Friday, April 1, 2011

Pragmatism Parts 1-3: Intro and Science

These are the thoughts that have led me to what I believe is some variant of epistemological pragmatism. Some might call me a Kantian after this post, though I don't think I really am. Note that this is a short essay, so I'm painting a picture here in very broad strokes; there are going to be gaps in my reasoning and missing details.

(1) A Rejection of Metaphysics

There is no direct metaphysical connection to reality that enables us to experience our sense data as anything but sense data. To use Kantian terms, we have access to phenomena (our experiences) but not to noumena (things as they are). In fact, it would be impossible to know noumena under any circumstances, since we can't know anything except by experiencing it, which means necessarily knowing an experience instead of the thing in itself.

There are also certain metaphysical assumptions that cannot even be experienced, but seem to be fundamental to how we organize our experiences. Causality, for example, is the assumption that certain events necessarily follow from certain other events. But we cannot experience causality. We can observe a brick hitting a window, followed by the window shattering, but we cannot experience "the brick causing the window to shatter." Causality is a way we make sense of our observations, but it is never observed as a thing in itself.

This doesn't mean that experience, assumptions about causality, etc. are bad, unuseful, etc. It simply means that we have no reason to believe, one way or the other, whether they actually represent noumena. It's impossible to know.

(2) Evolutionary Psychology of Truth

There are two popular theories regarding truth in the world of philosophy. These aren't the only theories, but they are the most common.

The first is correspondence theory, which says something like "a belief is true when it corresponds with the real world." As I stated before, we have no way to directly access the real world, so this theory can be thrown out from the get go. We could invent a modified version that states "A belief is true when it corresponds to our experience," but this theory isn't particularly useful, since there is no way to falsify an experience as such. Nobody ever looks at an apple, experiences the sensation of seeing a red color, and concludes that the apple is green. A good theory of truth needs to provide a practical means of differentiating truth from falsehood in order to be useful.

The second popular theory is the coherence theory, which says something like "a belief is true when it can coexist with your other beliefs without creating a contradiction." The problem with this theory is that someone could, in theory, invent several internally consistent bundles of beliefs. These bundles could be internally consistent but not consistent with each other, forcing someone to choose one. A coherence theory provides no means of determining which internally consistent bundle is the correct or truthful one.

This leaves us in a bit of a bind. If the truth isn't "beliefs that correspond with the real world," and it isn't "beliefs that coexist without contradiction," then what is it? What theory can we use to distinguish true propositions from false ones? This is where I think evolutionary psychology comes into play.

Roughly speaking, both the correspondence and coherence theories of truth are implicit to our thinking from birth. We naturally seek to remove contradiction between our beliefs, and to confirm our beliefs through experience. We don't come into this world with knowledge of a distinction between phenomena and noumena. By nature, we conflate the two. The distinction has to be teased out through careful, analytical thinking.

Let's step back a bit and ask; what is the purpose of truth? Why do we feel driven to distinguish true propositions from false ones? What is it about our nature that makes this necessary?

The basic principle of evolutionary psychology is that the human brain, like all other parts of the human body, is a product of natural selection. Your arms work the way they do because they enable your survival and reproduction, causing the genes that created your arm to replicate more successfully than competing genes. The same is true of your brain.

Knowing this, we can ask; what is it about distinguishing true ideas from false ones that enables our genes (most specifically, the genes that help in building a mind that is driven to distinguish truth from falsehood) to replicate more effectively? I think the answer is simple and practical.

If I believe a lion will eat me, and this causes me to run away from lions, I will survive and reproduce much more effectively than if I don't have this belief. Natural selection would lead to minds that believe ideas that result in more effective gene replication. The purpose of truth, in order words, is pragmatic; it serves a useful function. An idea is true to me if believing it leads to results that I prefer over the results of believing something else.

(3) Pragmatism

I think this pragmatic theory of truth is a very useful one for a number of reasons. First, it removes the need for any particular theory of metaphysics. Metaphysics is irrelevant if we are no longer trying to connect the concept of truth to noumena (things in themselves.) In theory, pragmatism is compatible with any metaphysical theory.

Second, it eliminates epistemological nihilism without needing to appeal to metaphysics. True and false aren't arbitrary and subjective, regardless of whether or not they correspond to noumena. True and false have measurable, understandable implications that enable us to continue to distinguish between the two.

