The Most Critical Question!

[AnssiH]

I think we need to hear from some biologists here. Perhaps I'm way off base here and don't understand the question

 

Yeah, you are a bit off base :)

 

I was just referring to a topic which Kriminal has got superficial familiarity of. If you are interested about what it is, I just so happened to write an overview about it to another potentially interested group;

http://tech.groups.yahoo.com/group/ai-philosophy/message/17841

 

Anyway, the crux of the OP shouldn't be lost. That particular analysis aside, the basic problem of epistemology is always that as soon as you make ontological assumptions of any sort, you are arguing on undefendable grounds (your arguments are valid only in so far that your assumptions are valid). And any assumption as to what the information to be explained "is" in any sense, is always an ontological assumption.

 

What's worse, ontological assumptions are necessary for human thought (because some "terminology" always needs to be used).

 

The only thing you really can analyze from purely epistemological angle, are properties that ought to be universally true for all ontologies (or world views). And that gets us to this thread :)

 

-Anssi

[Rade]

I rest my case.

Fine, but what you presented to be "your case" is a false argument, based on a false premise, based on your false "understanding", your false "use of terminology", that leads to a very convoluted philosophy on the relation between existence and what humans can know about existence. I understand you do not agree with this conclusion because it counters your worldview of the mental case you have invented. Resting a case, and resting a true case, are two completely different matters.

[Kriminal99]

Yes, you could say that.

 

Although the only thing I was pointing out in the previous post was how some things being "substantial objects" (photons, electrons) and other things being "stable processes" (waterfall, tornado, rainbow) are just semantical man-made categories in our world views. Much like "living things" and "dead things".

 

Ultimately all of these things are just defined things attached on some particular patterns on some "undefined information". Yet we are prone to think of some defined patterns as "real, substantial objects", which we think exist even when we are not there to define some patterns as those objects. And, we are prone to think of some patterns as "processes realized by substantial objects", and say that they only exist because we defined them in our heads.

 

What I'm saying is simply that the hard problem of consciousness arises from the tacit belief, that it is ontologically valid to separate reality in these types of categories. And as long as one thinks so, they must also think that consciousness cannot exist. Simply because our own explanation of "consciousness" must always be a representation of a process realized by a collection of defined persistent entities*; a single entity cannot represent any dynamics.

 

Very many people assume that the hard problem of consciousness will disappear once we manage to explain it correctly. That just reflects poor understanding of the epistemological roots of the problem.

 

Btw, that same undefendable categorization stands behind the quantum mystery as well; The seemingly idealistic features of Bell experiments go away when you realize exactly how those features arise from epistemological circumstances; the (so called) substantial objects reacting to our observation in such idealistic manner, is also a feature of an explanation of undefined information patterns, not a feature of any real entities "out there".

 

If you think about this, you can probably understand why I've been trying to put so much emphasis on the fact that the validity of our world view cannot be taken as an indication of its ontological correctedness, since we are really just measuring the validity of our expectations, not the validity of our particular terminology (set of definitions).

 

--

Now, you were able to connect this to the idea that we are talking about "an algorithm behind human intelligence", and yes, we are essentially talking about the properties of any algorithm that is capable of drawing expectations from some undefined information patterns. Any case of intelligent behaviour is a case of having pulled out valid expectations for some circumstance, and acted accordingly.

 

There are many specific possibilities for doing so (and we are not suggesting a single specific algorithm). But All those possible ways would have certain common properties to their definitions, as long as they don't make undefendable assumptions about the meaning of the information (i.e. as long as the definitions are explicitly attached to some recurring activities via inductive logic; only thing assumed is "the future will resemble the past")

 

An important thing to remember is just that, such an algorithm is already employed by the time we get to any idea that we are perceiving something. We cannot perceive undefined information by definition. On the other hand, we cannot think about/comprehend/perceive anything without having something defined (the definitions are our thoughts).

