What A Code Is – Code Denialism Part 3

My intent here in these recent posts on the genetic code has been to expose the absurdity of Code Denialism. The intent has not been to make the case for intelligent design based upon the existence of biological codes. I know some people find that disconcerting but that would be putting the cart before the horse. No one is going to accept a conclusion when they deny the premise. And please forgive me if I choose not to play the game of “let’s pretend it really is a code” while you continue to deny that it actually is a code.

First I’d like to thank you. It’s actually been pretty neat looking up and reading many of these resources in my attempt to see whether I could defend the thesis that the genetic code is a real code. I admit it’s also been much too much fun digging up all the reasons why code denialism is just plain silly (and irrational).

That the genetic code is a code is common usage and if “meaning is use” that alone ought to settle the matter. But this is “The Skeptical Zone” and Code Denialism is strong here. But I’m not just claiming that it’s a code because we say it’s a code in common usage. I’m claiming it is a code because it meets the definition of a code. The reason we say it is a code is because it is in fact a code.

My first two posts have been on some of the major players and how they understood they were dealing with a code and how that guided their research. I’ll have more to say on that in the future as it’s a fascinating story. But for now …

What A Code Is

: Information Theory and Coding
: Norman Abramson
: 1963
: Chapter 3

Definition. Let the set of symbols comprising a given alphabet be called S = {s1,s2,…,sq}. Then we define a code as a mapping of all possible sequences of symbols of S into sequences of symbols of some other alphabet X = {x1,x2,…,xr}. We call S the source alphabet and X the code alphabet.

Definition. A block code is a code which maps each of the symbols of the source alphabet S into a fixed sequence of symbols of the code alphabet X. These fixed sequences of the code alphabet are called code words.

“…the genetic code is a block code…” – Yockey, 1992. p. 102

For some people the word “symbol” seems to be a stumbling-block. Tt’s not quite clear why that is the case.

: Introduction to Coding Theory – Lecture Notes

Definition 1.1

Let A = {a1,…, aq} be an alphabet; we call the ai values symbols.

More on that shortly.

: Information Theory, Evolution, and the Origin of Life
: Hubert P. Yockey
: p. 6

The genetic code has many of the properties of codes in general, specifically the Morse Code, the Universal Product Bar Code, ASCII used in computer equipment, and the U.S. Postal Code. I shall explain the relation of these codes to the genetic code in the following discussion. Every code, as the term is used in this book, can be regarded as a channel with an input alphabet A and an output alphabet B. Here is the formal definition of a code.

Given a source with probability space [Ω A, pA] and a receiver with probability space [Ω B, pB], then a unique mapping of the letters of alphabet A on to the letters of alphabet B is called a code. (Perlwitz, Burks, and Waterman, 1988)

Here pA is the probability vector of the elements of alphabet A and pB is the probability vector of the elements of alphabet B.

My claim has been all along that the genetic code meets the mathematical definition of a code. This is hardly a matter of dispute.

To close out this OP I’ll leave you with the following:

Sydney Brenner remembered going to a talk Sanger gave at that time [1951] – the excitement, especially among the younger scientists, as they emerged. “At last, we knew what proteins were.” With that vanished any possibility of a general law, a physical or chemical rule, for their assembly. “With that, you absolutely needed a code,” Monod said.

119 thoughts on “What A Code Is – Code Denialism Part 3

  1. Repeating my question:

    Mung,

    Do you think the genetic code was designed by an intelligence? If you do, then why do you believe that?

  2. Mung: Are you saying that codes do not exist without someone to write them down using symbols? What makes it a code is in the writing of it? And of course, what the writing represents is the relationship, but it’s not the relationship that is the code?

    As long as you are not asserting a regular (as in syntactical) relationship between the sequence of DNA elements and phenotype, I’m OK with the word code.

    I suspect, however, that when an IDist uses the word code, he is implying a kind of language with which one could design.

    Just a suspicion.

  3. Mung,

    Are you saying that codes do not exist without someone to write them down using symbols? What makes it a code is in the writing of it? And of course, what the writing represents is the relationship, but it’s not the relationship that is the code?

    A map of a hill is not a hill. Representing the causal relationship in someone’s head or on a white board represents the physical as code – we can do a lookup of the map.

