AKG wrote:
Mariner, you seem to be taking this too personally.

You have asserted that algorithmic machines can't "derive" first premises. I have defined "algorithm" if you read my posts. It is a step-by-step process which solves a problem. I have written at length that deduction is a step-by-step process that does not solve the problem of establishing first priniciples, but still see no reason why a different algorithm can't.

You've asserted post after post that algorithms, by definition, can't solve that problem. I've continually asked why, and you've not responded appropriately. You've simply restated your position, claimed that we are contradicting logic, that we aren't analyzing our positions, we're focused on the wrong thing, we're ignoring your (non-existent) line of reasoning, etc.

Your last post addressed to Socrastein and myself has many words but is entirely devoid of relevant content.

This is why I think you're taking it too personally, as your posts focus on who debates like what, and you ignore the fact that up until now, you have given no reason why an algorithm cannot derive first premises.

I defined "algorithm" in the post right after you asked that I do it, and every post that followed, but you've just been placing the focus elsewhere, and worrying too much about the accusations that you're debating like a politician or asserting without arguing, rather than giving us reason to believe you're doing otherwise.

In your defense, I disagree with Socrastein to some degree. (...) The process of trying to tear another's arguments down is not "overly politician-like", it is essential and desirable.

Machiveli wrote:
Perhaps its us not understanding rather than you not saying - but you realy need to make the above something other than circular.

I was thinking about the phrase "derive first premises"
It seems to me that in general we..

Infer premises from observation.
Assume first premises
Are hard wired by our genes to assume premices

I wonder if you can give any other ways.

Finaly I have an intuition about your belief that I wonder if you would confirm.

(...)

Seems odd in the context of your belief how can you consider a room containing an agent which might as well be a robot to be capable of producing meaningful chinese by following instructions. Such a room is an algorithm and you have denied the ability of algorithms to make meaningful statements.

secondly your use of the word "purely"

thirdly You seem to be concentrating on "initial premices" i.e. those that we have had to assume to live and discuss meaningfuly i.e. logic, causality ect..

My intuition is .. I think you would have not too much of a problem with the idea that an algorithmic process with these premices encoded designed by us or by "the great programmer" can reach truth.

Mariner wrote:
Let's see the possible counter-theories:

a) yet unknown (non-algorithmic) natural process
b) miracle
c) alien intervention (which only takes the problem to a further step)

AKG wrote:
The only problem there is that if we or "the great programmer" are not deterministic beings, then you've only shown that a deterministic being in a non-deterministic universe can produce truth.

Truths themselves don't exist in our universe, and they are independent of whether or not there exists any being who can utter them. I differentiate between truth and true utterances, true beliefs, etc. Truth is essentially formless, truth is simply that which is, and true utterances are those truths expressed in the form of spoken language.

AKG wrote:
Mariner

You last post was more of the same thing. Accusing me of contradicting logic, and still providing no argument. You have still given no reason why an algorithm can't produce first premises.

Man's brain, it is reasonable to assume, is also hard-wired to associate certain grunts with certain classes of things. These things that man's brain is hard-wired to do follow some sort of deterministic process, there's no reason why they couldn't. Is it so hard to imagine that our brains evolved to have language, evolved in a deterministic fashion to differentiate objects, classify them, and associate grunts to them?

So far, your argument has been:

Deterministic beings could not produce first premises because they behave algorithmically.
Why?
Well, because algorithms are incapable of deriving first premises.

That's pretty much all you've given.

As far as I can tell, this post is the first time when you've said something, that being that algorithms don't work because they rely on first principles. Up until now, you've said only that they don't work because deduction doesn't work, or they don't work by definition, or they don't work and anyone who disagrees contradicts logic, or isn't looking at their own position, etc, etc.

...there's no reason to believe that the way in which we do those things is not algorithmic. They are non-deductive, and in fact, they have little to do with deduction.

Again, there are no assumed premises involved in defining things, or observing things and classifying them. We have direct access to the phenomenal world, and we are hard-wired to classify the information coming from phenomena. We see things and name them.

To my point about the nature of truths. We say that a proposition is true if it corresponds to fact. I simply said that truths can be more than just true propositions. A true proposition is a fact which has been put into a particular form (a sentence). There is transformation made from fact to sentence. That transformation is a correspondence between the sentence and the fact. The fact itself is a transformation of the fact (an "identity transformation").

Clearly, this transformation of the fact corresponds to the fact, and so it is also a truth. Truth can take different forms. A belief can be true, even if it is had by a person with no language, beliefs in different languages can be true, utterances in different languages can be true, etc. Beliefs and utterances are completely different kinds of "stuff" but both can be true. Fact is a different kind of "stuff" but it can also be true, so long as it corresponds to fact, which it does most naturally.

Finally, to address another point of yours: you claim that a deterministic being has no reason to trust what is pre-wired into it. First of all, how does a non-deterministic being come up with first premises, and why are they any more trustworthy?

Secondly, noting that we aren't pre-wired with truths, but the tools to create truths (like observation and the tendency to assign grunts to classes), your point doesn't hold.

Thirdly, if it is your concern that our experiences don't correspond to reality, I would argue that for us, reality is the phenomenal world so, in a sense, experience is reality. It may not correspond to the noumenal world, but we don't make truths about the noumenal world, we make truths about the world we live in and experience every day, the phenomenal one.

AKG wrote:
But the argument is still not worth debating.

