Life is Not a Computer: “Information” Inanity in Evolutionary Biology

John Wilkins, over at Evolving Thoughts has been making a pretty persuasive case that trying to apply concepts like “information” to biology is mostly a waste of time. In Part One, he covers what the concept of information actually is and why it’s such a clunker when it comes to describing biological systems. In Part Two, he defends a view of biology where information is just an abstract way of modeling things, not a property of those things itself.

Wilkins gets a bit technical, but a key take-away point is that defining exactly what we mean by “information” is extremely tricky: there are several complicated formulations of the concept, and they are generally incompatible with each other. We can’t possibly apply the idea of information to biological systems unless we nail down what we are looking for (most Intelligent Design arguments trip up at this very first step: they never specify what they mean by information in the first place).

Even once we’ve done that it’s still not clear exactly what we should be measuring. What is the information of biology? DNA code seems like an obvious candidate, but its clearly an incomplete one: so much of what biology is and how it works is contextual: it assumes certain things about what’s in the cells (including things that have their own evolving RNA code) and what environment those cells will find themselves in. Nor is DNA the only candidate: there are entire gene pools to consider (where arguably most of the information relevant to evolution is best captured), the functional complexity of an organism (which is not simply an expression DNA code), and so on.

Finally, there’s the problem that information may be the wrong model altogether, missing the subtleties that make biology distinctive from computer science. DNA, for instance, isn’t simply information read and then translated by a program. It’s a physical, tangible, and causal object, and its particular chemical properties and reactions are part of the process, rather than deviations or errors in the sole task of record keeping. Biology, the fantasies of creationists about “random processes” aside, is all about complex chains of causality, and those chains extend backwards from our DNA all the way to the origin of life.  Any meager measure of “information” is going to simply miss all of that.

With all those difficulties in defining and applying the idea of “information” to biology, it’s little wonder that many pseudo-scientists have gotten accustomed to talking about “information” in much the same way they talk about “energy”: as a sort of cipher for whatever unexplained bit of mysticism they happen to be pushing at the moment. Some woo-mesiters have even claimed that energy is information and vice-versa. This guy even cutely claims that Energy is “In-formation.” Get it?

No? Well, good: there’s nothing much there to get. There’s not even a hint of any understanding of information theory or energy in such babbling (other than “lots of really complex stuff!” and “glowing ball of lifepower!”)

And that’s the real problem with “information”-talk: because it’s such a difficult concept to apply to biological processes, the potential for error and confusion are always high. And if even biologists and information theorists aren’t sure what “information” should mean in terms of biology, you can bet that tossing the term out to laypeople is a recipe for intellectual disaster and creationist exploitation.


12 Responses to Life is Not a Computer: “Information” Inanity in Evolutionary Biology

  1. Mike says:

    Thanks for linking to this.

    I roll my eyes every time I see someone trying to apply Shannon’s information theory to DNA sequences. First off, just because he chose the term “information” does not mean it corresponds to the casual, intuitive usage of the word. Second, Shannon’s theory will give you quantitative properties of probability distributions and of an individual item in the context of a probability distribution of items. But the information in Shannon’s theory is not an inherent, invariant property of items (DNA sequences) themselves. (Kolmogorov complexity might be a better choice for a measure of information content, but still not that great)

    Talking about probability distributions over DNA sequences strikes me as silly. Like probabilistic arguments for the uniqueness of physical constants, life on Earth, etc., it is totally unconvincing. It assumes complete knowledge of all natural processes on this universe and all other possible universes.

  2. Bad says:

    It’s that last part that really drives me nuts: supposed calculations of probability of something like nature, where we simply have no clue at all how likely or unlikely the particular state of our universe is. Compared to what? As I’ve argued before: it’s basically compared to people’s very limited imaginations. For all we know, our universe could be so dramatically chaotic and unordered and poorly tuned compared to what we should have expected that we don’t even know what we’re missing. In fact, it could be so bad that we would have to posit an “Intelligent Ruiner” to account for it!

    I’ve seen even supposedly mathematically inclined people try to run supposed probability calculations when they themselves admit that they have no idea what the odds are, and only one example to go on for any one of the variables. Then why/how the heck are you talking about probability in the first place?

    Probability only makes sense in a defined context: the probability of finding a particular character in a string. The probability of rolling a particular number on a die with a known number of sides and inscribed values… or at least a random sampling of rolls over time. You gotta have something to start with in order to be surprised by the likelihood or unlikelihood of the result!

  3. Mike says:

    … something like nature, where we simply have no clue at all how likely or unlikely the particular state of our universe is.