 

Association rule data mining for instance, it first requires someone to decide what exactly to measure. Before we can even refer to anything of interest inside the information, some decisions had to be made about how some undefined information is to be interpreted.

 

So we are really talking about the first very fundamental steps of that algorithm; something that manages to pull out expectations without knowing at all what some information is supposed to be or how is it supposed to be read. All the possible ways to consciously analyze that data, can only arise after some definitions have been generated.

 

 

 

Still wanted to explicitly say that no, that's not what I'm talking about. This should be viewed as a discussion of very real epistemological issues. The problems mentioned above arise from those epistemological issues that we just can't get around because of what "explanation" (or "world view") actually is (i.e. how we measure its validity).

 

A lot of people think those problems arise from our explanations being subtly wrong, which reflects their belief that ontological correctedness is a valid topic, which reflects their poor understanding to the fact that such a thing can never be measured, by the definition of "explanation"...

 

-Anssi

 

Ok. The first steps of algorithm are related to Nietzsche's philosophy actually. It's more like a law of biology. Whatever doesn't kill you makes you stronger. This rule drives some aspects of how DNA works, our motivation, our learning algorithms etc etc. Isn't it cool how plant cells and animal cells have some of the same functions realized by different means? The obvious explanation for this was that those functions are needed to survive. Lots of randomness till a functioning cell, and then remember what you did that worked.

 

If an arrangement of cells was bad, then by definition of bad you wouldn't be able to keep reproducing it. So do it again.

 

So yeah... if a brain had any purpose at all it would be first and foremost to record events. So it would pretty much end up realizing an algorithm similar to association rule data mining on it's own. (A few slight modifications involving how to generalize from experiences)

 

If an experience killed you, you wouldn't be around to miss it. So simply miss (wish to re-experience) everything you experience. etc etc.

 

If you look at the explanation for consciousness, it may not answer the problem you are trying to answer but it may help draw lines around exactly what you are asking about. We constantly recognize correlations in our environment and store them for future use. But as you know (as a student of epistemology) these correlations can fail... They might be oversimplifications or "Overfit" as we call it in Machine Learning. Checking them constantly is of course prohibitively expensive.

 

So what can you do? Well assume all this recognition is to help you work towards some goal. You can have a second version of the algorithm, with less permanent memory. All it does is check on how correlated the current conditions are with achieving your goal. It outputs emotions to the conscious mind that may cause you to go back and redo some old correlations that you had in place if they led you to a bad outcome.

 

So with this in mind we can separate memories and the reasons and perhaps causes of emotions from consciousness. That just leaves raw feel for your question. Fundamental property of matter maybe? What does raw feel without any input even feel like? Probably like fast forward until some input.

[Kriminal99]

I was just referring to a topic which Kriminal has got superficial familiarity of.

Huh?

[AnssiH]

Huh?

 

I was referring to DD's epistemological analysis. I've seen you posting to those threads few times and you seemed to make a connection.

 

Fine, but what you presented to be "your case" is a false argument, based on a false premise, based on your false "understanding", your false "use of terminology", that leads to a very convoluted philosophy on the relation between existence and what humans can know about existence. I understand you do not agree with this conclusion because it counters your worldview of the mental case you have invented. Resting a case, and resting a true case, are two completely different matters.

 

You seem to misunderstand what I was saying, and I must say now I have doubts about your understanding of the hard problem of consciousness. I was saying that it arises as a direct consequence of categorizing reality into "objects" that are substantial, and "processes" that are only what we defined. To rephrase the problem; "according to our understanding of reality, our consciousness should not exist".

 

I pointed out that this categorization is something we make in our explanation of reality; it is required for being able to represent expectations at all. Yet, the underlying information doesn't explicitly tell us what exists. Epistemologically all "substantial objects" and "stable process" are on similar footing; each is based on recurring information patterns of some sort. It is simply a feature of our specific explanation that we categorize some of those things as "real objects with identity" and other things as not (and what exactly we put into which category).

