Philosophy Discussion Forums

John Searle says you can't get semantics from syntax and that's why the Chinese Room can't be conscious. I recently posted an attempt to begin a detailed explanation of how you could get semantics from syntax, but I expect/hope the lack of response was due to the complexity of the description.

It occurs to me that there is a shorthand argument that can be made without the detailed explanation. Said argument goes like this:

If consciousness is the result only of physical processes in the brain, as Searle postulates, and "meaning" is generated by the human brain, then "meaning" is generated by the operations of the rules of physics, and the "rules of physics" is just a particular syntax.

Any flaws in this argument?

*

JamesOfSeattle wrote:John Searle says you can't get semantics from syntax and that's why the Chinese Room can't be conscious. I recently posted an attempt to begin a detailed explanation of how you could get semantics from syntax, but I expect/hope the lack of response was due to the complexity of the description. It occurs to me that there is a shorthand argument that can be made without the detailed explanation. Said argument goes like this: If consciousness is the result only of physical processes in the brain, as Searle postulates, and "meaning" is generated by the human brain, then "meaning" is generated by the operations of the rules of physics, and the "rules of physics" is just a particular syntax.
Any flaws in this argument?*

Is meaning generated by the brain or the mind? Chomsky writes:

People think, not their brains, which do not, though their brains provide the mechanisms of thought. I may do long division by a procedure I learned in school, but my brain doesn't do long division even if it carries out the procedure. Similarly, I myself am not doing long division if I mechanically carry out instructions that are interpreted as the very algorithm I use, responding to inputs in some code in a Searle-style" arithmetic room". Nothing follows about my brain's executing an algorithm; likewise in the case of translation and understanding. People in certain situations understand a language; my brain no more understands English than my feet take a walk. It is a great leap from common sense intentional attributions to people, to such attributions to parts of people or to other objects. That move has been made far too easily, leading to extensive and it seems pointless debate over such alleged questions as whether machines can think: for example, as to "how one might empirically defend the claim that that a given (strange) object plays chess"...or determine whether some artifact or algorithm can translate Chinese, or reach for an object, or commit murder or believe that it will rain.

Similarily, the neuroscientists M.R. Bennett and P.M.S. Hacker in their criticism of Searle write:

Our primary concern was with the neuroscientific cousin of this, namely the error of ascribing to the brain – a part of an animal – attributes that can be ascribed literally only to the animal as a whole...it was pointed out by Anthony Kenny in his brilliant paper ‘The Homunculus Fallacy’ of 1971... we now observe that to say that the brain is angry is as if one were to say that the brain weaves or builds. For it is surely better to say not that the brain pities, learns or thinks, but that a man does these. Accordingly, we deny that it makes sense to say that the brain is conscious, feels sensations, perceives, thinks, knows or wants anything – for these are attributes of animals, not of their brains...It is animals that are conscious or unconscious, and that may become conscious of something that catches their attention. It is the student, not his brain, who awakes and becomes conscious of what the lecturer is talking about, and it is the lecturer, not his brain, who is conscious of his students’ boredom as they surreptitiously yawn. The brain is not an organ of consciousness. One sees with one’s eyes and hears with one’s ears, but one is not conscious with one’s brain any more than one walks with one’s brain.

Bohm2, thanks for the references. Two points:

1. Saying the person is conscious, as opposed to the brain being conscious, does not change the conclusion of the argument, which is, "meaning" comes from physics.

2. The sources you quoted are either mistaken or are using a different conception of "consciousness". Searle and I conceive of consciousness as a process, specifically, a process that occurs in the brain. Saying "[p]eople think, not their brains" is equivalent to saying people digest food, not their stomachs. Regardless of how you wish to attribute the quality of consciousness, all of the pertinent processes occur in the brain. Thus, the quote from the second set of authors:

The brain is not an organ of consciousness. One sees with one’s eyes and hears with one’s ears, but one is not conscious with one’s brain any more than one walks with one’s brain.

is simply incorrect. One is conscious with his brain exactly as one sees with his eyes.