Third, it eliminates the need for any concept of absolute truth. In fact, pragmatism allows for degrees of truth. Let's say I have 3 theories to explain the same phenomenon; the theory that is most useful for understanding the phenomenon is the "most true", the theory that is least useful is the "least true", and the remaining theory is in the middle. The possibility remains of a 4th theory that is even more useful that the current theories.

The human mind is necessarily finite to a great degree, and we simply cannot hold nearly as much information in our minds as we would like. This forces us to generalize, omit data, make unproven assumptions, etc. in our thinking on a constant basis. The benefit of a pragmatic theory of truth is that the concept of truth remains useful even in this kind of environment.

(4) Truth is prescriptive

This is one of the more interesting conclusions I have drawn. I'll try to explain how I came to this idea.

What is it we mean when we say "truth", when discussing theories of truth? Each theory has its own definition; the correspondence theory defines truth as belief that correctly corresponds with reality. Coherence theory defines truth as belief that coheres with other beliefs. And so on. But what is the thing each theory is trying to describe? What is meant by truth that underlies all of these theories?

To answer this, I think we have to ask why we make the distinction at all. Why do we call some beliefs "true" and some "false"? What function does this serve in our lives? The answer is pragmatic. Some beliefs function better than others for achieving our desires, and the distinction exists to enable us to select the beliefs that function better.

This is where I think the prescriptive nature of truth reveals itself. Truth is preferable belief. That is, a true belief is a belief that one ought to have, and a false belief is a belief that one ought not have.

When the correspondence theory says "truth is belief that correctly corresponds with reality", what it really says is "It is preferable to believe things that correspond with reality." When the coherence theory says "truth is belief that coheres with other beliefs", what it really says is "It is preferable to have beliefs that cohere with other beliefs."

This turns the whole is-ought problem upside down. If descriptive claims, truths, are a CLASS of prescriptive claims, then there is no fact-value distinction and no is-ought problem. Ayn Rand said "Every 'is' implies an 'ought'," but it would be more accurate to say "Every 'is' is an 'ought'."

(5) What does it mean for one belief to "work better" than another?

To answer this, we have to dig deep into what a belief is. I think our everyday conception of what beliefs are and how they work, while intuitive and functional in everyday life, is flawed.

The human experience is essentially a big feedback loop. We receive input through our various senses. Our nervous system (mostly our brain) then performs a variety of calculations with this input. At the end of the loop, our nervous system sends signals to various parts of our body to produce movement, chemical production, etc. In reality, we aren't one big feedback loop, but many, many such loops running in parallel to one another, affecting each other, etc. But one big loop serves as a useful analogy here.

To put it more simply, someone sees a lion nearby, the brain does a calculation and concludes "Lion bad. Run away." This is followed by electrical impulses being sent to various muscles to cause him to run away.

Within the brain, we store ideas. Ideas consist either of past experiences (memories) or the results of calculations (a mix of what we would call memories, ideas, abstractions, values, etc.) Ideas are involved in the calculation process; this is why someone might experience a fear of dogs after being bitten by one. The past experience has changed how the brain responds to the stimulus of seeing a dog.

The purpose of an idea is to make our brains more accurate when calculating what actions to perform. If I have an idea like "Turning the tap causes water to come out," I will be able to more easily satisfy my need for water when I feel thirsty.

The idea of "water" is itself just an algorithm for getting what we want. We experience a desire, thirst, that is satisfied when we place water in our mouths and swallow. In our brain, the label "water" refers to an algorithm that basically gives a simple true/false response. If it returns true (that is, if our brain calculates that what we are looking at is indeed what we would call "water"), various other algorithms in our mind are engaged. We now know that, if we move our bodies in such a way as to get this "water" in our mouth, we can satisfy our thirst.

Now suppose there were two competing algorithms, each vying for the label "water." Algorithm #1 essentially defines water as "liquid." If we observe something and it seems to be a liquid, Algorithm #1 will return "true" and we will act as though the thing we see is water. Algorithm #2 is more stringent and involves more calculation. Liquids that have certain colors or textures are dismissed as not being water. Additionally, ice and water vapor are considered water.

If I am really thirsty, and I encounter a large bottle of bleach, which algorithm will get my needs met more effectively? Clearly Algorithm #2. Algorithm #1, in this case, might kill me, since I would assume the bleach was water, drink it, and be poisoned.

