is downstream a teleo-functional consumer of the messages/signals/signs that are transmitted: a kind of non-cognitive agent that does something with the signals and their structure, on a causal basis.
It is because of these approaches perhaps, and the heritage in Millikan's teleosemantics about mind or mental contents, that there are few truly objectivist statements of information realisation in the philosophy of molecular bioscience. Yet it should be noted that there is not any indication in teleosemantic or infotel-semantic versions of these theories that any kind of sentient cognitive or conscious agent is required for the realisation and deployment of the information in question: they are broadly naturalistic and evolutionary-reductionist in their motivation and underlying ontology. There is very little strong information-subjectivism to be found. However, when the debate about the naure of information moves into the neighbouring territory of debates about scientific realism and ontic structural realism in scientific practice and theories and their ontologies, then things become altogether less objectivist, and strong subjectivism about information starts to raise its head. One of the leading philosophers of information in the world today, Luciano Floridi (a naturalistic philosopher by all accounts), is famous for an attempt to unify (to some extent) epistemic structural realism and non-eliminative ontic structural realism under the banner of informational structural realism. His approach is transcendentalist in its underlying metaphysics of information (information is reducible to relations which can be abstract).
There is an enormous amount of literature about ontic structural realism from scholars like Steven French, James Ladyman and Don Ross (34, 37, 39), and Floridi's project has been followed with great interest by numerous scholars - including more recently the polymathematical specialist Ariel Caticha, and it is Caticha's work I would like to riff off now.
In a recent paper Caticha, who has to be one of the most interesting scholars to lend his resources to the debates about scientific realism and informational metaphysics, suggests that:
...similar pragmatic elements are an integral part of Putnam’s internal realism, of Floridi’s informational structural realism, and also of van Fraasen’s empiricist structuralism. I conclude with the conjecture that their valuable insights can be incorporated into a single coherent doctrine—an informational pragmatic realism.
Hilary Putnam was famously a theistic philosopher - at least in the sense that his nominal Jewish theism informed his philosophical discourse and argumentation at some basic level, even though his philosophy was itself carried out with eminent rigour (he is perhaps most famous for putting forward one version of the now famous Quine-Putnam indispensability argument for realism about the existence of abstract mathematical objects due to their being mentioned and used all of the time in our best scientific theories.) I am not aware of any specifically theistic or faith based claims by Caticha, but I would not be surprised to find he has similar beliefs (but that is an uninformed guess, the latitude for which I will allow myself in the spirit of friendly discussion and the context of a popular-scope article that is not intended to be very rigorous). Such a disposition would certainly provide some incidental - if not motivating - basis for both his familiarity with and deference to Putnam's thought, as well as his support of the scientific anti-realist disposition of excellent philosopher and professed Catholic Bas Van Fraasen. Perhaps it is drawing too long a bow to point out that Caticha's interesting and intelligent deployment of Bayesian statistics in his analysis - where classical information theory is known to deploy frequentist statistics - rests on the subjectivist elements of the theistic Reverend Sir Thomas Bayes' probability theory, which probability theory manages to shoe horn subjectivity in as a figurative term, ensuring the apparent necessary - at least implied - presence of an agent (and a doxastic agent at that). And adjunct note here is salient to the discussion: Many mainstay conceptions of the nature of information are probabilistic, and one of the primary alternatives - which is frequently linked with Claude Shannon's original Mathematical Theory of Communciation - is the idea that information is the reduction in statistical uncertainty about the state of an information source (a stochastic process) associated with receipt of a particular message or symbol via a communication channel (a signal pathway through a physical transmission medium that carries or transmits encoded messages/sequences of physical symbols). An important associated fact is that there are in philosophy many interpretations of probability theory and statistics: frequentist, subjectivist, classical, physical-propensity, Bayesian, hybrid, and so on. This of course has the upshot that even if somehow information is only statistical or probabilistic in its nature (and this argument is far from certain) then there are already different natures to chose from commensurate with the different available characterisations of what probabilities even are in the first place.
Returning to the example Caticha's interesting theory of the nature of information, my perhaps rather cheeky observation is that he calls upon every available professional theist philosopher that he can lay his hands on, and the naturalistic - but transcendental-Kantian approach of Floridi (Err. Yes. Kant was of course also a theist.) His affinity for epistemic structural realism, for the work of Putnam (a theist), and for the for the work of Van Fraassen (a theist) - as well as his aversion to determinism and to reductionism - all heavily implicate theism on his part as a motivation for making space in the ontology for god via transcendentalism about information.
I could be wrong about that, but I am letting myself speculate freely here, and not without some coherent premises.