But again, the point is that the processes associated with "meaning" occur in the brain. The processes are physical, and therefor syntactic.

*

JamesOfSeattle wrote:Saying the person is conscious, as opposed to the brain being conscious, does not change the conclusion of the argument, which is, "meaning" comes from physics.

I think meaning comes from us; that is, from our biologically-given cognitive structures. So, if I had been differently constituted, with a different cognitive structure...I would come to know and follow different rules (or none) on the basis of the same experience, or I might have constructed different experience from the same physical events in my environment. I suppose one could argue that everything comes from physics (at least, from an ideal physics) but that's not very informative because our physics/science is progressive, tentative and on-going (work in progress). And physics involves postulated theoretical entities that come from us. Physics isn't reality itself. It's our model of reality.

I don't think we have any clue how mental stuff like consciousness or meaning, etc. relates to neural stuff but this may be due our incomplete understanding of the more fundamental stuff (i.e. atoms, sub-atomic, quantum, etc.) that give rise to neural stuff. In a sense, I agree with you that meaning and I suppose everything else would come from an ideal physics/science. But, since we are limited organisms with epistemic boundaries, it's very likely that there are stuff we can never understand/explain. It may be (as Russell argued) that our sciences/physics is limited to stuff that allow for computational/mathematical/structural analysis leaving other properties of mind, like our qualitative aspects, untouched even though their existence is more certain than anything that physics postulates. Moreover, it would be a miracle if nature produced in man's minds the ability to develop models that allow for a one-to-one matching between our concepts/theories and the world. It may be that,

...what we call “natural science” is a kind of chance convergence between aspects of the world and properties of the human mind/brain, which has allowed some rays of light to penetrate the general obscurity, excluding, it seems, central domains of the “mental.”

JamesOfSeattle wrote:But again, the point is that the processes associated with "meaning" occur in the brain. The processes are physical, and therefor syntactic.

I'm not sure what you mean by saying that the processes are physical. Do you mean processes described by current physics or a future, ideal physics? Or do you mean processes that are non-mental? Moreover, until or unless physics completes its job of providing a full and final characterization of the ultimate constituents of the universe, we do not possess a final/definitive concept of the "physical". I'm not disputing that the brain is involved. But as I see it, theories of mind and language appear much more established in terms of explanation than are neuroscientific ("physical") theories (i.e. cells, neurons, neural networks, etc.).

In the present case, the theories of language and mind that seem best established on naturalistic grounds attribute to the mind/brain computational properties of a kind that are well-understood, though not enough is known to explain how a structure constructed of cells can have such properties. That poses a unification problem, but of a familiar kind. We do not know how eventual unification might proceed in this case, or if we have hit upon the right categories to seek to unify, or even if the question falls within our cognitive reach. We have no warrant simply to assume that mental properties are to be reduced to "neural network properties", to take a typical claim.

Two papers discussing this perspective:

The mysteries of nature: how deeply hidden?
http://isites.harvard.edu/fs/docs/icb.t ... 213110.pdf

Language and nature
http://www.ucd.ie/artspgs/meaningthree/ ... natobj.pdf

JamesOfSeattle wrote: If consciousness is the result only of physical processes in the brain, as Searle postulates, and "meaning" is generated by the human brain, then "meaning" is generated by the operations of the rules of physics, and the "rules of physics" is just a particular syntax.

Any flaws in this argument?

Consider the fact that all carbon atoms have the same physical characteristics.

Consider that all flesh has very, very similar physical characteristics.

Consider that brain material, in its physical properties, are not too significantly different from each other, when examining brains in mammalian species.

Consider that a male mammal will find a female mammal sexually appealing in his own species only. (As an overwhelmingly accurate but not totally true general rule.)

Yet all male mammals share the PHYSICAL properties of their flesh, bone, brain materials.

IF your assertion was true, that biological behaviour (such as generation of meaning) is a function of physics, and resulting from the rules of physics, then we'd all speak the same language, and all animals would understand it and speak it too.

But!! Or however!! we don't speak the same language, and we don't find all females/males of the opposite sex in all mammalian species attractive sexually.

