In the previous installment in this series I introduced the concept of information, which I defined as correlations between states. Commenter Samuel (whose profile says he is a Young Earth Creationist) pointed out that:
This is a fair criticism based on what I wrote:Shannon entropy implicitly requires a mind to decide which distinct states will be recognized in order to assign a value to "n" (where n is the number of possible states.) [Emphasis added]
Notice that when we say that a switch or a light has two states we have ignored a lot of details. The actual physical state of a switch or a light includes a lot more than just whether it is on or off. For starters, there is the actual physical location of the light or the switch. A switch can be mounted on a wall, or it can be part of a lamp, or connected directly to some wires and not be mounted on anything at all. But these kinds of details don't matter for the aspect of a light switch's behavior that we actually care about. [Emphasis added.]
This implicitly tied my definition of information to the concept of caring about things. What if there are correlations that we don't care about? Do those correlations also contain information?
The answer is yes! The reason I introduced things "that we actually care about" is not because this is essential to the concept of information, but just to make the concept easier to understand. Information is not limited to things we care about, it's just easier to think about if we focus our attention on familiar examples like lights and switches and coins. More specifically, it's easier to talk about if we limit our attention to things we have words for, and even more specifically, to words whose meanings are unambiguous. Given that even prosaic words like "chair" have fuzzy boundaries, this is a tall order.
The reason for using a coin flip as an example is not because there is anything particularly special about coins per se, but that the procedure for flipping a coin is specifically designed to separate the possible states of the coin into two groups of states that can be easily and unambiguously distinguished from each other. We call those groups of states "heads up" and "tails up", but those are not the only possible states the coin can be in. A coin can be standing on edge, or spinning in the air, or bouncing on a hard surface. And each of those descriptions actually encompasses a vast number of possible distinct physical states. In fact, because the actual physical state of a coin (or anything else) involves the positions of atoms, the number of potential states might even be infinite. We don't actually know if space and time are continuous or discrete. If they are continuous, and if a physical process can cause two atoms to have positions that are perfectly correlated with each other, then the amount of information contained in that correlation would be infinite.
But here's the thing: even if this were possible, there is no way we could ever know. The reasons are complicated and have to do with quantum mechanics which we haven't gotten to yet, so for now you will just have to trust me when I tell you that it is a fundamental fact of physics that any time you make a measurement the result must be discrete, and so can contain only a finite amount of information.
But all this is beside the point. What actually matters is the claim often made by creationists that information cannot spontaneously emerge in nature. This is absolute rubbish. Information spontaneously emerges in nature all the time. Any time the states of two systems become causally correlated, information is created.
However, there is still a potential valid criticism here. It is possible that my definition of information as causal correlations between states is wrong, that it fails to adequately explain some observations. Specifically, it fails to explain caring, i.e. the fact that we humans seem to be able to distinguish between information that we care about from information that we don't care about in the same way that we can distinguish chairs from non-chairs. The information contained in books and newspapers and computers and DNA seems qualitatively different from the information contained in, say, a rock exposed to the elements.
There is no technical term to distinguish information in these two categories, and there is a reason for this: it turns out that there is no distinction. But in order to explain why, I first have to lay out the argument for why there might be a distinction, and to do that I need words that distinguish between them. In order not to prejudice the discussion, I'm simply going to call them type 1 and type 2. Type 1 information comprises the correlations that arise from straightforward physical processes like rain falling on a rock. Type 2 information is the kind that lives in human brains and DNA.
There are two main features that distinguish type 1 information from type 2. First, type 2 information has a lot more complexity and structure than type 1, and second, type 2 information seems to be intimately related to intentionality, i.e. to caring about things. (Note that here I am using the common definition of "intentionality", not the philosophical one.) When we humans write books or blog posts or invent printing presses and computers, we do it in service of some kind of goal or purpose. Likewise, the information in DNA seems to exist in service of the goal of producing life, without which the goals and purposes of human activities would obviously be moot. In other words, type 2 information seems to have something to do not just with being correlated with something, but with caring about something. And that distinction is ultimately rooted in our subjective experience of caring about things. That subjective experience requires explanation. One possible explanation is that caring is a fundamental component of objective reality, one that cannot be explained in terms of the mere movements of atoms. It's kind of like "chairness" except that we ultimately were able to explain chairs in terms of the movements of atoms. But explaining caring is going to be a much tougher nut to crack. We can see and touch chairs. We can't see and touch caring. We can't measure it. Caring is a purely subjective experience (much like "believing").
Now, I am going to advance a hypothesis which I believe to be false, but which is designed to be a steel-manning of what some religious people believe. I'm telegraphing this for two reasons. If you are not religious, this is probably going to sound crazy, and I agree with you, it is. But showing that it is crazy is actually not as easy as you might think. And if you are religious, I'd be interested to know if you think I've gotten this right, if this is a fair representation of (part of) what you believe. This hypothesis is intended to be a peer to the objective reality hypothesis. I'm going to call it the transcendent mind hypothesis, but I could as well have called it the God-is-love hypothesis:
Our subjective sensation of caring is a sensory experience that reflects a part of objective reality that lies beyond atoms, and therefore beyond what can be objectively measured, but which is nonetheless real.
In other words, our subjective sensation of caring about things is every bit as much a reflection of reality as our subjective sensation of seeing chairs. In both cases we are experiencing something real, something outside of ourselves. But in the case of chairs that something is made of atoms, and in the case of caring it is made of something else, something transcendent, and, most importantly, something that cares. Specifically, something that cares about us, something that cares about our caring. Something that could fairly be called God, even if it might diverge from standard theology in some details.
It is possible to show that this hypothesis is false, that it actually fails to account for all of our observations. But doing that is much, much harder than I think most of my fellow atheists appreciate. I think most atheists look at specific theologies, reject those (for very good reasons), but then end up throwing out the intentional baby with the theological bathwater. In order to really explain caring you have to explain consciousness itself, and that is not easy.
I'm going to fast-forward through at least a dozen chapters and tell you what the answer is going to turn out to be: it is not necessary to posit a transcendent mind to explain our subjective experience of caring about things. Caring can be explained in a purely materialistic way, purely as Atoms Doing Their Thing. The TL;DR is that we were created by Darwinian evolution, a process which "cares" about reproductive fitness in the same way that water "cares" about flowing downhill to reach the ocean. That process was physically reified here on earth as biological systems, and more specifically as a division of labor between genes encoded in DNA, and a phenotype that those genes produce. One of the things that genes "figured out" out how to do is to build brains, because it turns out that genes that build brains have a reproductive advantage (in certain environments) over genes that don't. And genes that build brains that care about things (like staying alive) have a reproductive advantage over genes that don't.
Like I said, that is vastly oversimplified. It sweeps huge swathes of complexity and nuance under the rug. But I didn't want to make you wait months before the Big Reveal. In the chapters that follow I promise I will lift up the rug and tidy up properly. But it's a big job. The path to enlightenment is not an easy one.
No comments:
Post a Comment