Later in his paper Caticha distances himself from a correspondence theory of truth - as do all good scientific anti-realists - and espouses what most scientific realists would call scientific anti-realism (41-2), albeit a position which he later limits in an admirable pragmatic unifying effort. It is however, at this point that he invokes pragmatist outspoken theist/deist William James as informing the metaphysics of Neils Bohr:
Likewise, there is a close affinity between Bohr and the pragmatism of William James (see Stapp 1972): …in our description of nature the purpose is not to disclose the real essence of the phenomena but only to track down, so far as it is possible, relations between the multiple aspects of our experience. (Bohr 1937, p.18)
So at this point I am feeling perhaps less cheeky than insightful. Caticha goes on to quite systematically and respectably - although incompletely - defend a subjectivist interpretation of probabilities and consequently, of the nature of information that is just as doxastic or agent-belief centered (via reference to subjectivist probability interpretations):
Science requires a framework for inference on the basis of incomplete information. Our first task is to show that the quantitative measures of plausibility or degrees of belief that are the tools for reasoning should be manipulated and calculated using the ordinary rules of the calculus of probabilities—and therefore probabilities can be interpreted as degrees of belief (44).
Caticha goes on to develop some Bayesian subjectivist measures for maximum information by introducing a Bayesian conditionalisation rule to the entropy measures proposed in Shannon's classical and frequentist theory (57-64). This is the first time this has been done to my knowledge. The only observation that I will make is that Shannon's theory seems to have fared quite well in terms of its results without reference to Bayes' theorem or to subjectivism, and so perhaps shoe-horning subjectivism into the nature of information by way of insisting that it should be based upon subjectivist probability theory is a little baroque and premature?
The ontologist that does not have to make room for a god - or any other subjective observer or teleological progenitor - is in my view at an advantage, since this approach is compliant with Ockham's Razor and ontic parsimony in the right way. If there is an observer or interpreter - especially one with belief content - required in the ontology - or indeed a god in the ontology, then it is perhaps fair to expect that a metaphysics of information that is designed to correctly capture all of the anatomy or furniture of information in the ontology will reveal any god that might be lurking there. Caticha's fascinating and limber effort notwithstanding - Shannon's frequentism has done a fine job in information theory thus far. Introducing strictures and impositions upon the ontology of information and pretzelling one's efforts to make sure the interpreter/subject/god is lurking - or has room to - is evidently not a scientific approach according to the best scientific principles and ontic parsimony.
In the summary to his clever and lucid systematisation of a theory of Bayesian subjectivist adjustment to information theory, Caticha essentially spills the subjectivist beans in a scientific anti-realist sense:
Within an informational approach it is not possible to sharply separate the subject matter or contents of science from the inductive methods of science; science includes both. This point of view has two important consequences. The first is that just as we accept that the contents of science will evolve over time, to the extent that contents and methods are not separable, we must also accept that the inference methods are provisional too—the best we currently have—and are therefore susceptible to future change and improvement. (66)
Caticha's approach enlists pragmatism of James' and Putnam's variety, scientific anti-realism and construtive empiricism of Van Fraasen's variety, Bayes' subjectivist probabilities, and the transcendental elements of Floridi's ISR. It's an ambitious and admirable project and it fits with Caticha's larger endeavours. Allowing Caticha to speak for himself:
I have argued in favor of an informational pragmatic realism but I have not attempted its systematic development. Some of its features can be summarized as follows: There is a world out there which we must navigate. Physical models are inference schemes; they are instruments to help us succeed. Within such models we find elements that purport to represent entities such as particles or fields. We also find other elements that are tools for manipulating information—probabilities, entropies, wave functions. All models inevitably involve concepts and categories of our own construction chosen for our own pragmatic reasons. To the extent that a model is reliably successful we say that its entities are real. Beyond that it is meaningless to assert that these entities enjoy any special relation to the world that might be described as ‘‘referring’’ or ‘‘corresponding’’ to something independent of us....Within the informational approach to science the notion of a structural realism in which the world consists of relations and structures only—without any entities orobjects—makes no sense. The information or lack thereof is information about something—both the entities and the informational tools must appear in the models. Having said that, the central point of the informational approach to physics (Jaynes1957a, b, 1983; 2011, 2012) is precisely that the formal rules for manipulating information—Bayesian and entropic methods—place such strong constraints on the formal structure of theories that a label of informational structural pragmatic realism may, in the end, be quite appropriate. Therefore I welcome Floridi’s reconciliation of the epistemic and the (noneliminativist) ontic versions of structural realism. Indeed, the recognition that a model describes a structure at a given level of abstraction brings Floridi close to Putnam’s internal realism. ‘Internal’ because the entities and structures in our world are defined within a given level of abstraction or conceptual framework which is chosen by us for our purposes. And ‘realism’ because the conceptual framework is not arbitrary; it is not independent of the world; the chosen entities and structures have to be useful and succeed in the real world. An important difference, however,is that Putnam’s internal realism makes no mention of information. This is the gap that can hopefully be closed by the informational pragmatic realism advocated here...undertake.The definition of information that I have proposed—information as a constrainton rational beliefs—differs from Floridi’s definition as well-formed, meaningful,and truthful data. But there is considerable overlap. Indeed, within the entropic/Bayesian framework a mere set of numbers—or data—does not by itself constitute information. It is necessary that the data be embedded within a model. This is what endows the data with significance, with meaning.