Essentially, all ideas are algorithms for interacting with our experiences, and beliefs are a class of ideas. A belief is a predictive idea; if I encounter X, I will also encounter Y. If I see dark cloud overhead, I should expect to see rain soon. In the above example with water, water is a concept, but the belief that drinking the bleach will satisfy my thirst is, well, a belief.

Ideas (including beliefs) that get our desires met more effectively are preferable to ideas that don't, and "preferable belief" is essentially the definition of truth.

(6) Pragmatism and Science

One fear that has been expressed to me is that pragmatism prevents us from preferring one scientific theory to another. For example, when Galileo claimed that the Earth actually goes around the Sun (and not the other way around), would there have been any pragmatic reason to accept his new theory?

To answer this, we have to ask a basic question; what is it about scientific theories that makes them useful in general? If we can show that (A) scientific theories are useful for clear, understandable reasons, and (B) that those reasons can be a standard for judging scientific theories as better or worse than one another, then we have retained the pragmatic outlook without losing the benefit of scientific progress.

To find out why scientific theories are useful, I think we simply have to look at what scientific theories are used for. Someone committed to correspondence theory might say that a scientific theory is used to describe the world in some way, but as a pragmatist I recognize that "the world" is simply the world of my experience, and that usefulness is intrinsic to my ability to understand my experience.

The real power of a scientific theory is prediction. A good theory is one that predicts future phenomena. The word future is being used to refer to the future of the person making use of the theory, not necessarily the future in general. For example, a theory of evolution helps someone understand events that occurred in the past, but the theory is predictive in that it predicts what that person will experience in the future, even if those experiences are the result of something in the past.

It is in their predictive power that scientific theories are useful. But suppose you have two different scientific theories with comparable predictive power? How do we know which is the better theory?

Since usefulness is our standard, there is a second aspect of a theory that comes into play; efficiency. The less time and energy that we have to invest into using a theory in order to get the results we want, the better. A more efficient theory is more useful than a less efficient one. The key element of a theory that determines its efficiency is complexity. A simpler theory will, generally speaking, be more efficient than a more complex theory because it requires less time and energy to get the desired output. This is essentially the pragmatic justification of Occam's Razor.

So if two competing theories give similar predictions, but one theory is simpler than the other, the simpler theory is preferable. It gives us the same results with less time and energy.

Suppose we have two different theories with equal predictive power and equal simplicity. How do we know which theory is better? This is where a third factor comes into play; how easily can the theory be combined and integrated with our other theories? Our ideas, values, etc. don't exist in a vacuum. They have to coexist and intermingle. Cognitive dissonance is, essentially, the emotional experience of having ideas that don't harmonize well with each other.

If a theory does not harmonize well with your other thoughts, it loses efficiency. It becomes awkward, unwieldy, and difficult to use. Moreover, it is far less likely to contribute to other ideas that you might have. It becomes a tool with little use other than the one, exact task it was built for. It's over-specialized, basically.

Let's bring all of this back to the original question about Galileo's theory. Let's compare it, step by step, to the geocentric theory (where Earth is at the center of the solar system) and see how they stack up:

Predictive Power - In the context of what was known at the time, both theories at first appear to have equal predictive power. Equations existed for both theories that could predict where various heavenly bodies would appear in the sky. However, Galileo's theory accounts for phases in the other planets that the geocentric theory cannot. This means that Galileo's theory is better at predicting the future experiences of observing heavenly bodies. Winner: Galileo.

Efficiency - While both theories could predict where various heavenly bodies would appear in the sky, the mathematics of Galileo's theory were far simpler. This means less time and energy have to be invested when using Galileo's theory. Winner: Galileo.

Harmonization - This is where things start to get a bit tricky. For a Catholic, the geocentric theory harmonizes strongly with the teachings of the church. Earth is the center of God's creation, with Man being the greatest and most important creation of God. Since the Bible describes the formation of the cosmos in a very geocentric manner, geocentrism makes more intuitive sense.

At first, it would seem like pragmatism has been defeated. After all, if most of the population is Catholic, the gains in efficiency and predictive power might well be overshadowed by the loss of having a theory that easily harmonizes with their other theories, ideas, values, etc. Science is halted by the church, and it would appear that this is how it should be, pragmatically speaking. This is where things get interesting though; Catholicism is itself a bundle of theories that themselves can be assessed using the above criteria.