    Try answering the following:

    – If there were only one tRNA, would its docking codon be a code for its attached amino acid?
    – If there were two tRNAs but the same acid on each, would that be a code?
    – If there were two tRNAs and two acids, would that be a code?

    I’m interested in the point you would think it becomes a ‘coding’ relationship?

    Or think of a simpler element of genetic mapping – complementarity. Is Adenosine the code for Thymine? Is Thymine the code for Adenosine? Cytosine for Guanine? What insight do we gain by saying either ‘yea’ or ‘nay’?

  4. Mung: Are you saying that codes do not exist without someone to write them down using symbols? What makes it a code is in the writing of it? And of course, what the writing represents is the relationship, but it’s not the relationship that is the code?

    ETA: Agree with Glen. Look at that.

    The point is that there are different types of codes–if we define the term broadly. Fighting over definitions is really really silly.

  5. GlenDavidson: I wouldn’t think so. It’s mapping, but why “symbols”? Just because that’s what human-made codes use? That’s what I’ve never liked about denial of the genetic code as code, the assumption that a code must be just like a human code, rather than to simply work “like a code” in the sense of one thing mapping to another.

    It’s important to define “mapping” also, or exactly the same argument will break out over whether every time we have a certain type of correspondence in nature we have a mapping.

  6. GlenDavidson,

    That’s what I’ve never liked about denial of the genetic code as code, the assumption that a code must be just like a human code, rather than to simply work “like a code” in the sense of one thing mapping to another.

    I have certainly never argued that the code does not work ‘like a code’. But a code is something involving human-mediated mapping, according to the definition I have been using. Until, that is, the genetic version was discovered and we had to extend the definition to accommodate it. It didn’t fit into the original definition.

    Would we be justified in saying that gene control is a computer program, simply because it resembles one? I guess I have a resistance to analogy, or rather its over-extension.

  7. Is a cipher a code?

    A little thought experiment.

    I Take the most stochastic source of bits known — the output of a Geiger counter — and XOR this stream of bits with a message stream. Does the relationship between the input and output constitute a code?

  8. Allan Miller:
    walto,

    I would not disagree.

    Well the solution is simple: disambiguate. Define “code1”, “code2”, etc. And always be meticulous when we use the term(s). Equivocation becomes impossible and real disagreements become clear.

  9. Allan Miller:
    GlenDavidson,

    I have certainly never argued that the code does not work ‘like a code’. But a code is something involving human-mediated mapping, according to the definition I have been using. Until, that is, the genetic version was discovered and we had to extend the definition to accommodate it. It didn’t fit into the original definition.

    “Evolution” is another word that was extended to fit biology. I don’t really have a problem with doing so. I think that “evolution” extends less well, in fact, since etymologically it suggests “rolling out.”

    As I noted in another comment, the “genetic code” happened to fit the category of “code” better than it did other categories we had lying around, which seems to me to be why it’s called a “code”–and that seems to be standard practice in categorizing things.

    Would we be justified in saying that gene control is a computer program, simply because it resembles one?

    No, because “computer program” is the specific, namely, specific to computers. “Program” is the general, and it is not infrequent that “program” is used in biology, “development program,” etc. I’m not sure just how closely gene control really resembles a computer program, however, especially given the stochastic nature of gene control.

    I guess I have a resistance to analogy, or rather its over-extension.

    I certainly don’t like the claim that life is designed by evolutionary processes. That seems like over-extension, indeed (unless the person is using words rather incautiously, which is ok in some contexts). To me, though, there is little conceptual difference between the mapping of nucleotide triplets to amino acids and the mapping of dots and dashes to the English alphabet.

    Still, it’s all semantics and categorization with most of us, while to the IDists it’s typically a matter of absolutes. Meyer’s “killer argument” in Signature in the Cell was (paraphrasing) that if you found language that had been written out 20 million years ago in a cave somewhere with no evidence of anything that used language, you’d still conclude that intelligence did it. And the genetic code is the same thing, in essence, therefore intelligence did it. Uh, no, life with its parts is not in essence the same thing as a written language, so his reductionism is both breathtaking and illegitimate.

    Glen Davidson

  10. Mung,

    If this is aimed at me, saying I’m equivocating over the use of the word code, please read the OP.

    Nope, not aimed at you. While I would like to hear your definition of “real code”, I’ve been reading your comments on this thread with some interest.