The laudable ambitiousness and interesting nature of Caticha's efforts aside (and keeping in mind that he is a working physicist and I am not a physicist), the lesson I want to take from them is that subjectivism about information leads to some extravagant strivings against ontic parsimony in theories and philosophical thinking (not that naturalistic philosophers like Floridi could not be accused of the same). I rather suspect that Caticha's efforts will prove to be useful and pragmatically valuable at least. I cannot speak to their applicability in physics nearly as coherently as Caticha himself. However, whether they lead us to a better understanding of the nature of information is perhaps less clear, and a point upon which I can offer insight.
I've not met a really good foil to objectivism about information and the nature of information yet, and I think that means that subjectivism about information - rather than telling us that information is intrinsically somehow subjectivist in nature - is telling us that our handling of information is often understandably subjectivist, and that all of the usual subjectivist and instrumentalist confusions and conflations between epistemic content and the objective ontic referents of that content also affect much of our treatment of the nature of information itself.
DNA and quasars would have something that scientists and their theories refer to as information - even if neither scientists nor any other cognitive or sentient agent or observer ever existed to believe anything about it.
References and Bibliography
Adami, C. (2004). Information theory in molecular biology. Physics of Life Reviews, 1(1), 3–22.
Adriaans, P. (2010). A Critical Analysis of Floridi’s Theory of Semantic Information. Knowledge, Technology & Policy, 23(1–2), 1–16.
Barbieri, M. (2008). Biosemiotics: a new understanding of life. Naturwissenschaften, 95(7), 577–599.
Barbieri, M. (2009). Three Types of Semiosis. Biosemiotics, 2(1), 19–30.
Barbieri, M. (2014). Introduction to Code Biology. Biosemiotics, 7(2), 167–179.
Beni, M. D. (2016a). Epistemic Informational Structural Realism. Minds and Machines, 26(4), 323–339.
Beni, M. D. (2016b). The Code Model of Biosemiotics and the Fate of the Structuralist Theory of Mental Representation. Biosemiotics.
Berto, F., & Tagliabue, J. (2014). The World Is either Digital or Analogue. Synthese, 191(3), 481–497.
Bringsjord, S. (2015). The symbol grounding problem . remains unsolved. Journal of Experimental & Theoretical Artificial Intelligence, 27(1), 63–72.
BUENO, O. (2010). STRUCTURALISM AND INFORMATION. Metaphilosophy, 41(3), 365–379.
Cao, T. Y. (2003a). Can We Dissolve Physical Entities into Mathematical Structures? Synthese, 136(1), 57–71.
Cao, T. Y. (2003b). Structural Realism and the Interpretation of Quantum Field Theory. Synthese, 136(1), 3–24.
Caticha, A. (2014). Towards an Informational Pragmatic Realism. Minds and Machines, 24(1), 37–70.
Chakravartty, A. (2004). Structuralism as a form of scientific realism. International Studies in the Philosophy of Science, 18(2), 151–171.
Chakravartty, A. (2015). Scientific Realism. In E. N. Zalta (Ed.), The Stanford Encyclopedia of Philosophy (Fall 2015). Retrieved from http://plato.stanford.edu/archives/fall2015/entries/scientific-realism/
Clifford, J., & Adami, C. (2015). Discovery and information-theoretic characterization of transcription factor binding sites that act cooperatively. PHYSICAL BIOLOGY, 12(5), 056004.
Collier, J. (2008). Information in Biological Systems (pp. 763–787).
Collier, J. (2011a). Explaining biological functionality : is control theory enough? South African Journal of Philosophy = Suid-Afrikaanse Tydskrif Vir Wysbegeerte, 30(1), 53–62.
Collier, J. (2011b). Explaining biological functionality : is control theory enough? South African Journal of Philosophy = Suid-Afrikaanse Tydskrif Vir Wysbegeerte, 30(1), 53–62.
Cover, T. M., & Thomas, J. A. (2006). Elements of information theory (Second). Hoboken, N.J: Wiley-Interscience.
Demirel, Y. (2014). Information in Biological Systems and the Fluctuation Theorem. ENTROPY, 16(4), 1931–1948.
