John Crace, at the Guardian, has a nice series of “digesteds“. I’m no John Crace, but I’m going have a go at digest of Dembski’s article at ENV, Conservation of Information Made Simple, because it seems to me that Dembski could use an editor with a big blue pencil.
Conservation of Information Made Simple by William A. Dembski – digested read:
The structure and laws of the physical world permit intelligent multicellular life. Therefore the universe itself, can be viewed as the solution to the problem of making life possible.
When we talk of searching for solutions, we are talking about the probability finding one of a small relatively small set of solutions at the end of one of many different possible searches. Clearly, the probability of finding a solution will increase if we have information as to where to look. But finding the information as to where to look may be as difficult as finding the solution.
Bacteria sometimes evolve to be capable of utilizing a new food source, for example, nylon-digesting bacteria. So we can say they have gained information. But it is not new information, it has come from the environment – in this case, nylon-filled pools.
You can think about probabilities as information costs. A high probability has low information, but a low probability has high information. Information is expensive – we pay fees to agencies that will increase the probability that we will get what we want. Or you could buy all the lottery tickets, and increase the probability of winning to 1, but it would have cost more than the price of the jackpot.
We call this “conservation” of information because that the best we can do is break even. But sometimes we do worse, and the cost of making the search easier is higher than the cost of the original hard search. After all, some search strategies may that steer us away from solutions. So information can only be lost, not gained, when we try to improve our search strategy.
The No Free Lunch (NFL) theorems say that no search is better than blind search, when the target is unspecified. Conservation of information says that some searches are better than others, but only because of the additional information that the search is employing to boost its performance.
It also suggests that an object and the search for an object are not separate. We only become aware of an object when aspects of it become salient: “to be perceived is to be an object of search.” It therefore follows that to be is to be an object of search. Also, search can itself be an object of search.
This is why I regard the multiverse as incoherent: the known physical universe is knowable because it is searchable. The multiverse is unsearchable, therefore unreal.
Conservation of information can be formulated as: raising the probability of success of a search does nothing to make attaining the target easier, and may in fact make it more difficult, once the informational costs involved in raising the probability of success are taken into account.
When Dawkins wrote WEASEL, he included the solution to the search into the his fitness function, so there was no net increase in information when the search found the solution. AVIDA, Tierra and ev do the same thing.
Ken Miller says that to increase in biological information over the course of evolution you only need selection, replication, and mutation. But computer simulations that use selection, replication, and mutation don’t solve interesting problems, or produce interesting patterns, unless we add information about what we want to achieve.
The fitness landscape provides an evolutionary process with information. Solutions will only be found if the fitness landscapes is smooth, doesn’t isolate local optima, and rewards complexity. So where do such fitness landscapes come from? The only answer, apart from an Intelligent Designer, would seem to be the environment.
But where did the environment get that information? Evolution is supposed to explain subsequent complexity in terms of prior simplicity, but conservation of information says that there never was a state of primordial simplicity — the information had to be there from the start. In the absence of intelligent input, the complex biological information we observe now must have been present in the universe in some form or fashion, since the Big Bang. But the heat and density of Big Bang rules out any life form so where was the information needed for the emergence and development of life on planet Earth?
Evolutionists say that the way the environment inputs information into biological systems is by a gradual accumulation of information as natural selection locks in on small advantages, each of which can arise by chance without intelligent input. But Intelligent Design presents a Two-Pronged Challenge. On the one hand the evidence for common ancestry is weak. On the other hand, even if common ancestry is true, conservation of information is compelling evidence for intelligence design.
It would actually be quite a remarkable property of nature if fitness across biological configuration space were so distributed that advantages could be accumulated gradually by a Darwinian process. Frankly, I don’t see the evidence for this.
If biological evolution proceeds by a gradual accrual of functional advantages, instead of finding itself deadlocked on isolated islands of function surrounded by vast seas of non-function, then the fitness landscape over biological configuration space has to be very special indeed.
If evolution is so tightly constrained, then the conditions that allow evolution to act effectively in producing the complexity and diversity of life is but a tiny subset of all possible configuration spaces, and therefore itself a small-probability target.
Evolution can certainly make improbable outcomes probable, but it requires a very special kind of fitness landscape that is itself improbable. Yes, you can reach the top of Mount Improbable, but the tools that enable you to find a gradual ascent up the mountain are as improbably acquired as simply scaling it in one fell swoop. This is the lesson of conservation of information.
So what is the source of information in nature that allows targets to be successfully searched? The answer will by now be obvious: from intelligence. If intelligence is real and has real causal powers, it can do more than merely redistribute information — it can also create it. That is the defining property of intelligence.
Also, Descartes.
The digested digested:
I was wrong about organisms being improbable, given a fitness landscape searchable by evolutionary processes. I should have said that a fitness landscape searchable by evolutionary processes was improbable. I’d still be wrong.
As long as this “argument” keeps cycling on “information,” it will never converge. Dembski is simply following in the tradition of misdirection initiated by Henry Morris and Duane Gish.
ID/creationists do not understand the concept of a multiverse. They constantly abuse “fine tuning”. They constantly make use of the Lottery Winner Fallacy. They don’t understand the essence of the second law of thermodynamics. They don’t understand that matter condenses at all levels of complexity.
I read through Dembski’s “Conservation of Information Made Simple,” and all I kept seeing were mischaracterizations and misrepresentations of fundamental scientific concepts that have well-established roots in basic physics and chemistry.
