Science and philosophy presuppose intelligibility
Science and philosophy both presuppose that reality is intelligible, meaning that it can be understood and explained by human cognition. This presupposition is necessary for several reasons, including the possibility of making sense of our experiences and the belief that there is an underlying order to the universe that can be discovered through investigation (Snyder, 2019).
One reason that science and philosophy presuppose that reality is intelligible is that it allows us to make sense of our experiences. As human beings, we are constantly interacting with the world around us, and our ability to understand and explain those interactions is essential for our survival and well-being. Without the presupposition of an intelligible reality, it would be difficult, if not impossible, to make sense of our experiences and to navigate the world in a meaningful way (Maurer, 2018).
Further, we believe that there is an underlying order to the universe that we can explore and access. This belief is based on the observation that the natural world is governed by laws and principles that can be described and predicted through scientific inquiry (Snyder, 2019). Without the presupposition of an intelligible reality, it would be difficult to believe that such laws and principles exist and that they can be discovered through investigation.
In short, if we don’t presuppose intelligibility, then all of science and philosophy are moot.
However, we’d ideally like to have a stronger logical argument than the above for the necessity of reality’s intelligibility. For that, we’ll examine language, perception, and their implications for both us and our technology.
Specifically, this paper argues that the structures of reality, perception, cognition, and natural and formal languages are isomorphic to one another. It is this isomorphism that logically ensures reality’s intelligibility, and provides an ontology for theories and models themselves. Moreover, that isomorphism allows us to learn about the nature of reality by studying linguistic syntax. Further, a key challenge facing the advancement of artificial intelligence (AI) is that of relevance realization, or the way in which an embodied conscious agent “reads” the language of reality, derives meaning from it, and acts upon that meaning, essentially serving as reality’s reflexive read-write functionality.
Evolution and perception
Donald Hoffman’s Interface Theory of Perception (ITP) posits that the objects and events that we perceive in the external world are not necessarily an accurate representation of reality, but rather are constructed as a means of efficiently interacting with the world (Hoffman, 1998). This theory is based on the assumption that evolution has shaped our perceptual systems to prioritize fitness over accuracy, meaning that our perceptions are designed to help us survive and reproduce, rather than to provide a completely accurate representation of reality (Hoffman, 2015).
For example, your senses tell you when your environment has too much or too little oxygen, rather than telling you the total quantity of oxygen present. If your senses informed you of the latter, it would be accurate, but largely useless to your survival fitness.
Instead, evolution shaped your senses to communicate fitness payoff information from your environment. That process factors in the state of the world, the state of the organism, the interactions between organisms, and the frequencies of their competitive strategies at any given moment of time and for any given arena. It is truthful and vital information, but translated into a meaningful string of data that a given organism can find intelligible and actionable.
One of the key implications of the ITP is the Fitness-Beats-Truth Theorem (FBT Theorem). This theorem suggests that, in any world of competition, an organism that sees the truth about the world and uses that knowledge to maximize fitness will always be outcompeted by an organism that sees none of the truth but is just tuned to fitness (Hoffman, 2019b). In other words, even if seeing the truth about the world would theoretically lead to better fitness outcomes, an organism that prioritizes fitness over accuracy will ultimately be more successful in an evolutionary sense.
The FBT Theorem has several implications for our understanding of perception. First, it suggests that our perceptions are shaped by the demands of the environment in which we evolved, rather than by any inherent accuracy of our perceptual systems (Hoffman, 2015). This means that our perceptions may be biased in certain ways that are not necessarily reflective of the true nature of the world.
Second, the FBT Theorem implies that our perceptions are optimized for action, rather than for knowledge (Hoffman, 2019b). This means that our perceptions are designed to help us interact with the world in a way that maximizes our chances of survival and reproduction, rather than to provide us with a complete and accurate understanding of the world.
Additionally, the ITP and the FBT Theorem suggest that our perceptions are highly individual and context-dependent. Different organisms, or even different individuals within a species, may perceive the same objects or events in vastly different ways, depending on their evolutionary history and the demands of their particular environment (Hoffman, 2015).
In other words, the ITP and the FBT Theorem have important implications for our understanding of perception. They suggest that our perceptions are shaped by the demands of the environment in which we evolved, and that they are optimized for action rather than knowledge. They also imply that our perceptions are highly individual and context-dependent, and may not necessarily reflect an accurate representation of reality.