If Catholicism itself is a less predictive, less efficient worldview than the scientific, atheistic worldview, then Catholicism is in crisis from a pragmatic perspective. And if Catholicism is thrown out, Galileo moves in quite easily. While it is true that, within the context of Catholicism, Galileo's theory is less useful, the context of Catholicism is not itself the basic human context. It is only in the barest context of the human experience that we can, in the ultimate sense, assess the pragmatic value of a theory. That doesn't mean that we always would/should do this in practice; it simply means that pragmatism is not simple affair only concerned with immediate needs and desires. Sometimes it can uproot long held ideas and entire worldviews in a process that is painful but ultimately beneficial.

(7) Pragmatism and Scientific Motivation

Another fear that has been brought to my attention is the fear that pragmatism essentially kills the motivation for scientific inquiry "for its own sake." The argument, if I understand it correctly, is this:

"Sure, Galileo's theory might be pragmatically better than competing theories, and maybe a pragmatic approach does account for most of the positive advances in scientific theory in the last few hundred years. But what use is it if all it does is predict my personal experiences? I'm a scientist because I want to understand the world as it is! I want to show how things actually are, and if I truly believed that this goal was impossible, I would have no motivation to do science anymore!"

I don't think anyone, truly, pursues "truth for its own sake." The idea is that the noble scientist is only concerned with truth, rather than his own interests, and will pursue truth come hell or high water. This is, frankly, not the case. Everyone pursues everything out of some interest that they have. The scientist that pursues what he calls truth is pursuing it out of self-interest.

The important thing to understand is that coming to terms with our experiences is a very important and gratifying psychological experience. Whether we call those experiences "my personal experience", "the truth", "the real world," etc., we have a profound drive to develop an understanding of that experience and share that understanding with others.

One need only ask a scientist "Why are you a physicist and not a biologist?" to discover that there is no such thing as the impartial pursuit of truth. Every pursuit is related, in some way, to the values and desires of the pursuer, if for no other reason than the fact that we cannot pursue everything at once. This is also why there is no such thing as unbiased news reporting; since one cannot say everything at once, one necessarily has to choose to say some things and not other things, and one's values and ideas will determine which of those things you say.

Fundamentally, though, this particular scientist is operating under false pretenses. First, the moral nobility he assumes for his pusuit; everyone is after the same truth that he is after in all of their attempts to come to terms with their experiences. The "noble altruist" meme (which is certainly not exclusive to scientists) should simply be thrown out altogether. I won't go into the psychological nitty gritty here, but it is not a sign of psychological well-being but of personal grandiosity and narcissism.

Secondly, the history of science shows that his belief that he is finding "the truth" (in a non-pragmatic sense) is completely false. Theories are frequently revised or thrown out in favor of newer, better theories. No scientific theory is ever final or "the truth"; it's truthfulness is always contingent on there not being another, better theory. A pragmatic epistemology is the only epistemology that gives any meaning to the word truth in the field of science.

Thirdly, I think the argument itself falls apart because I don't see most scientists as having this worldview. Scientists (any profession, really) are driven to science for any number of personal, subjective reasons. Some might just be in it for the paycheck. Others might have a profound desire to understand a certain slice of their own experience (even if they call it "the world" or "the universe"). Others still might be in it for fame, recognition, or honor. I don't think there is any actual risk of science being abandoned due to epistemological pragmatism.

1 comment:

Jonathan Connelly said...

“To use Kantian terms, we have access to phenomena (our experiences) but not to noumena (things as they are). In fact, it would be impossible to know noumena under any circumstances,”

By what means did you discover something that is by its very definition is “impossible to know… under any circumstances,”? If something is impossible to know, then you do not know it; nor could you even conceive of it.

Parmenides said this 2,500 years ago, "What is, is; and what is not, is not." The not, cannot be considered, nor even be thought about. What is not, is NOT.

To claim a fact as “unknowable”, while claiming to be knowledge; is a contradiction in terms. Identification presupposes knowledge. So then, how exactly did you identify what is “impossible to know.”? In some other reality, a divine revelation from God or “through a glass darkly”?

Or, are you just going to take Kant’s word for it, “Hey kid, you can’t know facts, or the ultimate constituency of existence. Your senses are invalid, so don’t try, just trust me on this. It’s a fact!” How? Did Kant have some special ability to know the “knowable” facts of existence? Kant's metaphysics is like some cruel philosopher’s joke.