Therefore meaning is not purely a function of the rules of physics.

-1- wrote:IF your assertion was true, that biological behaviour (such as generation of meaning) is a function of physics, and resulting from the rules of physics, then we'd all speak the same language, and all animals would understand it and speak it too..

This is simply incorrect. Just because all animals are made of physical stuff does not imply that all animals have the same physical organization. But it is true that everything they do, and everything that happens inside them, must follow the rules of physics.

*

JamesOfSeattle wrote:

-1- wrote:IF your assertion was true, that biological behaviour (such as generation of meaning) is a function of physics, and resulting from the rules of physics, then we'd all speak the same language, and all animals would understand it and speak it too..

This is simply incorrect. Just because all animals are made of physical stuff does not imply that all animals have the same physical organization. But it is true that everything they do, and everything that happens inside them, must follow the rules of physics.

*

You completely ignored what I had written, and focussed on one sentence, without apparently even reading the proof.

This is not good.

-- Updated 2017 March 8th, 5:55 pm to add the following --

Plus you committed the equivocation fallacy. "physical organization" and "rules of physics" are two different concepts.

Rules of physics apply to all physical objects equally, without exceptions and without degrees of influence.

Yikes. Stay away from me, and I stay away from you. What's your alias? James of Seattle. I'll try to remember that. If I ever respond to your posts again, please gently remind me if you would please, that I swore off of doing that. Thanks.

I would argue that there is a sense in which there can be meaning without consciousness. Translating from one language to another at the level of Google’s Neural Machine Translation system is more than just following a set of rules for correlating sets of symbols.

It turns out that if you teach the system to translate English to Korean and vice versa, and also English to Japanese and vice versa, the system is able to translate Korean to Japanese, without resorting to English as a bridge between them. This raises the following question:

If the computer is able to make connections between concepts and words that have not been formally linked… does that mean that the computer has formed a concept of shared meaning for those words, meaning at a deeper level than simply that one word or phrase is the equivalent of another?
In other words, has the computer developed its own internal language to represent the concepts it uses to translate between other languages? Based on how various sentences are related to one another in the memory space of the neural network, Google’s language and AI boffins think that it has.
techcrunch.com/2016/11/22/googles-ai-tr ... -language/

I think this is the crux: the brain is a causal mechanism that is specific to a particular electrochemistry. Searle would emphasize that the physical substrate matters; we just don't know how the brain/organic material instantiates it.

On the other hand, a program is a formal, not causal mechanism. Anything that carries out the syntactic steps of the program, if the Chinese room argument is wrong, would instantiate consciousness. I think that's a reduction ad absurdum, but I guess some functionalists bite the bullet and say the substrate (whatever carries out the program) doesn't matter. We might not know enough about how "the laws of physics" give brains/other organizations of organic material their causal powers, but Searle would say appealing to syntax is a huge mistake.

I guess I should have pointed out that the purpose of the OP was to continue the discussion of why computers can or cannot be conscious. Thus:

Fooloso4, I absolutely agree there can be meaning without consciousness, but not in the way you intend with your example. In my understanding, the translating computer you mention has consciousness. It just doesn't have the competencies you usually associate with human consciousness. For me, meaning is an affordance. If something makes use of that affordance, that thing has some level of consciousness.

Roquentin, the brain is a causal mechanism just as a computer running a program is a causal mechanism. A program by itself is not a mechanism. If the syntactic steps of the program duplicate the syntactic steps of the physical events in the brain, the results are equivalent. To the extent that the results are informational, the results are identical. I know that Searle likes to say that rain simulated by a computer is not wet, but a computer can simulate a calculator performing 2+2, or a person calculating 2+2, and in each case the result is identical: 4. Yes, I'm saying consciousness is an information process.

*

"If the syntactic steps of the program duplicate the syntactic steps of the physical events in the brain, the results are equivalent." I'm probably not understanding you correctly, but that sounds like a tautology. Of course if a given substrate duplicates the physical causal powers of the brain, then it's conscious. Searle himself has said that.