Perception is a language
There is ongoing debate among scholars about the extent to which perception is linguistic in nature and structure. Some argue that language plays a fundamental role in shaping our perceptions of the world, while others contend that perception is largely independent of language. In this response, I will provide an overview of some of the arguments and evidence for the linguistic nature of perception, including its use of symbols, tense, associations, and subject-predicate attributions.
One argument for the linguistic nature of perception is based on the idea that our perceptions are organized around symbols. This idea is rooted in the work of linguist Benjamin Lee Whorf, who argued that language shapes our perception of the world by providing us with a system of symbols that allows us to categorize and organize our experiences (Whorf, 1956). According to this view, our perceptions of the world are fundamentally shaped by the symbols that we use to describe them.
In this case, every detail of our perceived world, down to the smallest level of each of our senses, can be considered a member of the perceptual language’s alphabet. These can then be combined to form strings and associations, from which we derive meaning that is relevant to our goals, chief of which has always been survival.
Another argument for the linguistic nature of perception is based on the role of tense in shaping our experience of time. Cognitive linguists argue that tense is not just a grammatical feature of language, but is also an essential component of our perception of time (Boroditsky & Ramscar, 2002). According to this view, our perceptions of events are structured around the temporal relationships between them, and language provides us with a way of organizing these perceptions into a coherent narrative.
Associations are also considered to be a crucial component of perception that is heavily influenced by language. Cognitive psychologists have shown that our perception of the world is shaped by the associations that we make between different sensory stimuli (Barsalou, 2008). These associations are often shaped by the linguistic context in which they occur, such as the words that are used to describe the stimuli.
Subject-predicate attributions are another aspect of language that is thought to shape our perceptions of the world. In his book “Philosophical Investigations”, philosopher Ludwig Wittgenstein argued that our understanding of objects and events is structured around subject-predicate attributions, which are themselves dependent on the structure of language (Wittgenstein, 1953). According to this view, our perception of objects is not simply a matter of seeing them as they are, but is instead shaped by the language that we use to describe them.
While there is ongoing debate about the extent to which perception is linguistic in nature and structure, there is evidence to suggest that language plays a fundamental role in shaping our perceptions of the world. This includes its use of symbols, tense, associations, and subject-predicate attributions, all of which map isomorphically onto perception.
Perception can then be seen as a language, in and of itself; one that is vastly complex but structurally isomorphic to the syntax of our natural and formal languages, which of course are based upon perception.
Furthermore, when paired with the ITP and FBT Theorem, we can extrapolate this idea of a perceptual language to all conscious agents, a list currently restricted to metabolizing organisms but that also has immense implications for AI. Recall that, according to Hoffman, different organisms, or even different individuals within a species, may perceive the same reality in vastly different ways, depending on their evolutionary history, their current state, and the demands of their environment.
In essence, each species, and even individuals within the same, may have different perceptual and cognitive alphabets of symbols that can be combined in associations to form meanings. Because all organisms behave as if they share the same reality, we can infer that the syntaxes of these respective perceptual and cognitive languages are isomorphic to each other, and also to the structure of reality itself to a non-trivial degree.
Affordances
In his work, John Vervaeke defines affordances as the possibilities for action that are inherent in the environment and that are available to an agent with the requisite capabilities (Vervaeke, 2016). This concept has its roots in the work of psychologist James Gibson, who argued that perception is an active process that involves the detection of the affordances that are present in the environment (Gibson, 1979).
Vervaeke expands on this idea by emphasizing the role of perception and action in the detection and exploitation of affordances. He argues that perception is not simply a matter of passively receiving sensory input, but is instead an active process of exploration and interaction with the environment (Vervaeke, 2016). According to Vervaeke, the perception of affordances is closely linked to the development of skills and expertise, as agents learn to detect and exploit the affordances that are relevant to their goals.
For instance, if I grasp a water bottle, I am affording it the attribute of being graspable. I am an agent who has that capability, and so I am able to realize the bottle’s graspability. By contrast, a spider cannot do so. However, the spider could afford the bottle the attribute of habitability, whereas I, an agent much larger than the bottle, could not. Meanwhile, the bottle simultaneously, reciprocally, and dialogically affords me the attribute of being a grasper. Such a relationship is isomorphic in structure to that of a subject-predicate coupling in linguistic syntax, whereby a predicate affords some attribute (including action) to its subject.
Vervaeke’s work on affordances draws on a wide range of sources from psychology, neuroscience, and philosophy. He cites the work of Gibson, as well as the ecological psychology tradition that has grown out of Gibson’s ideas. He also draws on the work of neuroscientist Walter Freeman, who has argued that perception and action are closely intertwined in the brain, with perception serving to guide action and action shaping perception (Freeman, 1991). Vervaeke also cites the work of philosopher Maurice Merleau-Ponty, who argued that perception is not simply a matter of sensory input, but is instead an embodied and situated process that involves the active exploration of the environment (Merleau-Ponty, 1962).