Biologists generally have a pretty good intuitive understanding of fundamental laws of chemistry and physics because biologists work with concepts that comport with those fundamental laws. Variation and natural selection are observed processes that occur in nature because of the laws of chemistry and physics. Biologists are not mistaken in what they observe in nature; so using these concepts in computer models is perfectly justified.
Furthermore, Dembski doesn’t appear to recognize that there are literally billions of representations of living organisms that exist and have existed. Sample spaces are being well-sampled, and many viable forms have been produced. Focusing on specific examples as a search for the “needle in the haystack” is engaging in the Lottery Winner’s Fallacy.
Evolutionary algorithms make use of concepts that are variations on minimizations of potential energy concepts; and those concepts find their roots in the second law of thermodynamics and the fact that matter condenses at all levels of complexity. Huge sample spaces are being sampled constantly in literally billions upon billions of parallel processes. All kinds of things are being generated; and some of them we call life.
No ID/creationist I have ever observed over the entire 50+ year history of ID/creationism EVER gets the fundamental concepts in physics, chemistry, and biology right. They just refuse to be corrected and continue to engage in misrepresentation and misdirection instead. Dembski appears to have mastered the tactic of trying to provoke debates instead of actually learning the science.
Bacteria can buy all the tickets, and most invention has been done by bacteria.
Well, hidden in all the bluster, it looks to me as though he is conceding a pretty crucial point – that evolution works just fine to produce “complexity”.
So he’s shifted the locus of the designer back to the creation of a smooth fitness landscape. Here is the money quote (undigested):
What is presented as three separate qualities that the fitness landscape has to have is actually only one. A smooth fitness landscape is one that doesn’t isolate local optima, and rewards are intrinsic to the concept of a fitness function.
(Although he’s wrong that complexity has to be rewarded – complexity, I would argue, is epiphenomenal, because in an incremental system, retrofitting is going to be more probable than back-to-the-drawing-board, even when simpler would be better, so easier, if you want a snake, to keep the developmental program for a tetropod and cancel the legs, than undo all the leg-making gear at the potential cost of messing up something else.)
And a high dimensioned fitness landscape will tend to be smooth (lots of potential directions that may have ramps), while a system in which variant offspring are far more likely to be like their parents than not, will also be smooth.
So, given self-replicators that produce similar, but not always identical, offspring (a self-replicator whose offspring were radically different wouldn’t actually be a self-replicator), with heritable variance in potentially fecundity-promoting features, you will have a fitness lanscape that is rich and smooth. Oh, also given an actual landscape as well.
So his argument boils down to: but self-replicators are unlikely.
And of Szostak fills that gap, it will be back to Big Bang and the parameters that led to carbon. (His argument about Big Bang is illogical it seems to me – why should the fact that there were no life forms at Big Bang mean that there was no information at Big Bang? If carbon is the thing you need, maybe all that the ID needed to supply were parameters that would lead to carbon, and bingo, all the information is there at the start.
Or maybe carbon isn’t that improbable. In which case nor is Life. In which case there is no Complexity to explain in the first place.
Oh dear, did this guy really get a PhD in philosophy?
The structure of modern houses is such that they permit cockroaches to find entry and to live. Therefore the houses themselves can be viewed as solutions to the problem of constructing places for cockroaches to live.
</sarcasm>
After you concede that evolution is possible you are left with one more horror for the ID advocates: contingency. I think Gould posed the question as what would happen if you rewound the tap and started over.
Would you get the same kinds of things? Is dinosaurness inevitable, or birdness? Would you always get a flagellum?
I think ID people hate this more than they hate the possibility that evolution happened.
I’m sure the point has been made before, but if Complexity is an improbable outcome (plus some other criteria), and it turns out that complexity is also inevitable, is it no longer complex?
In which case, is something only complex if it isn’t inevitable given (to echo Demski on Medawar) primary axioms?
If so, how does Dembski know that Life wasn’t inevitable given primordial conditions? And therefore not complex? Or, at any rate, containing no more information than present in primordia?
In which case, does his entire argument not simply boil down to Argument From First Cause?
Why bother with the math?
This is precisely where ID/creationists misrepresent the concept of “multiverse” and abuse “fine tuning.” Their “argument’ presumes that varying the “constants” of the universe produces nothing; that our universe is the only possible universe that generates condensed matter and, ultimately, life. It presumes that other universes have to generate life as it exists in our universe; in other words, the Lottery Winner Fallacy again.
However, many combinations of “cosmological constants” produce universes. And the “lifetimes” of these universes are not the issue. They can have their own “periodic tables” with their own “elements,” some of which take on the role of “carbon.”
Condensations into living organisms only have to be short relative to the “time span” of the given universe. To a sentient living organism in that universe, their universe would seem much older than life; the concept of time is relative to the events taking place in that universe. Once enough complexity has built up, life emerges; with the “elements” in the “periodic table” of that universe forming analogous roles to the elements of the periodic table in our universe.
Multiverses don’t imply that other universes have to have exactly the same periodic table and the same life time as our universe.
Which would bring us back to Wolfram! Are the complex sequences (which I think satisfy Dembski’s CSI criteria – they are complex, and of course specified) Wolfram gets from Rule 30 or whichever it is, not complex because we know what the information is that went into them? Or not complex because they are high probability, given the rule?
In other words, can a highly compressible sequence (compressible to a simple rule) i.e. specified, never be complex, no matter how many bits it has, because we can always reduce the bits by compressing it?
Or rather, is CSI simply a measure of our current state of knowledge regarding the causal chain, not an objective measure at all?
