Human GOLEM/TDE
Each human subject (ie the 'self') consists of a single global (human-level 3) GOLEM, which consists of an asymmetric pairing of the left and right local (animal-level 2) models.
The combination of the two lower-level animal GOLEM’s creates the single higher-level human GOLEM. Figure 4.1 below depicts this conceptual relationship.
It is this aspect of GOLEM architecture which successfully explains the most notable feature of the brain's neuroanatomy, namely its marked left-right lateralisation of function. The success of this hybrid model suggests that human evolution consisted of a series of small incremental transformations that occurred over evolutionary time, in which the two cerebral hemispheres became increasingly functionally divergent (specialised). As noted by [6], our own logical reasoning ability (eg we can assign truth values to assertions, and usefully combine those assignations), though seemingly discontinuous with the map-like cognition of non-linguistic animals, actually conceals our continuity with the cognitive functioning of non-linguistic animals.
The simplest reason (Occam's Razor) for the brain's global lateralisation is that it reflects existing local asymmetries, ie the brain has a fractal architecture. This idea is depicted generally in figure 4.1(a), and in more detail, in figure 4.1(b) below-
(a) (b)
Figure 4.1 - note this figure is incomplete
In GOLEM theory, the four functional roles within each cerebrum are anatomically grounded in the four main anatomic lobes [1]. These same four functions (P,T,L and F) are abstracted globally by being realised within the left and right halves of the cerebrum - see figure 4.1(a).
There is empirical evidence to support this fractal view. Tulving's memory model [13] is shown in broad terms in figure 4.2(a) below, while figures 4.2(b) and 4.2(c) depicts a cognolinguistic interpretation of GOLEM. Originally, Tulving followed the orthodox approach by separating declarative from procedural memory. He subsequently partitioned declarative memory into autobiographical (he called it 'episodic') memory in the left cerebrum from general knowledge (he called it 'semantic') memory in the right cerebrum. Specifically, he identified episodic memory as an instantiation (ie a special case, a subtype) of semantic memory.
As shown in figure 4.1, GOLEM theory suggests that there may be a straightforwardly computational basis for this distinction - that episodic memory is formed from serial semantic categories, ie those selected from one time line, (the pseudopast recordings sampled from the individual's LTM) while semantic memory consists of a compendium of semantic categories, ie selected by the simultaneous (parallel or concurrent) application/availability of a variety of coding constraints [12], as if they came from multiple time lines eg from several individuals [14]. Tulving believed that his discovery supported the viewpoint that memory was essentially multimodal, and contradicted the conventional view that memory was monolithic. GOLEM theory tends to swing the pendulum back toward the monolithic view, not by contradicting Tulving's findings, but by placing a computational (ie parallel processing) interpretation upon them. There are points of agreement between the 2nd and 3rd levels of the GOLEM model and Tulving's memory scheme as follows-
(i) vertical separation into declarative (level-3) and procedural (level-2) governance paradigms
(ii) horizontal specialisation into left-side episodic (ie autobiographical) and right-side general knowledge (ie ‘common’ facts) semantic subtypes.
(a) (b) (c)
Figure 4.2
More than matching any particular data, GOLEM predicts and explains the overall pattern of observations. Figures 4.2(b) and 4.2(c) above are derived from a linguistic view of cognition. In this conceptual framework, cognition itself is considered to be a linguistic process. That is, the concepts we use to understand the role and behaviour of language outside the brain, such as semantics, syntax and symbols, apply equally well inside it, ie they can be used to explain mental processes[2].
the subjective stance (TSS)
It is possible to explain the consciousness function of each animal (level 2) GOLEM by using the GOES programming model, as was demonstrated in the previous section. However, a greater level of theoretical sophistication is needed to extend the model of consciousness to the adjoint human (level 3) GOLEM. This level of sophistication is provided by a modelling framework called 'the subjective stance' (TSS). TSS builds on our intuitive understanding of the subject's point of view as represented by the phrase 'inside looking out'. At the lower end of the memory, the semantic hierarchies modelled by the neural circuits represent objective (physical, existential, controlled/perceived items) data. However, at the upper end of the memory, the semantic hierarchies represent subjective (psychophysical, experiential, controlling/perceiving selves).
A necessary insight (and one that is surprisingly hard to put into words) is that each subject (a self) has all the information needed to model other subjects, by adopting the default 'subjective stance'- this is the default assumption (made by any subject S[0]) that every other subject (say S[i]) sees all that it (ie S[0]) sees, and feels what it (ie S[0]) feels. Only when it (S[0]) has direct evidence of different informational access, can a self change that default view of the similarity of other subjects. Each self S[0] must therefore assume that every other subject S[i] is internally identical to itself, except where proven otherwise.