Deutsch, D., & Hayden, P. (2000). Information flow in entangled quantum systems. Proceedings of the Royal Society of London. Series A. Mathematical, Physical and Engineering Sciences, 456(1999), 1759–1774.
Dorato, M. (2008). Is Structural Spacetime Realism Relationism in Disguise? The Supererogatory Nature of the Substantivalism/Relationism Debate. In D. Dieks (Ed.), The Ontology of Spacetime II (Vol. 4, pp. 17–37). Elsevier. Retrieved from http://www.sciencedirect.com/science/article/pii/S1871177408000028
Esfeld, M. (2013). Ontic structural realism and the interpretation of quantum mechanics, 3(1), 19–32.
Esfeld, M., & Lam, V. (2009). Structures as the objects of fundamental physics. In Epistemic objects : [research colloquium: “Epistemic objects”, May 16 - 17th 2008, Technische Universität Berlin] (pp. 3–16). Berlin: Max Planck Institute for the History of Science.
Esfeld, M., & Lam, V. (2010). Ontic structural realism as a metaphysics of objects. Retrieved from http://philsci-archive.pitt.edu/5531/
Floridi, L. (2003). From data to semantic information. Entropy, 5(2), 125–145.
Floridi, L. (2004). Outline of a Theory of Strongly Semantic Information. Minds and Machines, 14(2), 197–221.
Floridi, L. (2008a). A Defence of Informational Structural Realism. Synthese, 161(2), 219–253.
Floridi, L. (2008b). The Method of Levels of Abstraction. Minds and Machines, 18(3), 303–329.
Floridi, L. (2009). Against Digital Ontology. Synthese, 168(1), 151–178.
Floridi, L. (2010). Information, possible worlds and the cooptation of scepticism. Synthese, 175(S1), 63–88.
French, S. (2006). Structure as a Weapon of the Realist, 106(1), 169–187.
Gillies, D. (2010). Informational Realism and World 3. Knowledge, Technology & Policy, 1–18.
Godfrey-Smith, P. (2007). Information in Biology (pp. 103–119). Cambridge: Cambridge University Press. Retrieved from Harms, W. F. (2004). Information and meaning in evolutionary processes. New York; Cambridge, U.K: Cambridge University Press. Retrieved from
Hawking, S. W. (2005). Information loss in black holes. Physical Review D - Particles, Fields, Gravitation and Cosmology, 72(8). Retrieved from Jablonka, E. (2002). Information: Its Interpretation, Its Inheritance, and Its Sharing. Philosophy of Science, 69(4), 578–605.
Ladyman, J. (2011). Structural realism versus standard scientific realism: the case of phlogiston and dephlogisticated air. Synthese, 180(2), 87–101.
Ladyman, J. (2014, 2007). Structural Realism. In The Stanford Encyclopedia of Philosophy. Stanford University CLSI.
Landauer, R. (1991). Information is physical. Physics Today, 44(5), 23–29.
Levy, A. (2011). Information in Biology: A Fictionalist Account. Noûs, 45(4), 640–657.
Long, B. R. (2014). Information is intrinsically semantic but alethically neutral. Synthese, 191(14), 3447–3467.
Morganti, M. (2011). Is There a Compelling Argument for Ontic Structural Realism? Philosophy of Science, 78(5), 1165–1176.
Page, D. N. (1993). Information in black hole radiation. Physical Review Letters, 71(23), 3743–3746.
Psillos, S. (1995). Is Structural Realism the Best of Both Worlds? Dialectica, 49(1), 15–46.
Psillos, S. (2001). Is Structural Realism Possible? Philosophy of Science, 68(3), S13–S24.
Resnik, D. B. (1994). Hacking’s experimental realism. Canadian Journal of Philosophy, 24(3), 395.
Rovelli, C. (2016). Meaning = Information + Evolution.
Sarkar, S. (2000). Information in Genetics and Developmental Biology: Comments on Maynard Smith. Philosophy of Science, 67(2), 208–213.
Saunders, S. (2003). Structural Realism, Again, 136(1), 127–133.
Sdrolia, C., & Bishop, J. M. (2014). Rethinking Construction: On Luciano Floridi’s “Against Digital Ontology.” Minds and Machines: Journal for Artificial Intelligence, Philosophy and Cognitive Science, 24(1), 89.
Taddeo, M., & Floridi, L. (2007). A Praxical Solution of the Symbol Grounding Problem. Minds and Machines, 17(4), 369–389.
van Fraassen, B. C. (2006). Structure: Its Shadow and Substance. The British Journal for the Philosophy of Science, 57(2), 275–307.
Votsis, I. (2003). Is Structure Not Enough? Philosophy of Science, 70(5), 879–890.
Worrall, J. (1989). Structural Realism: The Best of Both Worlds? Dialectica, 43(1/2), 99–124.