    “As far as the laws of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality.”
    — Albert Einstein

  4. hughvic says:


    Even I, a caustic critic of yours, feel compelled to admit that this is a superb essay. There is too much pat functionalism in biology, as in the social sciences. Functionalism, good. Too much, not. Trouble is that analysts get caught up in their anthropomorphic metaphors when they want every biological attribute, for example, to be an instrument of some purpose they happen to recognize or find plausible. Also, an extreme version of the information metaphor is the cybernetic one, which these days sometimes seems to have overtaken our thinking and even our self-perception and perception of the outside world. Very few artists are capable of full-blown allegory; the rest of us run away with our metaphors, or let them run away with us. We end up in the realm of the categorical error.

  5. Bad says:

    Yes, and the problem with biology, particularly down at the cellular level, is simply that nothing about it really mirrors anything from our daily experience. Analogies are a very common shortcut to understanding things quickly: but in the case of biochemistry, there really isn’t a shortcut that won’t ultimately make for, as it were, a long delay. You really do have to do all the very hard and heavy lifting of figuring out this strange alien world from scratch in order to get a good grasp of what’s going on. And of course, we have a long way to go in understanding it fully in any case.

    If biology is all information, then by the same token pretty much everything is, and the word ceases to have distinctive and distinguishing meaning, and hence little power to inform about the complex specifics.

  6. hughvic says:

    I think this is brilliant, and I would wish you to keep writing about its various facets.

  7. slpage says:

    Mike writes:

    “I roll my eyes every time I see someone trying to apply Shannon’s information theory to DNA sequences. First off, just because he chose the term “information” does not mean it corresponds to the casual, intuitive usage of the word.”

    INterestingly, I have recently been engaging the latest in a long line of creationists with computer programming/engineering backgrounds on this very subject. I asjed him repeatedly to supply a defiition of information. Finally, he replies:

    “Generally speaking you can say that information is a pattern that has meaning.”

    As are most creationists with engineering backgrounds, he is exceptionally condescending and arrogant, but also very ignorant of the relevant science. For example, he insists that the human genome is not nearly large enough to “code” for the “complexity” of the human body. I asked him how complex the human body was. No answer. I asked him how big should the genome be. No answer.
    These folks are incredible.

  8. hughvic says:

    I never did like that LOL acronym, but I gotta admit I am indeed laughing —see? had to stop a moment—out loud right now. You go, slpage!

    Yes, the word “information”, in the Life Sciences, is proper (necessarily delimited and specific) nomenclature. You’re right: the problem is definition creep; in particular, creep into the prosaic. Something tells me that Bad could write a book’s worth of very fine essays on quicksand biological metaphors, into which both biologists and lay critics sometimes sink, the latter obviously more often than the former with some metaphors, and with others (e.g. “function”), vice versa.

  9. insomniac says:

    Howdy folks,

    Hope i’m not too late to jump in on this one. Lest you forget that information theory and cybernetics were both developed from studying biological systems in the first place.

    Here’s a bit from the intro of my book in progress, LifeOS: exploring the system that executes DNA…

    “All creatures alive today, plus the remains of all living things that have gone before, all the organic compounds, all fossils, all fossil fuels, all the biomass accumulated by this planet over billions of years, exists because, information coded into DNA was accessed, read and acted upon by a cell. Before any one of those cells could grow, before any living tissue could be manufactured, before any polypeptide chains could be assembled, before anything could happen in ANY cell, DNA information had to be processed. Information processing is the very first act of Life.

    Why is this an important distinction? Because information processing involves a set of concepts that are independent of the processing method being used. For one, there must exist a consistent set of rules or protocols that govern information processing within the system. This is called an operating system, or OS.

    All living organisms follow the same rules for accessing and reading DNA code. These universal rules infer the existence of a biological operating system. I call it LifeOS.”

    Where have i gone wrong. Let me have it. :-)


  10. Hugo says:

    God, you are diabolically clever, jim. Shit, where’d you come from.

    I’ll let you “have it” when I stop laughing.

    Hugh de Ste-Victoire

  11. insomniac says:

    Howdy Bad,

    The details of your argument aside, your efforts to stem the tide of paradigm shift seem to me to be doomed to failure. History shows us the we humans tend to describe our universe in terms of our technology. It is only natural for our world view to shift to accommodate our new understanding of information processing. For those of us steeped in a mechanical universe, adding information as a controlling factor is an easy step. For those who have been raised in the information age, dropping the mechanical/physical to a subordinate role to information is also an easy shift. But hanging on to the old idea that information is a non-factor hasn’t got much of a future.

    Besides, once you get into it, the new model is far superior to the old one.


  12. My brother would be amazed this website. We were just speaking about this. lol

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: