A theory cannot be more real than the experiences it is based on, or? ====================================================================== https://medium.com/@sila_22801/a-theory-cannot-be-more-real-than-the-experiences-it-is-based-on-or-3ea3c1bbbc4b Simon Laub July 18th, 2025. Experiencing the world precedes being able to practice science, authors Adam Frank, Marcelo Gleiser, and Evan Thompson tell us in the book ''The Blind Spot''. The book, the Blind Spot, is a tour de force that guides us around the many intractable puzzles that have followed, when theory no longer remembers the experience that started it… Sure, it is a very obvious insight: Everything we know is based on who we are. From the books preface: The authors want a science that includes — rather than ignores, or tries not to see — humanity's lived experience as an inescapable part of our search for objective truth. The authors sees science not as discovering absolute reality, but rather as a highly refined, constantly evolving form of human experience. - - - We should remember, that when we experience the world, we are not really in the objective position of looking at something from the outside: - In cosmology we can only know the Universe from inside. Not from an outside position. - When it comes to our own experience of the world. We can not fathom consciousness without experiencing it from within. P. VIII. And we must be careful not to confuse the map with the terrain. To think that idealised laws of mathematical physics describe inherently being of nature is fundamentally mistaken. - To think this way is to confuse the map with the territory. In the author words: ''Confusing an instrument of prediction for how things are in themselves. It is a kind of category mistake''. P. 9. E.g. we should be careful when theories substitute "clock time" for duration, computation for meaning, information for consciousness etc. [P. 24]. Intractable problems are likely to follow… And we should think carefully, when we are asked to take an ''objective view'', not simply independent of subjectivity , publicly checkable, and invariant under transformation of perspective, but a view from nowhere and no when. A ''gods-eye'' view that our own experience can't give us [P. 49]. - - - Time. When it comes to time, the authors warn us that we must experience time as duration, before we can abstract time as time in a space-time diagram… We can't make sense of ''clock time'' without referring to something that is not a reading of a clock. P. 113. But, sure time is a very tricky thing: The brain constructs the perception of the duration of the present moment by careful integration of asynchronous sensory and motor activities. Since sensory stimuli, such as sound and light, travel at different speeds, as do exteroceptive and interoceptive signals along nerve paths. Every event that we perceive as happening ''now'' reflects a cognitive performance by the brain. P. 63. About the ''blockworld'' theory of time, with no flow of time, the authors write: The block universe is all of spacetime in a unchanging four-dimensional block. Here the objective world simply is. It does not happen. There is no time flow. But it is only possible, if we adopt an impossible "gods-eye" perspective. External to the Universe and passage of nature... The ''blockworld'' theory robs nature of passage and time of duration. ... eliminating the becoming inherent in time. The block universe confuses a mathematical picture with what is being pictured, it confuses the map with territory. P. 123. According to the authors: ''To imagine that we can abstract away from the flow of time to obtain an objective view of reality — Is to replace the passage of nature with a lifeless symbol''. P.124. You might think that time is an illusion, and that it is an illusion that we are living in time’s flow, right now, in the process of reading these words. But then you will likely have to go with theory that tell you that time is really a variable in equations. A (theoretical) time that is highly unlikely to go backwards (other equations tell us why), but potentially can. I.e. Prominent physicists might tell us that time is a variable in equations, a series of measurable instants, and while this series is highly unlikely to go backwards (other equations tell us why), it potentially can. If time seems like a forward flow to us, it is because our imperfect mental and bodily instruments blur what's really happening. (But) We can always observe the system for arbitrarily long timescales, and at these long times, an occurrence of a state that effectively decrease the entropy (making time go backwards), and reverts the system to its initial low-entropy state will occur... This perspective on Loschmidts paradox is called Zermelos recurrence objection... But a problem comes with how we interact with such natural systems, which is always coarsegrained. And the assumed infinite precision in some of these models. Leaving our felt experience can easily lead us to (non physical) scenarios, outside what we can actually experience..? Again, we should be careful, when we are asked to forget the role of experience, as the ground of all scientific practice. And “move forward with timeless laws, viewable from a perfectly objective, perfectly perspectiveless perspective, a view from nowhere” P. 107. - - - Biology and artificial life. Understanding biology, and how it differs from artificial life, also needs to be based on something we actual experience. An AI system is not alive... It is purely computational intelligence. With no living body and no metabolism, and hence with no need to constantly renew and maintain itself. Its knowledge is limited by its having a certain kind of world model (A cognitive ontology - a specification of what kind of things exist in the world and the relations between them). There are many things such an artificial system could