 

I commented that most people don't seem to be capable of grasping this problem at all (from the level of epistemological fundamentals), and your response reflected exactly that inability. Just as soon as I pointed out that there's no way to defend the ontological validity of that categorization choice, you commented that "we are categorizing things into these categories [add your examples], thus it is fair to do so", without the slightest attempt to try to argue why.

 

Essentially you completed my argument, that is why I said "I rest my case".

 

You know it's a bit frustrating that everyone seems to interpret this as if I'm talking about my beliefs, when I'm just trying to point out what sorts of things we cannot have explicit information of, beyond simply believing so. Any attempt to argue otherwise can only be a statement of someone's belief, like your statement towards your belief about how photons are substantial things. It's not really that different from the argument that "the bible is infallible voice of God because it says so in the bible". Just because your explanation/interpretation of some information contains some defined object, it doesn't actually summon such an object into existence out there.

 

There are two paradoxical features in our common world views that make the ontological validity of our current definitions very suspect to me, and those are the hard problem of consciousness, and the seemingly idealistic features of modern physics (i.e. the complications arising from the role of observation). It should be interesting to all, that both can be traced to exist as entirely epistemological consequences of representing expectations of recurring patterns.

 

But I guess disbelief of one's own beliefs is also a paradox.

 

-Anssi

[joekgamer]

An object is a recurring pattern of matter/light/etc. For example, an airplane is a collection of matter that we call an airplane (i.e. stays in the air using an airfoil to create a pressure differential [lift]).

 

A process is the way an object changes (an action). For example, flying is the process of staying in the air.

 

In other words, the difference is purely semantic.

 

Regarding your post, since conciousness is simply how we percive objects, processes, etc, it is not an actual object.

 

Was this more relevant?

[AnssiH]

Was this more relevant?

 

A little bit I guess, but let's see if we can get little deeper :)

 

An object is a recurring pattern of matter/light/etc.

 

So then, what is "matter"? I.e. what is it that tells us something is "matter"?

 

-Anssi

[joekgamer]

Matter is simply something that has mass. This excludes photons, because while they do have mass-like properties, they have zero mass.

[AnssiH]

Matter is simply something that has mass. This excludes photons, because while they do have mass-like properties, they have zero mass.

 

What is it that you do to find out whether something has got mass?

 

-Anssi

[joekgamer]

If something is 'solid' in that it contains particles, then by definition it has mass.

[Kriminal99]

If something is 'solid' in that it contains particles, then by definition it has mass.

 

You keep responding as if there is some objective reality, but this objective reality is created as a function of our subjective reality. We think of it as more objective because other people seem to experience the same subjective reality so it seems the objective is a function of many people's equivalent subjective realities. But then again, we only have our subjective perspective to imply those people even really exist. Once we have reasoned that this objective reality exists, we forget and automate this process. But philosophers often find it useful to go back and consider perspective to answer some more difficult questions. It is useful for providing more precise definitions and understanding of concepts like time and many others. This is how people like Einstein amaze everyone else with theories that seem to come out of nowhere.

[Kriminal99]

So the question is about what is the difference between an object and a process?

 

In the algorithm I have created, it goes something like this. I input a picture, in the picture is a book inside a room. I move the book to a new room and input a picture of that. The algorithm identifies that the book is similar, but the background is not. Thus the image of the book is separated from the two backgrounds. How?

 

You use the first image to create a network of nodes with certain strength connections to one another. The connections cause one node to light up if enough connected nodes are also lit up. Then you feed the second image in to the network. The book nodes light up the way they are supposed to but the background nodes do not. You create a new network from the inputting of the second image to the first image's network. The rule you use to do this causes the book's network to be severed from the background. This process theoretically involves something called hopfield nets, but actual hopfield nets are computationally expensive so I used a reduced version that is theoretically equivalent in the way it is used.

