Somehow, watching talking heads of Ramsay and Rockwell made me a little sad.
The language is generative. The generation of language is governed by the grammar. Computer languages differ from human languages mainly in the types of grammars they use. Computer languages are defined by formal grammars which are unambiguous. (see Chomsky hierarchy) “Code” and “language” are interchangeable, where both serve the same function: encode-transfer-decode. What goes before and after is where we have the “processing unit” or “the agent”. The difference between machine and human is in the structure of the code and the interpretative machinery, yet both are inseparable and represent a unified metaphor of the informational system.
Whether computer program is a form of discourse or not… This is where the dialog was one sided. When we talk about computers, we (pretend to) know precisely what we are talking about. Despite enormous complexity and stochastic nature of the external processes, at any instance we can halt the machine and examine it’s state bit by bit and determine precisely what and why happens. Hence, we can easily (or at least have comfort in possibility) understand what computer program is and how it is executed by the machine. This part was clearly addressed in the dialog. Program is a set of unambiguous commands interpreted and executed by the computer. Program is so formal that it can proven to execute in the most strict mathematical definition of the proof.
Machine language is unambiguous in such a way that we can see how each command translates into the action of the automaton, be it shifting a bit in the memory register of an mp3 player or releasing a missile from the airplane drone.
Farther, there is no difference between the machine and the program, hard-coded solid-state hardware and software instantiated in the configuration of electromagnetic fields are one and the same. They both can be sufficiently described by the same formal language. Computers can emulate other computers by executing programs which are software metaphors of some other hardware or even themselves.
Now, how about the human language… Generation of human language is also governed by the grammar, but grammar not as strict as machine languages. Humans use ambiguous code for communication. The discourse does not transform itself into easily observed action as well. Ramsay and Rockwell noted this correctly. And yet, this is not where difference is to be sought.
The part we are missing is the interpreter and the execution unit, the metaphoric analog of the CPU, the wetware, the mind, the consciousness, the self-aware agent.We do not know a squat about our own minds. Rockwell admits to that. We do not know how languages are interpreted into the meaning. Hey! We don’t have the most important part of the metaphor of human languages, we do not know how we interpret them, how words and sentences transform into meaning, we have no clue! Yes, we know a great deal about the grammars, how to classify and design them. We can recover dead languages (like Phoenician) and invent new ones (like Esperanto).
However, we have no clue what does it mean to “understand”. We do not understand how to understand the understanding.
The hardware is inseparable from the software, they are essentially one in the Turing machine (universal model of the computer). It is probably so for human language.
And thus, how the incomplete metaphor of human language can be compared to the complete metaphor of the software/hardware monad of the Turing Machine? Both serve the same encode-transfer-decode cycles, but what performs this cycles is the real question. The machine language describes both hardware and software. Yet, we do not know how to capture one’s consciousness in the language, something is amiss in our self-expression. Why than we venture in the futile speculations over the nature of things we clearly do not understand and pollute clear and well-defined metaphor of formal languages with our incomplete metaphor of the discourse? This is why I am sad.
If we had the capacity of language compared to that of the Turing machine, we would be able to project our complete self through the language and thus replicate ourselves into the language. The text would be indistinguishable from the self, and self could exist in infinite copies. We would be able to make copies of ourselves n our own thoughts, so our thoughts would be as self-aware as we are. Neither of it ever happens though (Plato points to it too), there is something in our minds we utterly cannot put into the language, there is something in us that stays fixed in the stream of discourse, struggling to make itself understood and striving to understand others. Somehow, our languages are not as suited for us as machines and their codes. As if we cannot establish the identity between the perceived realty and the self, as if they are irreconcilable, disjoint, alien. Language fails us to reflect its agent completely. Language certainly reflects perceived reality, but language reflects only partially, if reflects at all, the self that utters it. So, if language reflects the reality, language must inherit the properties of reality by isomorphism, so what we attribute to the reality must be attributed to the language. So, we can say for once that language is objective and real. Now, if self is not reflected by language, is self not real? Or does self belong to the part of the reality which is not reflected in the sensual reality and thus cannot be captured by the language? If only Ramsay and Rockwell could figure this out in their dialog, it would be much more interesting.