Thursday, July 23, 2009

Clichéd Philosophical Question: What is truth? Propositions and Neural Nets.

The following is the result of fevered overcaffeination:

What is truth??

Well, like the questions “what is the moral good” or “what is evil”, we have to begin with the obvious and proceed in a step by step way to some sort of final answer.

As with the moral properties, truth is indeed a property.

How’s that for trivial? Hey, trivial is obvious. So, good starting point, right?

Well, if truth is a property, we need to next delineate all and only those things that possess truth as a property.

Just as we might designate the sorts of things that can have the property of redness, we should be able to generate a list:

Unlike the list of things that can be red, it seems this list of the things that can be true, is relatively short.

1. Beliefs
2. Sentences
3. Propositions

Taking these one at a time we now can ask just what it is about these that ‘gives them the ability’ to be true.

Remember, that we did something like this with moral properties. We first asked what sorts of things can have these properties. We answered; acts, and persons. Persons come to have moral properties thanks to their acts. Acts come to have their moral properties because they are done in a ‘knowing’ manner, as was defined in the earlier work. Knowing acts being those acts that are done with knowledge of their likely impact on eudaimonic values of other sentient beings.
So, applying this sort of reasoning to truth, what can we say?

Beliefs come to have the property of being true just in case they accurately represent their objects of representation. In plain English; if they paint a picture of the world as it actually is.

Trivial seeming.

Sentences are true for similar reasons. Propositions also.

Why include sentences and propositions as two distinct things in our list? Simply because there is a logical distinction between the content expressed by any given sentence, and the means of expression used to convey that content. Simple, well-worn example: the sentences “Schnee ist weis” and “Snow is white” express the same content by different means (different languages). What makes each of these sentences true is the fact that the content expressed represents the true state of affairs with regard to the optical properties of snow. Content that somehow lies behind the clothing of language is the truth bearing object. We reserve the term “proposition” for reference to that content.

So, strictly speaking, we can eliminate (2) from our list, or say that sentences can be true in a derivative sense, only because propositions can be true.

This raises the question then of the relationship between beliefs and propositions. Are beliefs and propositions identical, or are they distinct things?

Reverting to the obvious; beliefs are dependent for their existence upon sentient beings. It needn’t be the case that the sentient beings are language using sentient beings.

Beliefs are also things that affect behaviors and dispositions, although it is possible to have beliefs that do not obviously impact either behaviors or dispositions.

Since beliefs are things that can be true or false, it would seem then that they either are propositions themselves, or related to propositions in some way.

A sentence comes to have a meaning because it is part of a code. That code, or set of rules establishes semantic content to be associated with each discreet symbol or word. Somehow we are able to teach each other this association, creating consistency of use across individuals within a language community. In addition to the semantic content, there are teachable rules of construction, which reflect logical relationships between objects, their properties and other objects. For example, in English it is typical to place as the first element in a sentence, a word that indicates an active subject, the thing doing the acting, or that indicates the primary object of discourse, the possessor of some salient property. It is also typical to place a word indicating the action, or that has the function of ‘appending’ a property to that first object after this initial word. If there is a direct object of the action, a word indicating that thing will come last.
Obviously, this is simplified, but illustrative.

How does it relate to beliefs?

Hypothesis: Beliefs are not linguistic entities, but often are entertained or communicated via language. For instance, I might carefully consider my belief that Ben Franklin was a funny man by writing an essay about him.

Propositions are not beliefs, but things that can be believed (indeed they are things toward which we can have many attitudes like hope, fear, love, etc), and they are not linguistic entities but things that can be linguistically communicated.

Beliefs cannot exist without propositional content. Propositional content cannot exist without structure. What makes for a true proposition is a structural similarity between the proposition and the structure of something it is taken to represent.

Now, if a proposition isn’t a belief, and it is not a bit of language, and it also has structure that may or may not match some other structure, it must be constructed of something in order for it to have structure. (No duh.)

Analogously, atoms are ‘constructed’ of discreet sorts of things, electrons, quarks, etc, related to each other, that is structured, in various ways. These relationships give the specific elements.

So, what are the basic elements of propositional structure if they are neither linguistic, nor beliefs? What sorts of things would allow for structural specificity, allow beliefs to latch onto them, and would also be such that they could be interpreted as mirroring or failing to mirror other structures, and finally what sort of thing would also allow language as a means of communications, as a sort of transmittable analog of their structure?

Heck if I know. Maybe neurons?

If sentience, and consciousness are emergent properties of neurons, neuronal nets are malleable, and among the ways that neuro-structures can be formed is by conscious efforts (for example, we know that learning lays down new neural connective structures), then there may be some way to give a physicalists account of propositions as neural structures, which mirror structures outside the brain, and which in turn can be mirrored by language. Only with the emergence of sentience can these structures take on the role of being propositional content.

This requires that we conceptualize the emergent property of mentality as being something that can effect structural change within its own material basis. This would be a problem only if one were predisposed to epiphenomenalism. Heck, if a computer system can affect changes in its material basis, why not a conscious mind?