not possibly know: How would it be able differentiate between a healthy process and a pathologic one? There is no distinction between normal and pathological in physics and mechanics...? P. 141. Concerning the difference between organisms and machines the authors write: To be alive is to be self-individuating. To cleave existence into the dyad of self-versus-world. A living system constantly constructs itself, thereby making itself distinct from its environment while altering it's environment in ways conductive to its continuation. In other words, a living system possesses autonomy and agency: It produces, maintains and regenerates the parts and processes that constitute it's function as an integrated, self governing whole. While also promoting environmental conditions favourable to its own existence and actively avoiding conditions that threaten it (agency) P. 149. Living is sense making. Autonomy and agency imply sensemaking. Organisms are sensemaking beings. They create world of relevance. They engage with environments that are structured by their actions and by what is relevant to their viability. Living beings partition their environments into things that have positive or negative value, or that are neutral. Or have no significance at all (That fall outside the sphere of significance altogether). In biologist Varelas words: "Living is sense making". The purpose or function of a machine is extrinsic. It lies outside the machine in its maker or user. In an organism, however, the properties of the parts (the functional components) depend on the whole. ... In addition, an organism acts on its own behalf. So it is intrinsically purposive: It acts towards its own ends. Such as maintaining it's viability in precarious conditions. Thus, intrinsic purposiveness derives directly from biological autonomy and agency. P. 157. Life most salient features, such as autonomy and agency, are split of as epiphenomenon in mechanistic treatments of life. P 143. And, whenever we set up a framework, for understanding life, we should remember that life has a tendency to escape the framework: According to physicalism and reductionism the fundamental laws of the physical govern the spatiotemporal arrangements of matter and energy at the lowest levels. But life - to say nothing of experience - already escapes this framework. If the evolving biosphere involves an ever changing phase space that continually open up new ''adjacent possible'' pathways, then life generates new forms and possibilities in ways that the metaphysics of physicalism and reductionism can not capture. ... I.e there can be no "final theory" for the evolution of a Universe having at least one evolving biosphere. P. 162. Certainly, when it comes to life, we should always notice what can ''experience'' and what can not: Chatgpt does not know what it is actually like to be in the world. Just as face-recognition systems have no knowledge that the data structures pertain to images of faces. Let alone real people that exist in the world. P. 177. We should notice how these systems are set up: Ai systems are not autonomous, rational, or able to discern anything without extensive, computational training with large datasets or predefined rules and rewards. P. 178. Not seeing this, and the rhetoric concerning this, is a blind spot. Cognition and a sense of self also goes back to (self) experience: Cognition is not about transposing a world of predefined significance into the inside of an agent. It is about agents moving within the world and singly or collectively changing it ways that are significant according to the forms of life they enact. P.179. The sense of self is not one thing. It includes many different elements. Such as the bodily self. The mental self experienced in memory and anticipation, the self as a personal storyline (the narrative self), and the social self. The sense of self is a construction, made from integrating different kinds of self-experience. P.197. - - - Consciousness. Theories about consciousness should tell us how consciousness is created: About the ''Bayesian brain'': As neuroscientist Luiz Pessoa remarks. The ''Bayesian brain'' concept is inflated. No one calls the brain the ''Newtonian brain'', because we can use differential equations to model it. So, just because we can use Bayesian probability theory to model aspects of brain function doesn't make the brain a "Bayesian brain". The map is not the territory. And neither is the math. P. 201. The issue is whether the theory explains consciousness by explaining how the brain generates experience. Does the theory solve the "generation problem" for consciousness. The problem of how a physical system can generate conscious experience. The answer is no... P. 201. In the early days of AI the field was based on assumptions that turned out to be problematic. That the world consists of determinate states, objects and properties, and events (like the positions, pieces, and moves in a game) and that perception requires having a computational representation of them inside the system. It was assumed that intelligence could be built into a system by making it perceive discrete, well defined as mesoscale objects exemplifying properties and standing in unambiguous relations. But we do not perceive the world in these terms. We perceive the world as made up of open-ended, fluid situations that solicit and afford actions. P.175. Indeed, a better understanding of our own perception, and experience , should have made it clear that these early AI systems was indeed built on problematic assumptions. - - - All models are wrong, but some are useful, as George E. P. Box would have said. In order not to get into too many intractable problems, the authors definitely have a point, we should not forget humanity's lived experience, as an inescapable part of our search for objective truth. An awesome book, indeed. ---- Simon Laub. July 18th, 2025.