In other words, Vervaeke’s work on affordances emphasizes the active and exploratory nature of perception, and the close relationship between perception and action. This concept draws on a wide range of sources from psychology, neuroscience, and philosophy, including the work of James Gibson, Walter Freeman, and Maurice Merleau-Ponty. It also converges nicely with Hoffman’s ITP, FBT Theorem, and the linguistic nature of perception.
Relevance realization
Relevance realization is another concept introduced by Vervaeke that describes the process by which the brain identifies and prioritizes information that is relevant to a particular goal or context (Vervaeke, 2018). Relevance realization involves three main components: attention, meaning, and value, and it is central to cognition (Vervaeke, 2017).
The first component of relevance realization is attention. The brain is constantly bombarded with a combinatorially explosive amount of sensory information, and attention allows the brain to selectively attend to the most relevant information (Vervaeke, 2017). The second component is meaning, which involves integrating the attended information with one’s existing knowledge and understanding of the world to create a coherent and meaningful representation of the information (Vervaeke, 2017). The third component is value, which involves evaluating the relevance of the information in relation to the individual’s goals or needs, and using that evaluation to prioritize actions (Vervaeke, 2017).
Relevance realization is a fundamental cognitive process that enables individuals to navigate the complex and dynamic world around them (Vervaeke, 2018). It allows individuals to focus their attention, make sense of information, and prioritize their actions in a way that is meaningful and goal-directed.
In essence, relevance realization is our capacity to “read” the language of reality via our perceptual and cognitive syntaxes, to assign meaning to the associations and subject-predicate attributions therein, and to act upon that information.
Perception translates information into a language we can “speak”
The claim that “because we have survived, our perception must give us truthful information” is a common intuition, but it is not necessarily a sound argument. While our survival as a species may suggest that our perception has been useful for navigating the world, it does not necessarily imply that our perception is always accurate or truthful, only that it is at least non-trivially partially truthful.
In fact, research in cognitive psychology has shown that our perception can be highly fallible, and that our brains often rely on heuristics or shortcuts to make sense of complex sensory information (Kahneman, 2011). These heuristics can lead to cognitive biases and errors in judgment, which can have serious consequences for our decision-making and well-being.
Furthermore, our ability to survive as a species is not solely dependent on our individual perception, but also on social and cultural factors, as well as luck and chance events. As such, it is possible that our perception has evolved to be adaptive in some contexts but not in others, or that our survival has been achieved despite our perceptual limitations rather than because of them.
In short, the fact that we have survived as a species does not necessarily guarantee the truthfulness or accuracy of our perception, and we must be cautious in assuming that our perception always gives us a reliable picture of the world. We should take perception seriously, but not literally.
This reinforces the idea that perception is a language, or a carrier of information. Some critics suggest that the ITP entails that our perception gives us no accurate information, but this is not correct – our survival does imply that perception gives us true information, if not fully accurate. We could then suppose that ITP entails that we receive partial truth, but this isn’t quite right either.
Instead, the implication of ITP is that our perception simplifies the information of reality, translating it into a language that we can “speak” (so to speak) in order to more efficiently accomplish tasks beneficial to our survival.
For instance, when I play a video game and enter its virtual world, that world gives me truthful but simplified information about the reality underlying it: the 1s and 0s, the transistors, etc. The states of the virtual world do provide truth about the states of that underlying nature. However, the virtual world is an interface that translates that complexity, which I could not easily find intelligible without expending tremendous energy, into a simplified perceptual language that is readily intelligible and therefore actionable.
With this in mind, it is no surprise that our perception is not always accurate. Think of how easy it is to have details get “lost in translation” when exchanging information across languages. Of specific concern in such work are the figures of speech, or linguistic heuristics (compare this to the perceptual heuristics mentioned above) that native speakers use in order to more quickly convey information.
However, because we have survived using our evolved perception, it logically follows that our perceptual language carries to us translated, simplified, non-trivially truthful information about reality.
Therefore, there is what we’ll call a weak isomorphism between perception, cognition, and reality. It makes reality intelligible by providing us with a simplified, fitness-tuned, approximate representation of reality, as opposed to a strong isomorphism that would make reality comprehensible by providing a 1:1 representation.
In other words, the weak isomorphism of our perception with reality makes reality intelligible, but not comprehensible.