What he said, about syntax not being sufficient for semantics, is something you just acknowledged. That a program is not a mechanism. But then you went on to say a computer adding 2+2 and a human adding 2+2 are equivalent, which is inconsistent with the earlier requirement that computers or any other substrate's having to duplicate the brain's physical causal powers. The two are equivalent at the "program" level only, although even that's untrue since computers use brute force algorithms to prove theorems, whereas mathematicians make jumps using what can only be described as "intuition."

JamesOfSeattle wrote:. Yes, I'm saying consciousness is an information process.

An argument against this view:

Computation and Consciouness
http://web.csulb.edu/~cwallis/labs/stan ... &consc.pdf

An eummary of the paper (short-version):

Is Consciousness the Result of Information Processing?
http://www.socphilinfo.org/node/190

Bohm2, thanks for the ref.'s, as always.

Of course, I think the argument presented has flaws. The main flaw is that it misrepresents the nature of a conscious experience, and this misrepresentation I think is common to those who think there is a "hard problem", or those who define something as conscious by saying there is something it feels like to be that thing.

With respect to computation, the misrepresentation becomes evident in statements like these:

More generally, whenever the “toothache algorithm” is being run, whether by you or something else, there is a conscious experience of having a toothache going on

Every time Klara runs the “toothache algorithm”, Klara will experience a toothache;

[Klara is the name of the computer which is performing an algorithm which is syntactically equivalent to what neurons are doing.]

The problem is that this view seems to suggest that something is experiencing the toothache while the algorithm is running, and that Klara is the thing doing the experiencing. Both of these are wrong, at least by my understanding.

So to restate my understanding: the entire calculation is the experience. There is no experience to speak of until the entire calculation is finished. After the calculation is finished, the output of that calculation may or may not be available for subsequent calculations by the same or a different agent. Klara could be said to "have" the experience if and only if Klara can respond to that output of the first experience, thus generating a new experience.. Klara's "consciousness" is defined by the repertoire of "experience-type" calculations she can perform.

When I say I have a toothache, what's happening is the toothache algorithm is being calculated repeatedly over a period of time. If it happened exactly once, I would probably never know, or more likely simply not remember.

*

Men genetics are compel men stick there ding-dong in a hole somewhere.

Women need a reason. men just need a place.

Male externalize *Y*---testes outside

Female internalized \**/ ovaries inside

Consciousness is shared by both men and women, however, the genetics are different.

The greatest difference in world--as far as humanity goes --- is between man and woman yet they have so much in common.

r6

JamesOfSeattle wrote:

If consciousness is the result only of physical processes in the brain, as Searle postulates, and "meaning" is generated by the human brain, then "meaning" is generated by the operations of the rules of physics, and the "rules of physics" is just a particular syntax.

Any flaws in this argument?

There are definitely flaws in this line of thinking that strongly undermine the ultimate argument. I outline the critical ideas that could attenuate the argument:

- If our focus is on “meaning”, first we should decide what meaning itself is. Meaning is a fundamentally “human” creation; there is no meaning in nature. “Rules of physics”, then, should be replaced by “principles of interpretation.” This would be a more plausible idea.

- Comparing rules of physics to syntactical formulae sounds like a very remote, and apparently impractical, metaphor. Syntax is a collection of socially and historically shaped conventions that can formally structure human language. These rules are subject to change and they have no definite epistemic foundation. As a result, different cultures produce different conventions that shape the format of their languages. On the contrary, the way nature works has never been mediated by human agency. You cannot tell the Sun how to work every day but you can modify, question and renew syntactical structures (as we see in language policy-making).

- Is there any proof that meaning is constrained by syntax and semantics? Today, many scholars have shifted their attention to pragmatics. As Bakhtin would argue, the meaning of a word is an aggregate of all of its actual uses. So, there is no essential meaning to a word. If meaning is continually produced and re-produced, how could we associate it with a system that is regulated by predetermined rules?

- Meaning is a creation of symbolic networks of relations. There are no symbols in natural processes. Everything in our perceptual world is symbolic: the legal system, the scientific system, the society. What is the role of symbol in the rules of physics (nature)?