Ah, I have been here before:
Lizzie wrote:
I have read a fair number of Dr. Dembski’s monographs and writings, although I have not read the book “No Free Lunch”. However, I have read his piece:
Intelligent Design as a Theory of Information
several times, and I agreed with it (after considerable reflection), on first reading, up until the final section entitled “The Law of Conservation of Information”. On re-reading it, I still find that up until that point in the article, Dr. Dembski makes perfect and elegant sense. CSI has to be the result of Actualisation-Exclusion-Specification, and Specification must be the result of choice. To quote Dr Dembski:
However to my mind, Dr. Dembski at that point takes a leap.
He claims:
His reasoning appears to be as follows: he states:
(citing Monod), and proceeds to rule out both chance and necessity as a source of information. He then states:
In other words, he defines chance as “blind, purposeless contingency” – and ascribes “intelligent causation” to any other kind. Now, I’d still agree with him, using his own operational definition of “intelligence”. Where I disagree with him is that such “intelligence” cannot be “natural”.
He writes:
This is where Dr Dembski’s reasoning is not clear to me. Indeed, earlier in the paper, Dr Dembski gives an example of rats learning maze, and notes:
But trial and error is the method by which rats learn to find their way through a maze – or at least, if it is not, there is no way of telling that it is not. All we observe is that the rat has learned the maze. And Dr Dembski tells us, correctly of course, that if the rat makes the correct sequence of left and right turns, we can infer from that the fact that the pattern of its behaviour exhibits CSI that it was produced by an intelligent agent, namely the rat. In his own next words:
So it would appear that Dr. Dembski believes that learned behaviour can be recognized by its CSI, and inferred to be intelligent behaviour. Which is fine. But we know that learning can proceed by trial-and-error – indeed many of the cognitive tasks we use in cognitive psychology can only be solved by trial and error. So it does not follow, to my mind, that a pattern that is arrived at by trial and error cannot generate CSI. The rat demonstrates that it can.
So I took a closer look at Dr. Dembski’s analysis of chance and necessity. Dr Dembski claims that:
He rules out chance as a source of CSI, as of course we must do. Chance, is, after all, the null hypothesis in any signal detection test. But he also rules out necessity. He does so by reasoning that if A must lead to B, observing B tells us no more about A than we know already from A. But consider this: if we have a “natural” choice maker, such as a perfect filter, or sieve, and a supply of particles of varying size, then we will find ourselves with a sorted arrangement of particles that cannot have arisen by chance – the pattern of the sorted particles exhibits CSI. We can infer the rule that generated the pattern:particles smaller than a certain threshold pass, but larger particles are retained. I assume that Dr. Dembski would not want to call such a perfect filter “intelligent” – although it clearly has “the power and capacity to choose between options”: it chooses the large particles and releases the small ones. And I would agree that a sieve is not what we would normally call intelligent, although it fulfills Dr. Dembski’s operational definition. So why would Dr Dembski not infer an intelligent agent from something a pattern that had resulted from a natural sieve?
Well, he tells us that the pattern generated by the sieve cannot exhibit CSI because no new information has been added. And it is true that if we knew the precise mesh size of the sieve in advance, and the sizes of every particle, the piles of particles wouldn’t tell us anything new about what pattern would be generated by the sieve. But in that case would a sieve with randomly fluctuating mesh size produce piles of particles that exhibited CSI? And could we then infer an intelligent sieve? Well, clearly not. All we’d have is an unreliable sieve instead of a perfect one. An unreliable sieve is not more intelligent than a perfect sieve.
In other words, the pattern of sorted sand, which appears to me to exhibit CSI, cannot, according to Dr. Dembski have CSI, simply because we know everything about the sieve. His argument therefore appears to be not that, as he claims, that we can infer intelligence from a pattern that appears to exhibit CSI, but from the degree to which we have less-than-perfect knowledge about about the mechanism that created the pattern. Thus we can infer that the rat is intelligent, not from the CSI generated by its behaviour, but because we do not know everything about what determined the rat’s choices. Conversely, we infer that the sieve is not intelligent, not because the pattern it generates does not have CSI (it does), but because we can know, in principle, everything about the sieve.
This is the problem I have with Dr. Dembski’s analysis. I agree with him that CSI is detectable. I agree with him that if is detectable we can infer that it was generated by something with “a power and capacity to choose between options”. I do not agree with him that that we can distinguish between a “natural” choice-maker, like a sieve, and an “intelligent” choice maker, like a rat, by observing the differences between the patterns they generate. Both will generate patterns that exhibit CSI. But if intelligence is to be inferred from the amount of new information contained in the pattern, this quantity will depend not simply on the pattern, but on what we know about the factors that determine the choice-making of the choice-making agent. The more we know about the choice algorithm (whether rat or sieve), the less new information will be gained by examining the pattern. Once we know everything about the rat’s brain, will the rat cease to be intelligent?
No, because, in my view, is that there is no difference between the two agents. The amount of new information (in Dr. Dembski’s terms) contained in the pattern generated by a trial-and-error learning algorithm in a computer only differs from the trial-and-error learning process in a rat because we know the algorithm – because we wrote it. And the pattern produced by a “natural” filter only differs from the pattern produced by the rat in that the rules that govern the pattern are more amenable to inference by a diligent scientist.
If you calculate the odds against all the contingencies necessary to produce my birth, my existence is inconceivable, so to speak.
Without knowing how physical constants originate, what is the point of discussing the odds?
Assuming they are arbitrary, what is the point of calculating the odds against a fait accompli? Some psychological need for certainty, for a cosmic father figure, some assurance that death isn’t real or final?
Science, for sure.
In thinking about the rat and maze example it occurs to me that what many ID advocates think is necessary is some uber-knowledge, some agency that knows, prior to choice, the probable outcome of the choice. In other words, and agency that is not blind to consequences.
The engineering projects they talk about are a bit like this. Architects and engineers have rules for assembling things so that the final product will work without trial and error. It’s true that one does not build skyscrapers by trial and error. At least not since the Middle Ages.
But of course living things also assemble themselves without trial and error. That’s what development is. Using known rules in known ways to build structures.
But it isn’t what evolution is. No one has proposed any way for evolution to know what is needed or how to make it. Evolution is Braille. Evolution has acquired a big bag of tricks — something emphasized by James A. Shapiro. A bag of tricks for feeling its way in the functional landscape, but it doesn’t see the future, and it doesn’t know the emergent properties of novel sequences.
I think that Dembski equivocates between the frequentist approach of thinking of a probability distribution as a normalised frequency distribution, and the Bayesians approach of regarding probability as a measure of uncertainty.
If probability is uncertainty, then something as deterministic as the outcome of a mandelbrot equation, or cellular automata, the uncertainty depends on what we know, and if we are running a cellular automaton for the first time, we will be pretty uncertain – we might have a shot at predicting the next iteration on the basis of past patterns, but we won’t know for sure what’s coming up until it happens. But if probability is a measure of frequency, and we position ourselves as someone who has run the program over and over, and computes the outcome as a probability of distribution with a single value, i.e. the outcome has a probability of one.
When Dembski talks about CSI, he thinks of probability as a frequentist would (despite lacking data on which to base his frequency distribution). But when he talks about the Law of Conservation of Information, he thinks of probability as a Bayesian would – that information regarding the search buys us greater certainty as to the outcome.
It seems to me that it gets progressively worse if one simply decides to increase “specificity.”
Mr. X being the only one out of a million who won the lottery is highly specified.
But Mr. X, living at address Y, on street Z, that just happens to have a specified arrangement of blades of grass on his lawn, with Mr. X wearing specified items of clothing with a specified number of coins in his specified pocket, and driving a specified car with a specified license plate number with a specified number of gallons of gas in the tank, etc., is even more specified; hence less probable.
No matter how specified one makes an outcome, one can always increase specificity by folding in more detail.
So where does it end? Complex SPECIFIED Information is the central theme of Dembski’s “argument. Choosing to always look at outcomes in this way allows one to make anything so improbable that no kind of search can land on it. Reaching the “500 threshold” is easy; just fold in more specificity.
Well, he tries to equate “specificity” with “compressibility”. But that isn’t consistent with his Conservation of Information argument because if a pattern has Specified Complexity pattern it is supposed to have both high Shannon information (unlikely to be found by blind search) and be highly compressible (can be found easily if you have the compressed code). But according to the Law of Conservation of Information, the information content of the the compressed code contains no less, and possibly more information than the unpacked version. And if the compressed code is dead simple (which it must be, if it is compressed), it will have low Shannon information. So there’s equivocation with two meanings of information there.
If he wants to say that CSI of a pattern is the probability of finding the most compressed version of the pattern by any non-intelligent search strategy, then he needs to redo the definition of CSI, because right now, that isn’t in there. In the Specification paper, He says that what you need to compute the Specified Complexity of a given pattern is the proportion of patterns of the degree of Shannon complexity of the given pattern that can be compressed as, or more, simply than the observed pattern; which isn’t the the same as the probability of finding a compressed formula that gives a pattern at least as Shannon complex as the observed pattern by blind search.
But it’s all moot, because given self-replicators that replicate with moderate fidelity, and heritable variation in reproductive success, you will get a smooth fitness landscape on which are situated complex solutions to breeding successfully challenging environments.
So the answer to hard it is to find such information is a simply practical one: can self-replicators that replicate with moderate fidelity and heritable variation in reproductive success emerge spontaneously from chemistry?
If so, it’s back to Big Bang for Dembski.
“Bacteria sometimes evolve to be capable of utilizing a new food source, for example, nylon-digesting bacteria. So we can say they have gained information. But it is not new information, it has come from the environment – in this case, nylon-filled pools.”
So, according to Dembski, the feature of the environment that an organism has adapted to, and the feature of the organism that is that adaptation are the same “information”? I’d like to see him spell out that information…
Also, by that logic, there is no such thing as new information, which he kind-of admits here:
“[…] conservation of information says that there never was a state of primordial simplicity — the information had to be there from the start.”
Of course, if there is no such thing as new information arising at any time during evolution, then this culminating question of his:
“So what is the source of information in nature that allows targets to be successfully searched?”
…is pointless because, according to him, there is no new information that needs to be found!
Well, you might want to check the primary source! But I think my precis is OK. I wanted to do it, as an exercise to try to find the core of what he was saying. The original is full of derails about WEASEL (yawn) and evolution meanies.
You can’t make carbon with the properties it has without making the rest of the periodic table have the properties they have. The fundamental forces and particles are pretty fundamental. We get a menagerie of elements due to both valency and the crowding and shielding of electron shells, generating complex emergent variation from the same handful of variables. The parameters that make Life ‘life-y’ are the same as those that make rocks rocky, planets planety and water wet. Which co-dependence the die-hard will still see as a brilliant act of fine tuning.
But I’m not convinced the parameters are variable across multiverses (or that the latter exist) either.
Yet it seems that Mike Elzinga is guilty of indulging in misrepresentations and misdirection himself.
To be sure, the MV is an unproven concept, hardly qualifying it as a satisfactory response. The idea is to provide an answer that is more compelling than FT.
As well, variation and NS do not happen because of the laws of physics. You have yet to show any chemical formula (or any formulas for that matter) that account for variation and selection. You simply assume it; another misdirection on your part.
Finally you’re continual portrayal of life as a mundane property of matter is your most obvious misconception and misrepresentation of what we observe.
Asking students to scale up energies as a way to instill a WOW factor is one of your other strategies to forego a rational assessment of our observations of living systems exhibiting intelligent properties.
Steve, do you think my digest is a fair summary of Dembski’s article?
Excuse my obtuseness, but what are the MV and the MT?
I don’t think Steve is interested in anything you have said. He has been a troll taunting me and others on Panda’s Thumb. He has a huge backlog of failed tests of his knowledge about scientific concepts.
His comment is nothing more than a taunt; directed at me in particular. I won’t help him pull the comments off topic.
You may be giving too much benefit of the doubt. In principle, everything is a pattern. Even random distributions are patterns, and we see constellations. So how can we distinguish between which patterns show CSI and which do not? How can we tell that an unsorted pile of sand is NOT a careful and deliberate cofiguration, for some unknown purpose?
Again, think of the space alien that leves something on your doorstep during the night. With no background knowledge whatever, you have no way of knowing if it’s an alien device, or some random part of an alien landscape, or the alien’s lunch, or even the alien itself.
So it seems to me that the identification of CSI, in the first place, requires considerable background knowledge about the history of the object. ID proponents, IMO falsely, compare their “design identification” to SETI or forensics or archaeology. But in each of these cases, a background history is either known or assumed in enough detail to be able to tell signal from noise.
The sorted sand “seems to you to exhibit CSI” because you have considerable background knowledge about the processes that move sand around. In Dembski’s terms, you are “smuggling in” the information necessary to identify CSI.
Excuse me butting in. I’m guessing you’re asking about MT and FT, Steve’s unusual abbreviations in that comment ? And if so, I’m guessing Steve means MultiVerse and FineTuning. It’s a plausible guess, in that xe disparages MV [Multiverse] as a “satisfactory response” to the “compelling” claim of FT [Fine Tuning]..
Ah, thanks.
Well, first, what is the question that you think that “multiverse” might be the answer to?
And second, would you agree that the observable universe must be a very small subset of the entire universe?
Well, yes, they do. Variation occurs, mostly at any rate, at the level of chemistry which is of course governed by physics – variant genotypes are variations in of a molecule, a chemical, and that molecule, like others involved in live, behaves as molecules do. There’s nothing in biochemistry that defies the laws of physics and chemistry! As for selection – well, sure it obeys the laws of physics. It’s stuff that happens – gravity, for instance, or terminal velocity, will be a factor that affects how fast a given phenotype of a tree-dwelling animal falls to the ground, and therefore whether it survives when it gets there. That’s a simple example, but we don’t have to invoke anything outside the laws of nature to explain how selection happens. We can even watch it happening in real time
Well that’s the issue, isn’t it? In what way do you think that life is something other than a “mundane” property of the matter of which it is made? Do you think that something non-material is required to make an organism function? Because I do not see anything that we observe that would make us think otherwise. Can you explain what you have in mind?
“Mundane” is a demonizing caricature, borrowed from sectarians railing against secular ideas, that Steve has chosen to use in his taunts. Neither I nor anyone else has ever characterized life as a “mundane” property of matter.
To understand Steve P’s trolling history, one must go back to exchanges that took place over at Panda’s Thumb.
Thanks Mike, but I shall ignore history, for now, and wait to see whether Steve responds.
William, at UD (why they can’t post their responses over here, I really don’t know – is William worried about a counter-argument?
William, I just accounted for the can-opener! – chemistry. What you now want me to account for is a universe in which can-openers are found!
Which is my point – Dembski’s is an argument from metaphysics, not an argument probability, as he claims it to be.
I’m supposing something for which we already have substantial evidence, as an alternative supposition of something for which we have none, and suggesting that, contra-Dembski, intelligent agents are not the only conceivable origin of smooth, high dimensioned fitness landscapes – that, to use Dembski’s term, “Necessity” is in principle capable of generating such a fitness landscape.
Maybe read my whole OP (I know it is long, but shorter than Dembski’s I think), not just cherry-pick?
Lizzie,
multi-verse and fine tuning
Lizzie,
Elzinga is only interested in demonizing anyone he decides is a creationist. His posts are frequently tied to his pet caricature of those that don’t accept Darwinian evolution as having a myriad assortment of social/political agendas. It never dawned on him that someone could reject Darwinian evolution on logical grounds.
He also displays a penchant for using the table turning rhetorical tactic as he is doing here, claiming my posts directed at him amount to taunts. Yet, he is always there getting his digs against anyone whom he deems creationists and plays indignant when that poster responds in kind.
Lizzie,
How does a can opener = chemistry? A can opener is a designed object.
This is one of the may logical flaws proponents of design detect when discussing evolution with those that reject design in nature.
They always seem to reply to a problem by referencing a concept that itself needs explaining.
This is a case in point.
Lizzie, I would like to pick your brain and ask you if you can describe what happens in nature without reference to any designed object; meaning can you describe what chemistry has done, in the manner of the can-opener but purely in the language of chemical formulas; ie this reaction causes such and such molecules to change their charge, resulting in a subsequent reaction, which increases the energy of the system, thereby causing the specific change in question…..etc etc?
Why is there this choice of using analogy instead of the formulaic language of chemistry, which your side is always claiming is the causal base?
Elizabeth says: William, I just accounted for the can-opener! – chemistry. What you now want me to account for is a universe in which can-openers are found!
You have not “accounted for” anything. You “suppose” a can opener that is not known to exist in the first place (that unguided chemistry is plausibly capable of creating a self-replicating system, variation, and a convenient-to-targets fitness landscapes), which you admit may be “impossible”. When called upon to pay for your search for a search with a sufficient cause, you go outside of the universe and call upon magic to pay your debt.
You then try to say that Dembski, or I, have taken the argument “outside of the universe” when we have not – you have. We have a sufficient causal agency inside they universe; we do not have to “suppose” intelligent design. You have to “suppose” that something unknown is going to pay the price for your search.
This is why ID is the best current explanation; it doesn’t rely upon the magic of an unknown commodity that may like “outside of the universe”.
Ironic, isn’t it?
William J. Murray,
you go outside of the universe and call upon magic to pay your debt.
And the Designer?
Lizzie: Well, first, what is the question that you think that “multiverse” might be the answer to?
Actually, it is Elzinga that seeks to use the multiverse as an answer to fine-tuning. The the MV is unproven making it a weak rebuttal.
Lizzie: And second, would you agree that the observable universe must be a very small subset of the entire universe?
Absolutely. But that which cannot be observed could not be characterized as a separate universe(s) now, could it? Further, if you believe much if not most of the universe is un-observable, yet that un-observable part universe has an impact, then understanding that impact in terms of information (un-observable) is not nearly as far-fetched as most posters on this and other pro Darwinian evolution blogs want to portray it.
Lizzie: Well, yes, they do.Variation occurs, mostly at any rate, at the level of chemistry which is of course governed by physics – variant genotypes are variations in of a molecule, a chemical, and that molecule, like others involved in live, behaves as molecules do.There’s nothing in biochemistry that defies the laws of physics and chemistry!As for selection – well, sure it obeys the laws of physics.It’s stuff that happens – gravity, for instance, or terminal velocity, will be a factor that affects how fast a given phenotype of a tree-dwelling animal falls to the ground, and therefore whether it survives when it gets there.That’s a simple example, but we don’t have to invoke anything outside the laws of nature to explain how selection happens.We can even watch it happening in real time
If that were true, we would have the chemical formulas for each and every reaction and be able to show which reactions had to be first, which logically came second and how the second must proceed from the first, etc, etc. Yet AFAIK, no one has provided the chemical formula for the advent of cellular organelles, DNA detection and repair, the replication process, etc etc. It is just assumed that there could be nothing else but chemistry involved.
But this is precisely what is in question. And it is not ‘creationists’ that see the elephant, but thinking scientists that understand the impasse we are at; that a key element is missing that is preventing our bio-chemical enlightenment. The emerging consensus is that information as a separate entity is a viable candidate to fill the gap in our understanding. Nothing ‘creationistic’ about that, just intelligent people plowing the field with new equipment, hoping for a better crop next year.
Lizzie: Well that’s the issue, isn’t it? In what way do you think that life is something other than a “mundane” property of the matter of which it is made?Do you think that something non-material is required to make an organism function?Because I do not see anything that we observe that would make us think otherwise.Can you explain what you have in mind?
If life was in fact a mundane property of matter, we would see it all over the universe, not necessarily carbon based, but nevertheless, we would see animated matter. Yet we don’t. Therefore, the logical conclusion is that life is a special property of matter. So we need to understand what is that special property that differentiates living systems from non-living systems. Obviously chemistry is not the answer, as inanimate matter also displays chemical reactions. So an appeal to chemistry is inadequate.
By the way, it cannot yet be determined if information is non-material or not. What we can say is that it is un-observable, like much of the universe. I will be quite honest that many propoents of design do in fact see information as non-material. And it could be so. But the physical characteristics (or lack thereof) of information does not have an impact of our ability to understand it as a causal factor. In the same vein that we understand gravity, we can understand information; from its effects.
Regarding needing something other than chemistry to understand how an organism works, I would say yes we in fact need something more for the simple fact that we have yet to understand the how the hierarchical nature of organism’s multiple systems came together and interact from a chemical POV. If we really did understand them, we would not be having this conversation.
Thanks, yes.
Oh, I think he accepts that someone may think they have logical grounds. It’s just that he, like I, have yet to find those logical grounds anything but flawed. So let’s talk about the logic, rather than lick old sores?
Well, I do try to move posts that are largely taunts, and I certainly discourage them. And I usually move discussions about taunts to the sandbox, but things are quite busy right now. I’m hoping to move us to a forum format shortly which will make things easier, I think. In the mean time, let’s all just try to stay focussed on the argument, not the personalities, or perceived motivations.
Well, William was riffing off Dembski’s metaphor. In my view, Dembski (and William) are accusing people making the argument I am making of saying “suppose a can-opener”. I’m not, I’m saying “suppose chemistry”. And, as you say, while a can-opener is designed (we know that for sure, we’ve seen people designing them) it is not self-evident that chemistry was designed.
And the larger point is that Dembski used to claim that Life must be designed, because evolution couldn’t add information. Now he seems to be saying that the information must be present in the Chemistry, not added by the Designer during evolution. Or possibly added at OOL, though he doesn’t say that specifically. Or possibly at Big Bang. So his argument has changed fairly radically, it would seem.
Well, I think there has been a bit of confusion here, but yes, Dembski seems to have moved the argument back to “well, explain chemistry”. Which is progress I think, because explaining the diversity and complexity of life, given self-replication is easy to do; explaining self-replication given chemistry, is harder, but may well be solved. Explain chemistry, given Big Bang, is harder still, but not impossible; Explain this Big Bang rather than some other, non-life-permitting Big Bang is harder still, and may be impossible.
From which some will conclude “we don’t know why the universe is life-permitting” and others will conclude “an Intelligent Designer must have wanted to create life”. But, as I keep saying, this is metaphysics, not statistics. Dembski’s probability argument cannot be applied once we get to Big Bangs, because we only have one example.
Well, I am neither a chemist nor a physicist, but I do recommend you read Mike’s posts on this issue (try to ignore his fulminations against people he thinks ought to get it but don’t!). But I’d point out that that is what I tried to do originally – it was William who introduced the term “can-opener” to characterise my attempt to describe a possible OOL scenario in terms of known chemical processes (lipid vesicles, double stranded polymers, cyclical temperature gradients).
But I’m not sure what you mean by “increases the energy in the system” – what “system” are you referring to? Energy, as you know, is neither created nor destroyed, but it can be stored. Are you asking for instance how biological organisms store energy?
Well, as I said, it was William’s analogy, which he borrowed from Dembski, not mine. Analogies can be useful, but also misleading, as I think is the case when Dembski uses the “search” metaphor. In evolution there is no “searcher” and no “target” to be searched for. What it is is a system that is biased, by physics and chemistry, at bottom, to producing complex well-adapted organisms, whereas a system that was unbiased – where all outcomes were equally likely – would be vanishingly unlikely to do so.
Such a system requires a starting population of self-replicators. We do not yet know just how these emerged, but the working model is chemistry.
All scientific accounts are suppositions aka models, theories, hypotheses, hypotheticals. And Szostak’s supposition/model/theory/hypothesis is a series of events involving systems that are already known to exist (the behaviour of lipids; the behaviour of polymers; the behaviour of complex molecules in temperature gradients; the existence of convection vortices; osmosis). These are not magic can-openers, they are perfectly good, perfectly well described, perfectly replicable forces, and while we do not yet know precisely how they could converge to produce a proto-cell, Szostak’s team have an active research team which I suspect even in my life-time, should I exceed my three-score years and 10 (9 to go) by a decent margin, they will succeed in synthesising. But I could be wrong.
The point is that we have a hypothesis, not a can-opener, and hypotheses are testable.
And if they do succeed, then the Design argument simply has to go back to (as Dembski has anticipated): but why is there a chemistry that could do this? Why is the universe life-permitting?
Which, as I keep saying, is a perfectly good question, but not one that can be answered with statistics.
Whoah, you’ve jumped a bit. I thought you were at OOL – are you talking about the origins of chemistry?
I haven’t called on “magic” to explain the universe. I haven’t attempted to explain the universe at all. Maybe the universe was Designed (in which case “magic” seems as good a word as any, as we have no non-magic design mechanisms that could explain a pre-existence Designer) – maybe it’s just that non-existence is intrinsically unstable, and eventually, a life-permitting set of universe parameters will show up.
William, I have reasons for rejecting a Designer as the creator of our universe, but they are logical ones, not evidential ones, and could be wrong (essentially, I don’t find the argument that mind can exist without matter, persuasive – I know we differ on this).
But that’s not what we are talking about here. Here, we are discussing whether Dembski’s argument for a Designer holds water. My case is that it doesn’t, because it invokes statistical improbability to account for Big Bang. And he can’t do that, for the simple reason that you can’t calculate probability distributions for a class of event for which you only have one exemplar.
What “sufficient causal agency inside the universe” are you talking about?
ID, whether the postulated ID is a constituent of the universe or its cause, is an “unknown commodity”, William, unless you are postulating that the ID in question is an embodied entity, in which case we’d have to explain that embodied entity.
We have absolutely no evidence at all that intelligence is anything other than a property of material beings. I know you think that it is not, but it is certainly not a given. If you want to make the case that disembodied Intelligence is a known causal agent in the universe, than make it, but do not assume it made.
And you cannot, of course, infer it from the observation that the world looks intelligently designed, because that would be assuming your consequent.
Which would, as you say, be
This seems an extraordinary paragraph:
Is Dembski claiming that the information to make a life-permitting universe must have been (” But where in the Big Bang, with a heat and density that rule out any life form in the early history of the universe, is the information for life’s subsequent emergence and development on planet Earth?”) or cannot have been (“Conservation of information says this information has to be there, in embryonic form, at the Big Bang and at every moment thereafter”), present at Big Bang?
Is he saying that it must have been there, but not in the environment? The noosphere?
Any ideas, anyone?
Humans are intelligent designers, capable of paying such search debts. Do we exist outside the universe?
Lizzie,
Darwinists so often misunderstand when ID advocates speak hypothetically from the point of view of materialists. IOW, you cannot escape your information debt. Intelligence has the power to pay the debt; without considering it as part of the sufficient cause of such searches, then the materialist must trace that information back to the big bang, which still must have had that information in it.