    I was referring to the whole “a code implies a code maker” argument, which does equivocate by first defining a code without reference to any external agent, then changes the definition implicitly after getting another participant to agree that genetic code is a code under the first definition.

    Personally, I don’t care if it’s called a code or a template or a process or a pathway or anything else so long as the definitions are clear and rigorously adhered to throughout the discussion.

  11. Besides pedantry, stubbornness and a certain amount of red-rag-waving, I have been trying to get people to focus upon the essentials of the code , and the relationship between its assumed code-iness and its evolvability. Whether one calls it a code or not is entirely a matter of taste, IMO.

    People seem reluctant to dissect the code and see where it breaks down – what would be a minimal code, however it may please them to define it. Hence my suggestion that one consider what one would call a single-tRNA-charging system, then 2-to-1-acid, then 2-to-2. Or, alternatively, the appropriateness of ‘code’ for the relationship between a base and its complement, or a series of bases and the complement of that.

    It is easy to see that a tRNA charging-docking system can evolve in principle, provided that 20-acid catalytic proteins are not essential from the get-go (which has nothing to do with codes, but the physical capacity of the various biomolecules).

  12. Allan Miller: and the relationship between its assumed code-iness and its evolvability.

    I think folks having IQs above room temperature have no problem with evolvability.

    If Dembski or Meyer or Upright or Mung had an actual argument against evolvability, they would present it directly instead of wallowing in horseshit.

    Perhaps repetition will make the claim true.

  13. Really, all you have to do is look at the actual history of OOL, in pathetic detail, and point out the step where the miracle occurred.

  14. code 1 any system of symbols, together with the rules for their association, that can be used to represent or transfer information; e.g. the genetic code.

    – Oxford Dictionary of Biochemistry and Molecular Biology.

  15. Patrick: While I would like to hear your definition of “real code”, I’ve been reading your comments on this thread with some interest.

    Definitions are in the OP. I’ve consistently referred to mathematics, more specifically to codes as defined in Information Theory/Coding Theory.

    “…the genetic code is a block code…” – Yockey, 1992. p. 102

    The “real code” bit is a nod to Neil (the mathematician), who claimed to not know what a “real code” is. At the time I should have asked as compared to what. Fake codes? Imaginary codes?

    The genetic code is a code. It “really is” a code. In the future I’ll try to remember to drop the “real” part as quite superfluous.

  16. Mung,

    For the third time, do you think the genetic code was designed by an intelligence? If so, what are your reasons for believing that?

  17. Mung: code 1 any system of symbols, together with the rules for their association, that can be used to represent or transfer information; e.g. the genetic code.

    – Oxford Dictionary of Biochemistry and Molecular Biology.

    Indeed, a somewhat unexpected definition there from the Oxford Dictionary of Biochemistry and Molecular Biology.

  18. …the existence of a molecular symbol-language and its semantic fixation in the genetic code suggest strongly that the concepts of information theory may be applicable here.

    Bernd-Olaf Kuppers, Information and the Origin of Life. p. 32

  19. Has it escape notice that Allan is the one arguing for a necessary connection between codes and the mental, while I am arguing for understanding codes as a part of the natural world of biology? Allan’s position is what makes intelligent design attractive. Now that is irony.

    Doesn’t mean he’s wrong. The existence of a code really could entail a mind.

  20. LOL at Mung desperately Googling for any definition of “code” he can claim supports his “designed” equivocation!

  21. Mung,

    For the fourth time, do you think the genetic code was designed by an intelligence? If so, what are your reasons for believing that?

  22. keiths:
    Mung,

    For the fourth time, do you think the genetic code was designed by an intelligence? If so, what are your reasons for believing that?

    +1.

    Are you working towards a point, or just looking to muddy the waters, Mung?

  23. Mung: Allan’s position is what makes intelligent design attractive. Now that is irony.

    It doesn’t actually. It only seems that way to you because you do not understand it.

  24. Neil Rickert: Mung: Allan’s position is what makes intelligent design attractive. Now that is irony.

    It doesn’t actually. It only seems that way to you because you do not understand it.

    You apparently missed his response to your original foray into this thread:

    Mung: …the genetic code meets the mathematical definition of a code.

    Neil: Well, quite obviously, it doesn’t.

    Allan: Unless you write it down!