TSS has a very useful corollary- If a self (S[0]) sees another self (S[i]) in a physical location that it (ie S[0]) has previously visited, it can safely model that other self by retrieving its own memory of that location. Because of TSS, S[0] can safely ignore the other contextual factors that might be different. Another way of stating TSS is that a self (S[i]) can only assume that another self (S[j]) has different informational content if it (S[i]) has direct evidence of that informational difference, such as if it sees that the other self S[j] is in a different spatial position [4].
This makes the task of one subject modelling a dialog with another subjective agent orders of magnitudes easier than if the agents were conventional (ie objective) robots, since the first subject's adoption of TSS means it can model the second subject as a more-or-less identical copy of itself. The upshot is as follows- the prime difference between the grammatical 'subject' and 'object' is not one of informational access (which we have shown to be identical, unless otherwise proven), but of agency, ie of conative or volitional role. The first subject must model itself as the controller, the one who leads the interaction/dialog, and must model the other self as the controlled, the one who follows it. There is a old roboticist's epithet [5] which applies in this situation- one of the best ways of predicting the future is to cause it to happen.
The purpose of this concept becomes clear when we construct a single level-3 human GOLEM from two level-2 animal GOLEMs. From TSS, the two halves of the same brain have identical affect, ie access to information, so any lateralisation must come from their different effect, ie status as agents of change. From this argument, it is clear that the basis of functional lateralisation is that the left side of our brains is the active ‘master’, and the right side is the passive ‘slave’.
The TDE
While the animal GOLEM model is best suited to the explanation of consciousness and emotionality, it has trouble explaining language and conceptual aspects of cognition. The human level GOLEM has an alternative form called the TDE, which was specifically designed to reveal the complexities of language, especially its functional lateralisation in left (production & reception) and right (prosody & pragmatics) hemispheres - see figure 4.3 below.
The TDE is a trilevel human GOLEM, made from two bilevel animal GOLEMS. This is an example of a fractal architecture, with the two cerebral hemispheres at level-2 matched by the two halves of the cerebellum at level-1. A 'by inspection' prediction of the model is that the language paths which are disrupted by lesions at Broca's and Wernicke's areas in the left cerebrum should also be disrupted by lesions which occur at similar areas in the right cerebellum. The fact that this unexpected result (ie unexpected using more conventional models of functional mapping) is empirically confirmed [3] represents one of GOLEM’s major modelling successes.
Figure 4.3
At first blush, the TDE overview depicted in figure 4.3 seems complex, but this is a false impression. Actually, the TDE overview represents belated confirmation that conventional notions of compiled program computation do indeed apply to our brains, minds and selves. It is a triumph of (relative) simplicity. From this diagram, it is possible to clearly identify a sequence of linear processes directly relating declarative goal states (semantic = compound level-3 semantic) to procedural script (syntactic = level-2 semantic transitions/differentials) sequences built from symbols (symbolic = simple level-1 semantic). This sequence of linear processes is a direct analogue of the familiar source-compilation-execution programming cycle used in all modern digital computer designs.
Now refer to the top level TDE in figure 4.3 above. The efferent red line is the production of syntax via Broca's area, while the afferent yellow line is the comprehension of syntax via Wernicke's area. But isn't the comprehension of syntax just semantics? Look at the figure 4.4 below. If you read carefully between the lines, you will see that the arguments all rely on the inaccurate 2-stage (saussurean) idea of coding, not the correct 3-stage (Piercean) model.
The ascending path through Wernicke's Area in the left Cerebral Hemisphere carries semantic information, but not about absolute semantic state (which is held in the global T-lobe, the Right Cerebral Hemisphere), but about relative semantic changes or semantic differentials . When we use speech, our goal is not to communicate semantics, but to communicate those semantic changes we believe are needed by the system or person who is receiving them. Without considering the mind-state of the person we are talking to, there is no meaning. "I saw Joe wearing a gold ring today" only has the intended meaning if the listener believes that Joe is single. If you don't know Joe, your response might be "so what! its not illegal, is it?". This seems to be the place where Chomsky seems to have erred.
GOLEMMA
Figure 4.4 above depicts Carl Sachs conception of Churchland's and Millikan's (respective) attempts to make sense of the issue of the truly confusionary(sic) nature of this chim-chim-chimaera we so glibly call 'semantics'. It would be funny if it wasn't so serious. Hey, lest you are prematurely tempted to emit a smug chuckle, be warned- this stuff beat Chomsky. That means it is HYPERHARD. Stuff that hard, you empty your life for at least ten years and do nothing else but walk around muttering to yourself like a rabid dog until you solve it. To adopt a 'sensible' mixed-mode solution strategy is futile. I watched the debate between Noam and Michel and came away with a dreadful fear in my heart. I'm not easily freaked, so thats how high the bar is! BTW, I ask people (scientists who should by all rights know) what Foucault's gig was. They must have read about his sado-masochistic sex life, which I admit is quite disgusting to straight people. So they assumed he is into some psycho-humanistic Sorbonne [8] garbage, like Pol Pot [9], Uncle Ho, or Steven Pinker, maybe (sorry, Steve, I couldn't resist). Mais non, he's not. Get this- Foucoult's gig is the history/historiography of science, you know, like Karl Popper and Thomas Kuhn. Crazy dude, biensur, but solid, sane even conservative work. Go figure. But then who would have picked iconoclastic linguist Richard Montague as...oh, you don't know? google it, but first make sure your kids are in bed. The lesson: Nasty sh*t occasionally happens to *nice* people. Unfortunately, clever people are, sadly, often among the nastiest. The benefit to humanity (not in the abstract, its you and me that benefit- the miracle medicines that save us, the magic machines that make everything easier, groovier and less boring) from their cleverness must be balanced against the evil from their nasty side. Leonardo did sick shit to cute little kittens. I'm a cat person, so I think he is not a nice man. He was a sick asshole, if I'm honest. His job was a weapons designer, for the (Medici) cartel. If he was alive today, he would be designing semi-submersibles (airship/dirigibles? [10]) to smuggle cocaine for the Sinaloa or Medellin family businesses. And Descartes...! [11]. Lets not even go there...
1. for the sake of this argument, the occipital lobe is treated as a part of the parietal lobe
2. The tri-level view of linguistic function can be attributed to Chomsky, while its teleological application to neurocognitive processes can be attributed to Marr. Chomsky (1986) characterizes these two approaches in terms of the technical concepts E-language and I-language, where E stands for 'externalized', and I stands for 'internalized'. Under the internalist perspective, an I-language consists of a computational system, including generative procedures and constraints on their application and output, plus a lexicon-in other words, a grammar. In Chomskyan theory, the computational system plus a lexicon defines the language under analysis. Thus the terms 'grammar' and 'language' (as in 'a language') identify the same mental object. Under this perspective, the study of language becomes a core topic of human psychology and biology, raising questions about acquisition, genetic endowment, and evolution.
3. Silveri MC, Di Betta AM, Filippini V, Leggio MG, Molinari M. (1998) Verbal short-term store-rehearsal system and the cerebellum.Evidence from a patient with a right cerebellar lesion. Brain: 121:2175-87.
4. TSS has unexplored implications w.r.t. the frame problem
5. I guess that means me.
6. Sachs, C.B. (2018) In defence of picturing; Sellars's philosophy of mind and cognitive neuroscience. Phenomenology and the Cognitive Sciences. https://doi.org/10.1007/s11097-018-9598-3
7. 'a weaponised info-virus disguised as a revolutionary new mobile telephony ADI-OS' is how a GOLEM critic might describe it. Key:: ADI = artificial digital intelligence. OS = operating system. GOLEM = ADIOS. Bye bye!
8. A joke from the time of Nelson and Waterloo. Q: what is the difference between toast and frenchmen? A. you can make soldiers from toast.
9. my sis calls me 'Brother no 1'. look it up, lazy! I really like killer hoes! go on, you will probably need to google that too.
10. I'm not a drug smuggler, but if I were, I'd be building airships. Millenial Medellin Zeppelins carrying cannabis, coke, meth anonymously (use sat phones to navigate) from Central America to the drug-fueled ad agencies and board rooms of mainstream USA. Just paint 'Goodgear' on the side. No one will be the wiser. Whoops, I meant 'Goodyear'.
11. Descartes makes almost anyone from the last two centuries look limpwristed and lame as a lambswool comforter, safe as a milk moustache. Wow! Mercenary, royal gigolo, reputedly poisoned to death by his lover, queen of holland. What? Google it. What was his catchphrase...OCCIDERO (NECO?) ERGO SUM...yeah, something like that. I kill therefore I am. When he wasn't swashbuckling, and being a hoodlum, he'd do some philosophy, maybe.
12. Tulving, E., & Thomson, D. M. (1973). Encoding specificity and retrieval processes in episodic memory. Psychological Review, 80(5), 352-373
13. Tulving, E. (1984) How Many Memory Systems Are There? APA Award Addresses University of Toronto, Canada
"A ternary classificatory scheme of memory is proposed in which procedural, semantic, and episodic memory constitute a 'monohierarchical' arrangement: Episodic memory is a specialized subsystem of semantic memory, and semantic memory is a specialized subsystem of procedural memory".
14. This is a feature made possible by The Subjective Stance, or TSS. TSS is a unique, innovative, and indeed, essential aspect of GT's philosophical framework.