On natural and formal languages
Natural language is a system of communication used by humans to convey meaning and express thoughts and ideas. It is characterized by its complexity, ambiguity, and variability. Moreover, it constantly evolves and changes over time, coupled with the culture that employs it (Chomsky, 1965). Examples of natural languages include English, French, Spanish, and Chinese.
Formal language, on the other hand, is a specific type of language designed for a particular purpose or application. It is typically more precise, unambiguous, and well-defined than natural language and is often used in areas like mathematics, logic, and computer programming. Natural languages may become formalized (Sipser, 2013).
Next, we’ll look at specific considerations regarding natural and formal languages, and their effectiveness at carrying true information about reality.
Language evolved with and from perception
Language evolved out of our perceptual abilities. This hypothesis suggests that early humans used their perceptual abilities to communicate with each other, and over time, this communication evolved into the complex system of language that we use today (Hurford, 2011).
One key aspect of this hypothesis is that perception provides the foundation for many of the features of language, as already discussed in this paper. For example, the ability to recognize and categorize objects in the environment is a fundamental aspect of perception, and this ability is reflected in the way that language uses categories and labels to describe the world around us (Lakoff, 1987). This is due to the linguistic nature of perception, providing an isomorphism between perceptual syntax and the syntaxes of languages that evolved out of and alongside perceptual faculties.
Another aspect of this hypothesis is that the evolution of language was closely linked to the development of the brain. The ability to use and understand language requires a high level of cognitive processing, and it is likely that the evolution of language was closely tied to the expansion and development of the human brain (Deacon, 1997).
Overall, the idea that language evolved out of our perceptual abilities suggests that language is deeply rooted in our experience of the world around us. Our ability to perceive and categorize the environment provided the foundation for the development of language, and the evolution of language was closely linked to the development of the human brain.
Language shapes cognition
Cognition and language are closely intertwined, and each one has an impact on the other. Language provides a means for individuals to acquire knowledge, communicate with others, and form abstract concepts. In turn, cognition plays a crucial role in the acquisition and processing of language.
According to the Sapir-Whorf hypothesis, language shapes the way people think, and the structure of a language can influence how individuals perceive the world around them. For instance, the English language has distinct words for colors such as “blue” and “green,” while some languages such as Tarahumara do not differentiate between these two colors, instead using a single term for both. Research has shown that speakers of languages with fewer color terms tend to have more difficulty distinguishing between different shades of colors (Winawer et al., 2007).
Moreover, language can influence the way people categorize objects and events. For example, speakers of Mandarin Chinese tend to group objects together based on their functional relationships, while English speakers tend to group objects based on their perceptual features (Boroditsky, 2001).
Cognition also plays a vital role in language processing. The ability to reason, plan, and problem-solve all depend on cognitive processes, and these processes are involved in language comprehension and production. Research has shown that cognitive abilities, such as working memory, attention, and executive function, are essential for successful language learning (Gathercole & Baddeley, 1993).
In these ways, reality, perception, cognition, and languages all shape each other in a dialogic, reciprocal manner. They converse with each other by transducing information between them, thereby recursively evolving together. In so doing, they maintain a syntactical isomorphism to the structure of reality that ensures and preserves their utility, and, ultimately, benefits the survival of the conscious agents who employ them.
Reality cannot be intelligible without this dialogue and the resulting isomorphism. In short, to deny the reality-perception-cognition-language isomorphism is to abandon science and philosophy as meaningful projects. Since both have been successful at discovering and working with reality, the isomorphism must hold true. It is the dialogic, linguistic interplay between reality, perception, cognition, and languages that brings about those syntactic similarities.
Generative grammar and predictive world modeling
The structure of reality and the structure of language have been compared in various ways, with some researchers drawing parallels between the generative grammar theory of language and predictive world modeling theories of the brain. Generative grammar posits that language is structured according to a set of rules or principles that generate an infinite number of possible sentences (Chomsky, 1965). Predictive world modeling theories of the brain suggest that the brain constructs internal models of the external world that allow it to predict future events and generate actions (Clark, 2013).
One way in which these two theories can be compared is in terms of their generative capacity. Just as generative grammar can generate an infinite number of possible sentences, predictive world modeling theories suggest that the brain is capable of generating a vast number of possible future scenarios based on its internal models of the world (Friston, 2010).
Another way in which these two theories can be compared is in terms of their hierarchical structure. Generative grammar posits that language is structured hierarchically, with larger units of meaning built up from smaller ones (Chomsky, 1957). Similarly, predictive world modeling theories suggest that the brain constructs hierarchical representations of the external world, with higher-level representations built up from lower-level ones (Clark, 2013).
Moreover, both theories rely on probabilistic models. Generative grammar posits that language is probabilistic, meaning that the probability of a particular sentence being grammatically correct can be calculated based on its adherence to the rules of the grammar (Chomsky, 1957). Predictive world modeling theories also rely on probabilistic models, as the brain must constantly make predictions about the likelihood of future events based on the available sensory information (Friston, 2010).
In other words, there are several ways in which the structure of reality can be compared to the structure of language, with some researchers drawing parallels between generative grammar and predictive world modeling theories of the brain. Both theories rely on generative capacity, hierarchical structure, and probabilistic models to generate and make sense of complex information.
Implications for physics and metaphysics
Moreover, the similarities between the structures of generative grammar and reality find parallels in physics and metaphysics, with respect to cosmological questions such as the origins of reality. For instance, any reality theory that describes reality as evolving from a ground state of potential and exploring all possible options necessarily displays an isomorphism to generative grammar, which entails the same kind of recursive process in language.
The implication is that reality (the set of everything that is real) is, by definition, self-contained and self-generating, with no external factors or entities necessary for its existence. It is capable of both generating and interpreting its own language and meaning – since there is nothing else besides reality, nothing else could perform these functions. In other words, since reality is intelligible to us, and since we are part of reality, reality must be intelligible to itself.
Reality is, therefore, capable of infinitely complex self-reference and self-description. Any process of identification, such as this self-actualization and self-realization, entails distinguishing “some-thing” from its logical complement. Indeed, “no-thing” is ever truly realized without its complement to provide logical context.
The conception of opposites supports human thinking in a number of ways, including our “everyday counterfactual thinking, classic deductive and inductive reasoning tasks and the representational changes required in certain reasoning tasks … it follows that opposites can be regarded as a general organizing principle for the human mind rather than simply a specific relationship (however respectable) merely related to logics” (Branchini et al 2021).
In other words, we make sense of the world by creating dualities, such as good and evil, hot and cold, tall and short, etc. We mentally position these pairs as opposites, allowing us to reason and grok important information about our arena.
For instance, we use the hot-cold dichotomy in order to know if the temperature of an entity or of the environment at large is dangerous or suitable to our survival. A hot stove delivers negative fitness payoffs. So does a frozen lake.
The dangerous properties of a hot stove and a frozen lake are not properties of the “things” in themselves, but rather are only realized as such once we, conscious agents, enter into a reciprocal, dialectical, agent-arena relationship with the things in themselves. For instance, many other organisms are able to survive intense heat or cold, but both the hot stove and frozen lake are outside the temperature range that humans need. Thus, the agents and the arenas co-realize each other, and that relationship is “re-presented” in our perceptual and cognitive frameworks as icons (physicality) and as the conceptual notions of “things” and opposites.
Duality implies the separate ontic existence of the two entities making up the dichotomy. In order for them to be opposed, surely they must exist independently of one another as two distinct “things.”
However, we instead find a more complex, self-realization of the conceptual, in which “thing-ness” is merely nominal, just as it was for the material. The “things” once again reciprocally realize each other in a kind of dialectical relationship, not so much opposing each other as depending on each other’s co-existence, and ultimately on a shared unity (McGill & Parry 1948; Lincoln 2021; Vervaeke & Mastropietro 2021), in order to be realized, and thus made real.
In all cases, we get back to the logical necessity that reality, as the only “thing” that exists, must realize itself in order to be real. It is in this way that conscious agents, as part of reality, “read” the language of reality, thereby fulfilling the role of self-contained reality’s self-identification and self-actualization (Campbell, 2003; Hoffman, 2019a; Kastrup, 2019; Azarian, 2022; Santos, 2022).
Quantum physics is then best interpreted along the lines of Carlo Rovelli’s relational model (Rovelli, 1996), and Markus Müller’s physics of the first-person perspective (Müller, 2023), both of which are consistent with the previously referenced interpretations that support non-locality and contextuality.
In that case, and consistent with the ITP and the FBT Theorem, quantum physics tells you about the probability of each outcome and what you will perceive next as an observer. It answers the question, “What will I observe to be the next state of the world?”
We can resolve the mysteries of the wave function under this model as well; it is not that consciousness collapses the wave function, as some propose. That statement implies a kind of ontological dualism, in which consciousness and the physical wave function are both ontic entities. This is not so, because reality logically must be one ontic entity (for instance, any two real “things” and a given real difference between them all share the similarity of being real, meaning they are part of an ultimate reality, the “One”).
Instead, because spacetime and the physicality that we perceive are like an interface, they should be considered epistemic entities, not ontic entities. Doing so resolves the quantum paradoxes that have plagued physics (the specifics are beyond the scope of this paper, but the reader should explore the work of the previously referenced physicists and others, such as Nima Arkani-Hamed).
In other words, quantum processes are artifacts of our “reading” the language of reality. They result from our perceptual and cognitive frameworks translating that vast stream of informational input into an intelligible language that gives us simplified truth about what state of reality will come next. That intelligible language is physicality and spacetime, complete with all of the linguistic syntax of perception, which is then isomorphic to our natural and formal languages.
It then logically follows that the only viable metaphysics is idealism (Campbell, 2003; Kastrup, 2019; Santos, 2022).
Explaining the unreasonable effectiveness of mathematics in the natural sciences
The “unreasonable effectiveness of mathematics” in the natural sciences is a phrase coined by the physicist Eugene Wigner in his 1960 paper of the same name (Wigner, 1960). It refers to the striking ability of the formal language of mathematics to accurately describe and predict natural phenomena, even when there seems to be no inherent connection between the two.
Mathematics has proven to be remarkably effective in describing and predicting natural phenomena across a wide range of fields, from physics and engineering to biology and economics. For example, the laws of physics are expressed using mathematical equations, and these equations have been able to predict a wide range of phenomena, from the behavior of subatomic particles to the motions of planets.
There are many theories as to why mathematics is so effective in the natural sciences. Some argue that it is because mathematics is a fundamental aspect of the universe itself, and that the laws of physics and mathematics are ultimately the same thing. Others suggest that mathematics is effective because it provides a powerful way to abstract and simplify complex phenomena, allowing scientists to focus on essential features and ignore irrelevant details.
However, this paper provides a simpler explanation by which to resolve this paradox: the structure of reality is isomorphic to the structure of perception, upon which our natural and formal languages are based, and with which the syntaxes of our natural and formal languages are isomorphic. Therefore, the structure of the formal language of mathematics is, via a kind of transitive property, isomorphic to reality, making reality intelligible but not comprehensible. It then logically follows that mathematics is effective at describing reality precisely because of that isomorphism.
Using the context of metaphysics, we can phrase this another way:
Physicality, which the formal language of mathematics describes, is how reality appears to itself when perceived from within itself.
More specifically, the physical is what consciousness looks like when perceived.
Gödel’s Incompleteness Theorem: intelligibility but not comprehensibility
Gödel’s Incompleteness Theorem is a fundamental result in mathematical logic that has important implications for the limits of knowledge and the possibility of a theory of everything. The theorem states that in any formal system that is powerful enough to express basic arithmetic, there will be true statements that cannot be proven within that system (Gödel, 1931).
This has profound implications for the search for a theory of everything, which is the quest to find a single theory that explains all of reality. The Incompleteness Theorem suggests that even if we were to find such a theory, it would be incomplete, because there would always be true statements about reality that could not be proven within that theory.
This is because any theory of everything would be a formal system, and Gödel’s theorem applies to all formal systems that are powerful enough to express basic arithmetic. As such, the theorem implies that there are limits to what we can know and prove about the universe using formal systems and mathematical logic alone.
In addition, Gödel’s theorem has also been interpreted to suggest that there are limits to what can be computed and predicted using algorithms and computers. This is because any algorithm or computer program can be viewed as a formal system, and Gödel’s theorem implies that there will always be true statements that cannot be proven or computed by that system.
Gödel’s conclusions for formal systems, which utilize formal languages, converge with what the ITP and FBT Theorem tell us about perception and the perceptual language. In other words, reality is intelligible to us, but not comprehensible to us.
An ontology of theories and models
We now have all of the premises needed to form a logical argument for the intelligibility of reality.
We evolved our perceptual apparati to provide true, simplified information about fitness payoffs in our external state. Our perception has all of the aspects of a language, including a linguistic syntax, and performs the function of a language: carrying information. Since we have been successful at surviving, it logically follows that the information carried by our perception is truthful. In other words, the structure of our perception is isomorphic to reality.
Our natural and formal languages evolved out of and are based upon our perception. As a result, our linguistic syntaxes are isomorphic to our perceptual syntax.
Because our perception is isomorphic to reality, and because our natural and formal languages are isomorphic to perception, therefore our natural and formal languages are isomorphic to reality. As a result, the theories and models that we develop using those natural and formal languages are capable of sharing that isomorphism and accurately exploring the nature of reality.