You continue, again and again, to say little more than, "You guys have to think about it," and "my position is still correct."

You're still not arguing. You still repeat the same flawed assertions about algorithms again and again and again. Much of your argument seems to be based on philosophies of logic/metaphysics that requires debunking on its own.

It would be too much of a project to point out the flaws in your argument because they are so fundamental that each point might require a separate thread of its own. I will try to give one-sentence responses to the points you've made, and surely you won't budge at all because you spend too much time telling us we're just not understanding instead of actually making a decent argument.

You refute the "natural selection" argument by saying its an algorithm, and assuming that this is a sufficient argument. I already explained how your understanding of deduction, algorithms, and premises was flawed, but you seemed to ignore it.

You say algorithms must "follow logic." Logic allows true premises to be manipulated to reach true conclusions. Deduction is an algorithmic process for solving the problem of concluding truths. If we want to tie our shoes, we don't need to follow logic in the same sense as when we want to deduce.

We're not manipulating propositions when we tie our shoes, we're manipulating shoelaces. Sure, there must be some logic involved in the instructions, but it's totally different. You say that natural selection is an algorithm, but what the heck does it mean for natural selection to follow logic?

Your assertion that we name things, and then they become things is patently absurd.

We name things before we see them?

How the heck do we know what it is we're naming if we don't even see it? Logic doesn't come from perception, logic is something like the structure by which we perceive things. It is like an inherent structure in our perception.

Logic is how we perceive things, roughly speaking. There must be a way by which we perceive things.

We evolved, we perceive things, and the way in which we evolved led us to perceive things in a certain way, with a certain structure. Unless you want to argue that deterministic beings are incapable of perception, but this is nonsense.

Also, there's no reason why hard-wiring can't be non-material, nor would that give any problems to the thesis that deterministic universes can have truth. Also, we say that the sensory input to the brain creates images/perceptions which the mind (non-material) sees. This is a causal, deterministic, even algorithmic process. Determinism is not materialism. The identity transformation maps anything to itself. Facts are true, there's no reason why they can be false as well. Again, your understanding of algorithms is hopelessly flawed, as evidenced when you ask, "If the tools to create truths are not true, themselves, then the "truths" derived from them won't be true at all." Nonsense. What does it mean for a tool to be true? When a person sees an apple and grunts something to assign the grunt as a label for that apple, that ability to see the apple and assign a grunt to it isn't a truth, but that the grunt refers to the apple now becomes true. Your question is like asking, "If the paper and pencil aren't true, how can they be used to write a true sentence?"

As a student who has taken a course on algorithm analysis, I assure you, I know what an algorithm is. It has some relation to deduction. Namely, that deduction is done algorithmically. Machines are hard-wired and we implement algorithms on these machines to solve problems.

. Not only have you demonstrated no real knowledge of the two and their relationship, you even manage to be rude about it. You won't even do us the courtesy of explaining properly the relationship, you simply brush us off as idiots who are just too dumb to even see the relationship.

The closest you've come, even though it's been demanded post after post, is your one remark that they both rely on premises. Of course, that's so terribly flawed (and I have refuted your claims that deductions reliance on true premises means anything for all algorithms, but your posts suggests you ignored that) but it doesn't seem to matter to you.

I will try one more thing. I will try to address the point that is at the heart of the matter, and perhaps anyone else interested in this thread can pick up from here, because the style and content of your posts has totally put me off from this thread.

We want to establish that a deterministic being can express something meaningful that corresponds to some fact, some aspect of the way things are. If a being is able to percieve an object and assign some sound to it, and can then recognize that same object and repeat that associated sound, it can express that it recognizes that thing, and that, "there is that thing," and so we have a truth. So what is required here? First, that this being perceive this thing. We know simple animals can perceive things.

I don't know much about animal language, but it is not so hard to imagine that an early human saw its children, and decided to assign some special noise to it.

Even if it is algorithmic (which I guess it is), don't get confused by the ideas of "true premises" as that is entirely irrelevant. A man can see a child, and assign some grunt to it.

Given the basic axioms and rules of inference of propositional logic, no, a being can't make grunts.

Humans don't use axioms to name things.

Although it would be too complex for me to detail the evolutionary process and the brain-processes which associate the grunts with the children, I think it's very hard to imagine that we require a non-deterministic process to do this.

Please note that the assertion that an algorithm cannot create truths because it requires truths to begin with is false. That is true of deduction, not algorithms. The early human didn't deduce the grunt that would represent his child from a previous truth.

It produced the grunt (output) through some deterministic process using tools like observation and the tendency to name things (the processing) and the visual input of his child (the input). The tools like observation and the tendency to name things work deterministically, and arose deterministically through evolution. It's quite plain to see that, although I don't have evidence that this evolutionary account is certain, it's certainly very possible.

By no means is it contradictory to logic to suggest that things could have happened this way.

Evolution may not have accounted for our tendency to name things and observe things and organize our sensory data in a certain way, but to say that it is logically impossible for evolution to account for these things because it is algorithmic is borderline meaningless.

Machiveli wrote:
One last question (since you were so complimentary about me)

I believe you are confusing the question of whether truth can exist in a world without consiousness with whether it can exist in a deterministic world (assuming consiousness is deterministic)

Why do you require algorithms to conceptualise?

(I expect your answer will be along the lines of...
Truth requires correspondance and this requires statements to have meaning, meaning requires conceptualisation.)