Therefore, for a materialist, the convenient information for the advent of novel, functioning biological information had to be present in the big bang. ID theorists don’t require that information to be present in the big bang because we have a commodity that can produce seemingly unlimited quantities of it – intelligence.
Elizabeth says: “The point is that we have a hypothesis, not a can-opener, and hypotheses are testable.”
You don’t have a hypothesis germane to the debate at hand, which is about how to pay the information debt, unless you reach outside of the known universe to pay it – which is why Hawking and others write books and theories about multiverses – to pay the debt of the search for a finely-tuned universe.
You have a hypothesis that assumes the can opener. That is the point Dembski is making; you are not entitled to the can opener you are assuming. Sans that baseless assumption, you have only intelligence left to pay the debt for the search for biological novelty.
This is nonsensical. No intelligence I am aware of can produce ‘unlimited quantities’ of disordered matter and the physical forces to condense it in a pre-specified direction. ID simply hijacks a term – ‘intelligence’, ‘information’ – then pins whatever additional qualities to it as are necessary to force the conclusion that All Was Created. Nor does ‘Darwinism’ (FFS!) require any information to be present at the start, simply a process that generates it.
The kind of ‘information’ that drives life resides in entropy gradients – the physical quantity of entropy, not the informatic one. At the Big Bang, entropy was minimal, and any subsequent ‘information’ relating to future configurations effectively non-existent. But proximal matter responds to the forces acting upon it – each particle ‘informs’ the other – and because the universe winds down in a non-uniform and gradual manner, the change in entropy can permit local ordering, an ‘informative’ state that is not pre-specified in a highly energetic matter-less plasma, but emerges from it.
Lizzie,
Try again?
Wow, I entered a long post, and was told that the post was forbidden – and the software threw it all away. I worked hard on it. Sorry.
That has nothing to do with the debate. The debate is about the ability to fine-tune searches (one way or another) towards the construction of complex, specified, functional machinery. There is no known commodity that can do this other than intelligence.
Please explain precisely why you (you, William J Murray) think there is an information debt.
Which information? How are you defining information?
Do you think that Dembski is saying that Information is NOT conserved, and therefore must be provided by Intelligence as we go? Or do you think he is saying that information IS conserved, and therefore must have been put there in the beginning by an intelligence?
Or neither?
And please state, clearly how you are defining Information, and Intelligence.
Because William, not only do ID proponents not agree on many of these things, as far as I can see, their positions are frequently mutually incompatible.
Flint,
ouch. Did you mention p0ker?
My hypothesis was about how a system consisting of a smooth high-dimensioned fitness landscape might have arisen.
I hope you will tell me what the information debt is.
William, we are really at cross-purposes here. Let’s track back: what do you mean by “information debt”? Please be precise and specific. I am not even trying to “pay” this debt. Far from reaching outside the known universe to pay it, I am not convinced there even is one.
Until you tell me what the information debt is, I have no idea what you mean by can-opener.
What I did do was assume that Dembski was asking where a system capable of finding adapted organisms might have could have come from, and claiming that such a system was unlikely to arise by chance.
I’m saying that, given chemistry, such a system is quite likely to arise by Necessity. I’m not “reaching outside the universe” at all, because I don’t think there IS a debt.
Please put your case that there is, and what you mean by it – Dembski defines Information as the improbability of a pattern given random shuffling of its constituents, and Specified Information as the improbability of some specified class of pattern. He accepts that fitness landcapes, in other words, systems involving replication with variation in heritable reproductive success) can result in patterns with high Specified Information, but claims that such systems, are themselves improbable, given random shuffling of systems, and therefore there is an Information Debt. I am simply saying that, given chemistry (not a can opener – we know that there was chemistry on early earth, we do not know there were can openers, or indeed Intelligence), so we do not have to worry about random shuffling of systems – we have mechanisms that shuffle things highly non-randomly. So there is no debt. Or, if there is, it is paid by the chemistry we know exists.
Yes, there is. We know that a smooth high-dimensioned fitness landscape can do it.
Are you suggesting that only Intelligence (disembodied if necessary) can produce a smooth high-dimensioned fitness landscape?
Elizabeth,
All you are doing now is, IMO, attempting to derail the argument into semantics.
You agreed in principle that there is information required, in terms of search for novel biological function, or else you would not have attempted to show how an evolutionary algorithm working in a fitness landscape can generate sufficient information. You agreed that given the “chemical properties” of the universe and only chance & necessity to work with, it might be impossible to manufacture a self-replicating machine – another concession that there is an informational price that must be paid to acquire such a target that might be beyond the computational resources of the “chemical” and lawful properties of the universe.
You brought in the multiverse as an explanation for a convenient fitness landscape – another concession that something must pay the search information debt. You – and others – think this debt can be paid simply by providing enough brute chance. Dembski’s point is that no amount of “chance” – even that of near-infinite universes – can pay the debt in any plausible manner, nor can any amount of physical law – because physical law is incapable of producing the kind of contingency chance and ID can provide.
Others try to pay this search/information debt via multiverses, natural teleology, panpsychism, deep time, or other commodities, but what all these things have in common is an attempt to pay the search debt for acquiring the kind of information needed to construct novel, complex, functional machinery.
Now, after all this, you wish a “definition” of “information” and “intelligence”? I think rather you just wish to find semantic diversion from the corner you’ve painted yourself into.
BTW, the reason I don’t post here much is because I have a limited capacity to endure your ilk – this very kind of endless rabbit-holing and obfuscation of terms. But, out of a sense of fair play, I’ll spend some time here.
For as long as I can stomach it.