And if that’s the case, then reality is intelligible to us, and our theories and models have an ontology of their own.
The challenge of relevance realization for AI
The challenge of relevance realization for AI is the ability for machines to identify and comprehend the significance of information in a particular context.
Recall that relevance realization is a fundamental cognitive process that allows humans to understand and navigate the world around them. This process involves identifying relevant information, integrating it with existing knowledge, and using it to guide behavior and decision-making. However, Vervaeke notes in lectures and commentary that current AI systems struggle to achieve this same level of relevance realization, as they often rely on static rules or algorithms that are unable to adapt to changing contexts. For instance, how could we ever program a computer to not just zero in on the pertinent details of a situation, but also to ignore the combinatorially explosive number of other details and combinations of details (Vervaeke, 2020)?
In other words, when we’re reading a natural language on paper or on a screen, we’re able to focus on just the symbols and strings relevant to us at any given moment. We can ignore the rest of the alphabet, the rest of the potential strings, etc., allowing us to find only the words before us salient.
In the case of our perceptual language, we do the same thing, but with a vastly more complex linguistic system. For instance, every fine detail of your sense data could be considered a symbol of the alphabet (giving us far more than English’s modest 26 to choose from), and there are practically speaking infinitely many combinations of them. The task of “reading” only what is relevant and ignoring everything else becomes far greater when accessing the information of reality through the language of perception.
Furthermore, Vervaeke argues that relevance realization is closely tied to our embodied cognition, or our ability to use our physical bodies and sensory experiences to interact with the world. This embodied cognition is difficult to replicate in AI systems, which typically operate in a purely symbolic or computational domain (Vervaeke, 2020).
The challenge of relevance realization for AI has important implications for the development of intelligent machines. Without the ability to perceive and understand the significance of information in a particular context, AI systems may struggle to make sense of complex, dynamic environments. As such, researchers are working to develop new approaches to AI that incorporate more embodied, context-sensitive forms of cognition (Vervaeke, 2020).
In other words, if reality’s structure is isomorphic to the syntax of our languages, including the language of perception, then relevance realization is our way of “reading” the language of reality. In this sense, reality itself can be viewed as a formal system.
For AI, it remains to be seen if relevance realization can be programmed or learned. Rather, because it seems dependent on an embodied conscious agent with an evolutionary history, perhaps relevance realization must be evolved.
In that case, the scales tip in favor of evolutionary models of AI development, though in practical terms it remains to be seen how effective such a model would be. After all, the natural evolutionary processes that shaped the cognition, perception, and relevance realization of metabolizing organisms has taken an immensely long time, and it is an open question whether or not the guidance of a human engineer could speed up the process in AI.
Until AI has the ability to “read” the vastly complex language of reality, as we do, it will be missing a core aspect of our cognitive functions. Moreover, that restriction will also limit its potential actions in the world, or its ability to “write” reality, which has implications for its capacity to be a truly general problem solver.
Therefore, AI must (likely through an evolutionary process) develop relevance realization in order to perform the reflexive read-write functions of reality that metabolizing conscious agents do.
Bibliography
Azarian, B. (2022). The Romance of Reality: How the Universe Organizes Itself to Create Life, Consciousness, and Cosmic Complexity. Dallas, TX: BenBella Books.
Barsalou, L. W. (2008). Grounded cognition. Annual review of psychology, 59, 617-645.
Boroditsky, L. (2001). Does language shape thought? Mandarin and English speakers’ conceptions of time. Cognitive Psychology, 43(1), 1-22.
Boroditsky, L., & Ramscar, M. (2002). The roles of body and mind in abstract thought. Psychological science, 13(2), 185-189.
Branchini E., Capitani E., Burro R., Savardi U., Bianchi I. (2021). Opposites in Reasoning Processes: Do We Use Them More Than We Think, but Less Than We Could? Front Psychol. 2021 Aug 26;12:715696. doi: 10.3389/fpsyg.2021.715696. PMID: 34512474; PMCID: PMC8426631.
Campbell, T. (2003). My Big TOE: A Trilogy Unifying Philosophy, Physics, and Metaphysics. Lightning Strike Books.
Chomsky, N. (1957). Syntactic structures. Mouton.
Chomsky, N. (1965). Aspects of the theory of syntax. MIT Press.
Clark, A. (2013). Whatever next? Predictive brains, situated agents, and the future of cognitive science. Behavioral and Brain Sciences, 36(3), 181-204.