 

Anyways the book is an object. Less than infinite precision allows you to recognize the book as the same thing even from different angles. You can even reduce the precision to make it easier to recognize things on purpose, and you can perform comparisons at different levels of precision.

 

The lines between an object and a process are indeed blurred. (Epistemologically and in my algorithm) We think of an object as 3 dimensional, but to do this we think of a 2 dimensional object from different angles. Rotation is a process...

 

Similarly we can identify a connection between say a person going through different stages of walking up and picking up an apple and eating it. We could watch someone give a command and then watch that person carry out the command. Then we can associate the command or name of the process with our observation of the process. The commander can express dissatisfaction with superfluous steps, or can give new commands prior to new processes being carried out. The point here is by observation such an person (or my algorithm) can pinpoint the beginning and end of the process.

 

I haven't given every mathematical precise detail, because there are a lot of such details. But there are no barriers to this type of approach. I believe it is how human beings learn as well.

[joekgamer]

So the question is about what is the difference between an object and a process?

 

In the algorithm I have created, it goes something like this. I input a picture, in the picture is a book inside a room. I move the book to a new room and input a picture of that. The algorithm identifies that the book is similar, but the background is not. Thus the image of the book is separated from the two backgrounds. How?

 

You use the first image to create a network of nodes with certain strength connections to one another. The connections cause one node to light up if enough connected nodes are also lit up. Then you feed the second image in to the network. The book nodes light up the way they are supposed to but the background nodes do not. You create a new network from the inputting of the second image to the first image's network. The rule you use to do this causes the book's network to be severed from the background. This process theoretically involves something called hopfield nets, but actual hopfield nets are computationally expensive so I used a reduced version that is theoretically equivalent in the way it is used.

 

Anyways the book is an object. Less than infinite precision allows you to recognize the book as the same thing even from different angles. You can even reduce the precision to make it easier to recognize things on purpose, and you can perform comparisons at different levels of precision.

 

Yep. This is exactly (or, at least, close enough to be almost so) how we recognise patterns of matter (objects) when they are represented in a less-than-perfect media (e.g. photographs, television, 3-d models, etc.).

 

The lines between an object and a process are indeed blurred. (Epistemologically and in my algorithm) We think of an object as 3 dimensional, but to do this we think of a 2 dimensional object from different angles. Rotation is a process...

 

Remember - an object is a pattern of matter. If we see a representation of it (e.g. a photograph), then we make the connection that the photograph is of a book. Seeing the book/photograph/ect. is a process, and so is taking the photograph, but the end result is an object.

 

I believe it is how human beings learn as well.

 

Yes, it is, only we learn by modifying the synaptic weights in order to make our neurons fire only when certian inputs are triggerred, not just by activating the neurons. If you have some experience in programming, then look up artificial neural networks. They work on the same principle, but use math functions to simulate neurons, instead of using actual physical neurons (although there is some research into that area).

[Rade]

To rephrase the problem; "according to our understanding of reality, our consciousness should not exist".

This comment makes no sense to me. According to my understanding of reality, consciousness must exist.

 

..the underlying information doesn't explicitly tell us what exists.

I claim that underlying information always tells us explicitly what exist. See next comment...

 

It is simply a feature of our specific explanation that we categorize some of those things as "real objects with identity" and other things as not

All actions are caused by entities' date=' this is a fact. Every entity has certain attributes and no others, this is a fact. Both of these facts are known by observation, not by explanation. Stated another way, causality is a fact independent of consciousness (thus explanation). Regular patterns of information are the real objects with identity, stated another way, Existence "is" Identity.

 

I commented that most people don't seem to be capable of grasping this problem at all (from the level of epistemological fundamentals)...

Well, have you considered that this is because they do not so grasp because there is no such problem to grasp.

 

Just as soon as I pointed out that there's no way to defend the ontological validity of that categorization choice' date=' you commented that "we are categorizing things into these categories [add your examples'], thus it is fair to do so", without the slightest attempt to try to argue why.

Well, in fact, you never pointed out any such thing. What you did is make a number of commandments that there is no way to defend categorization, but your arguments are not convincing.

 

You know it's a bit frustrating that everyone seems to interpret this as if I'm talking about my beliefs' date=' when I'm just trying to point out what sorts of things we cannot have explicit information of, beyond simply believing so.[/quote']But don't you see, this comment is "your belief"...your belief concerning "things we cannot have explicit information of". What I have been trying to communicate is that I do not agree with your belief about those "things" that we cannot have information about.

 

like your statement towards your belief about how photons are substantial things

I never said such was my belief about photons' date=' please re-read what I said. What I said is that it is a scientific fact that photons are a substantial thing, an elementary particle of nature. Are you trying to argue that it is not such a fact ?

 

it is not really that different from the argument that "the bible is infallible voice of God because it says so in the bible".

So, in your philosophy, there is no difference between a scientific fact (i.e., that photons are a substantial thing) and what is claimed in the bible--correct ?

 

There are two paradoxical features in our common world views that make the ontological validity of our current definitions very suspect to me' date=' and those are the hard problem of consciousness, and the seemingly idealistic features of modern physics (i.e. the complications arising from the role of observation). It should be interesting to all, that both can be traced to exist as entirely epistemological consequences of representing expectations of recurring patterns.[/quote']Well, no, direct observation (the act of perception) has nothing at all to do with representing expectations. Perception of objects that exist is prior to expectations (i.e., explanations of what is perceived), in fact, many (nearly all) of the patterns perceived never reach the consciousness to allow for expectations to be made.

 

But I guess disbelief of one's own beliefs is also a paradox.

No, such is called a contradiction and hence always false. A paradox must by definition be a statement that appears to be a contradiction, but may in fact be true.

[joekgamer]

I never said such was my belief about photons, please re-read what I said. What I said is that it is a scientific fact that photons are a substantial thing, an elementary particle of nature. Are you trying to argue that it is not such a fact ?

 

Photons are techincally just fields with mass-like properties, so therefore they have no mass and so are not "substantial" in a colloquial sense. It seems that the misunderstanding is simply a semantic one.

[Rade]

Photons are technically just fields with mass-like properties, so therefore they have no mass and so are not "substantial" in a colloquial sense. It seems that the misunderstanding is simply a semantic one.

Thank you, but the difference in philosophy between AnissH and I is more than semantics, but one of using different premise to form a worldview. In short, I build my worldview from the primacy of existence, AnssiH does not. It really is that simple. Now, if he responds that I am incorrect, that he also builds his worldview from the primacy of existence, then I would have to agree with you, our differences are only one of word use or semantics. Perhaps I error, I'm sure he will let us know.

 

Photons are a type of "gauge boson" that are without rest mass. They can take the form of particle or wave. They have relativistic mass via momentum. They have spin. They are force carrier. In other words, photons "exist" independent of the entities that they interact with, that they luminate. Photons are substantial, they exist independently and can be coupled to another entity and to human consciousness via the filter of visual perception.

 

There are many different types of reality, only one of them meets the definition of naive reality. Naive realism holds that human senses, such as visual perception, provide the consciousness the pure "thing-in-itself" essence of some entity that exists. Thus naive realism holds that the human senses are valid because the sensory qualities of the entity exist independent of human perception. I find this to be a false attempt to uphold the validity of the senses. However, just because one holds that naive realism is false, this does not mean that the evidence provided by the senses is not connected to reality, that the senses do not provide the consciousness what I would call a veiled reality of the entity that exists independent of the human mind. What I mean is that, although the senses can not provide the pure "thing-in-itself" essence of entity, they can provide a veiled reality of the "thing-in-itself" essence of the entity coupled to the force carrier of the information. What enters the consciousness is not the thing-in-itself essence of the entity, but the thing-in-itself coupled to the force carrier of the information about it (photon) which interact with perception prior to consciousness. In other words, what enters the consciousness is a process (the fulfillment of the thing-in itself as a potential entity) and not the entity that exists as actual thing-in-itself. But, of critical importance is that this does not mean that there is not actually an entity that exists prior to perception with an actual thing-in-itself essence that is outside the ability of human consciousness to directly interact with.

 

I find that the term used by AnssiH to describe the concept "entity", his "substantial entity", has no meaning in particle physics. The physicist does not discuss difference between substantial entity and non-substantial entity. Given that substantial means "exists" to AnssiH, it would be nonsensical for a physicist to say they study entities that exists (substantial) and try to compare them to entities that do not exist (non-substantial).

 

So, we come to the question asked by AnssiH, is a "process" substantial, does a "process" exist in the same way that an "entity" exists ? Do both provide similar patterns of undefined and unknown information that can be observed (known) and placed into past experience ?

 

Suppose a pile of bricks and a bricklayer tasked to build a wall. Are the bricks and the completed wall, both being entities that exist, the same as the process conducted by the bricklayer to position each brick to form a wall ? I argue that while both entity and process exist, they exist in fundamentally different ways. That is, entity can exist fully in a state of completion as actual (brick), or in a state of an actual with potential (wall that can be taken apart), but process can only exist fully as a potential, never actual. Let me try to explain my thinking.

 

All process is a form of motion. I define what is substantial can (1) exist fully in a state of completion (a single brick) and (2) exist fully in a state of potential (the process of building a wall), (3) exist in both states simultaneously as a completed potential (the wall of bricks completed with potential to be taken apart to become again single bricks).

 

Thus I conclude that while all entity exists actually (brick and wall), all process is constrained to the fulfillment of what exists potentially (i.e., the pile of bricks with possibility of becoming a wall during the building actions of the bricklayer). Entity is actual (with possibility of becoming potential), however process is constrained to be fully potential. Here then we find the reason that I disagree with AnssiH, he would find actual in process, whereas I claim that only potential exits.

[AnssiH]

Hi Polymath,

 

I've been asking those questions in order to get to the fact that we are talking about our own definitions. If you've been trying to think of the issue when you gave the answers;

 

- An object is a recurring pattern of matter.

- Matter is simply something that has mass.

- If something is 'solid' in that it contains particles, then by definition it has mass.

 

you have probably also noticed how hard it is to actually explain what we mean by "substantial things", without getting to circular definitions. We started from me pointing out that "substantial objects", i.e. what you would call "solid objects" are, from epistemological standpoint, some patterns of information that we interpret as meaning "there is a solid object there".

 

Observe the circularity of your attempt;

 

"What are substantial/solid objects?"

 

- Objects that are a recurring pattern of matter...

... matter is something that has got mass...

...mass is something that solid things have.

 

I.e. solid objects are those things that are solid.

 

The actual issue is of course that there exists some definable behaviour (or information) that you interpret, according to your world view, as meaning "there is a solid object there". The definitions of "mass" and "particles" and all those things are just your terminology with which you comprehend that information. (And btw in the terminology of modern physics, what you call "solid" is an electromagnetic process, not an intrinsic property of an element)

 

You probably already saw Kriminal discussing his perspective about how "objects" are supposed to be recognized from some information. Like he is pointing out (I think he is anyway), it is actually entirely a function of your world view, what sort of information gets recognized as a substantial object, and what is thought to be just a persistent process (like a rainbow or a tornado).

 

Notice how this line is indeed entirely blurred in epistemological sense. It is your world view which decides where to draw that line, just like it is your world view which draws the line between "dead" and "living" things. Epistemologically and ontologically, in both cases, we are really just talking about our own definitions.

 

So think about the "hard problem of consciousness" argument, that we should be "unconscious zombies", because having a process that interprets reality, does not entail that a subjective experience exists.

 

-Anssi