Deacon, T. W. (1997). The symbolic species: The co-evolution of language and the brain. W.W. Norton & Company.
Freeman, W. J. (1991). The physiology of perception. Scientific American, 264(2), 78-85.
Friston, K. (2010). The free-energy principle: A unified brain theory? Nature Reviews Neuroscience, 11(2), 127-138.
Gathercole, S. E., & Baddeley, A. D. (1993). Working memory and language. Lawrence Erlbaum Associates.
Gibson, J. J. (1979). The ecological approach to visual perception. Houghton Mifflin.
Gödel, K. (1931). On formally undecidable propositions of Principia Mathematica and related systems. Monatshefte für Mathematik und Physik, 38(1), 173-198.
Hoffman, D. D. (1998). Visual intelligence: How we create what we see. W. W. Norton & Company.
Hoffman, D. D. (2015). The case against reality. Aeon. Retrieved from https://aeon.co/essays/the-case-against-reality
Hoffman, D. D. (2019a). The case against reality: Why evolution hid the truth from our eyes. W. W. Norton & Company.
Hoffman, D. D. (2019b). The fitness beats truth theorem. Quanta Magazine. Retrieved from https://www.quantamagazine.org/the-fitness-beats-truth-theorem-20190320/
Hurford, J. R. (2011). The origins of grammar: Language in the light of evolution II. Oxford University Press.
Kahneman, D. (2011). Thinking, Fast and Slow. New York: Farrar, Straus and Giroux.
Kastrup, B. (2019). The Idea of the World: A multi-disciplinary argument for the mental nature of reality. iff Books.
Lakoff, G. (1987). Women, fire, and dangerous things: What categories reveal about the mind. University of Chicago Press.
Lincoln, C. (2021). The Dialectical Path of Law. United States: Lexington Books
Maurer, K. (2018). The intelligibility of reality: How science and philosophy reveal the universe. The Imaginative Conservative. Retrieved from https://theimaginativeconservative.org/2018/01/intelligibility-reality-science-philosophy-reveal-universe-kenneth-maurer.html
McGill, J. & Parry, W. T. (1948). The Unity of Opposites: A Dialectical Principle. Science & Society, vol. 12 no. 4 (Fall 1948), pp.418-444.
Merleau-Ponty, M. (1962). Phenomenology of perception. Routledge.
Müller, M. (2023). The physics of first-person perspective: an introduction by Dr. Markus Müller. (n.d.). www.youtube.com. Retrieved January 16, 2023, from https://www.youtube.com/watch?v=cAUpmd_gGMM.
Rovelli, C. (1996), “Relational quantum mechanics”, International Journal of Theoretical Physics, 35: 1637–1678.
Santos, M. (2022). The Melody of Reality: A theory of life, death, the universe, and everything. Raleigh, NC: Bad Cat Press.
Sipser, M. (2013). Introduction to the theory of computation. Cengage Learning.
Snyder, T. (2019). Why science presupposes the intelligibility of nature. Big Think. Retrieved from https://bigthink.com/errors-we-live-by/science-intelligibility-nature
Vervaeke, J. (2016). The relevance realization revolution. Journal of Consciousness Studies, 23(9-10), 139-165.
Vervaeke, J. (2017). Awakening from the meaning crisis [Video file]. Retrieved from https://www.youtube.com/watch?v=54l8_ewcOlY&t=3830s
Vervaeke, J. (2018). On the nature of relevance realization. Journal of Consciousness Studies, 25(5-6), 6-36.
Vervaeke, J. (2020, April 8). The Meaning Crisis (Part 6) – Relevance Realization & Cognition. [Video]. YouTube. https://www.youtube.com/watch?v=H9Kj6rRm1Q4&t=1194s
Vervaeke, J. & Mastropietro, C. (2021). Dialectic into Dialogos and the Pragmatics of No-thingness in a Time of Crisis. Eidos. A Journal for Philosophy of Culture 5 (2):58-77.
Whorf, B. L. (1956). Language, thought, and reality: Selected writings of Benjamin Lee Whorf. John Wiley & Sons.
Wigner, E. P. (1960). The unreasonable effectiveness of mathematics in the natural sciences. Communications on Pure and Applied Mathematics, 13(1), 1-14.
Winawer, J., Witthoft, N., Frank, M. C., Wu, L., Wade, A. R., & Boroditsky, L. (2007). Russian blues reveal effects of language on color discrimination. Proceedings of the National Academy of Sciences, 104(19), 7780-7785.
Wittgenstein, L. (1953). Philosophical investigations. Blackwell Publishers.