  25. Mung: You apparently missed his response to your original foray into this thread:

    Mung: …the genetic code meets the mathematical definition of a code.

    Neil: Well, quite obviously, it doesn’t.

    Allan: Unless you write it down!

    No, I didn’t miss it. I agreed with it.

  26. But the point here is that it is hard even to think about the problem if one does not think of genes sending signals, and if one does not recognize that the signals are symbolic.

    – John Maynard Smith. The concept of information in biology.

  27. That’s right Neil, far be it from you to contradict yourself.

    The important point here is that what makes something a mathematical object is human intentions, rather than its physical properties.

    Yes, you agree with Allan. Yes, you both adopt the mentalist approach. Yes, that’s irony.

  28. Mung,

    code 1 any system of symbols, together with the rules for their association, that can be used to represent or transfer information; e.g. the genetic code.

    – Oxford Dictionary of Biochemistry and Molecular Biology.

    Sorry, that’s just wrong. Next!

    eta – the last bit, not the first. Argumentum ad dictionarium

  29. Mung,

    Has it escape notice that Allan is the one arguing for a necessary connection between codes and the mental, while I am arguing for understanding codes as a part of the natural world of biology? Allan’s position is what makes intelligent design attractive. Now that is irony.

    Your reasoning does not stack up. If one accepted a definition of codes as necessitating a mental connection (not strictly what I’m saying, but let it pass), then the genetic code either is or is not a code. One cannot simply declare it so, because a necessary condition has not been established.

    Doesn’t mean he’s wrong. The existence of a code really could entail a mind.

    You think? Wow, get onto the DI; they love this kind of thing!

  30. I guess my view might be best summarised as: a code is a map – a map of inputs to outputs. Morse code is a map. The genetic code written in a textbook is a map. But the physical system involved in generating proteins is not a map, ergo not a code (in that sense, I should hardly need add). It may be a map in God’s head, it may not.

  31. The further point being, if one thinks one can define God into the picture, it is a breeze to simply define him out again.

  32. I think semiosis is the dumbest line of ID argument ever.

    It is a textbook example of map/territory fallacy. Maybe worse than the information fallacy.

    The standard form seem to be: find a metaphor, reify it, proceed to argue that the properties of the reified abstraction impose entailments on the original object or process.

    It doesn’t rise to the level of cargo cult science. It ‘s cargo cult philosophy.

  33. One observes that landscape gardeners produce some kind of a plan. Then they implement it. Then someone comes along years later and maps it out. There is a ‘prior-map’, there is ‘the thing’, and there is a ‘post-map’. But we don’t (typically) argue that the only way to produce landscapes is through a Gardener. For a natural landscape we can produce a post-map, but there is no prior-map.

    The genetic code is a ‘real code’ when it is a map of the system – a post-map. The system, however, is not a code but ‘a thing’. Whether or not there is a ‘prior-map’ is the Quod-Erat-Demonstrandum of the whole Argumentum. Get Demonstranding!

  34. I would like to see mung demonstrate that it is possible to produce a pre-map of a genome.

    I followed his links on semiotic biology. There is a casebook selling for $118 and several blog essays dating from the last century. No evidence of any follow-up or progress at all.

    I read the blogs. Several times. They are completly confused about ordinary evolution. They address nothing except 19th century Darwinian selectionism. No molecular evolution. No drift. Nonpopulation genetics. It’s very close to Lamarkianism.

  35. Allan Miller:
    I guess my view might be best summarised as: a code is a map – a map of inputs to outputs. Morse code written in a textbook is a map. The genetic code written in a textbook is a map.

    Fixed that for you.

    But the physical system involved in generating proteins is not a map…

    And yet, oddly enough, codon sequences get mapped to amino acid sequences. And who is claiming that “the physical system involved in generating proteins” is a map. What would that even mean?

    Also, one can only wonder why people thought a code was present before they actually started to put the code in writing.

  36. With the self-replication of nucleic acids, we have arrived at the fundamental step leading from chemistry to biology: the four bases of the nucleic acids begin to play the part of linguistic symbols. The sequence of these symbols can encode a message. There thus arises an attribute that is new to the world of physics and chemistry, one unknown to the science of material interactions, of atoms, molecules, and crystals: that of information.

    The definition of information requires, first of all, a restricted set of symbols; secondly, the concatenation of these symbols into chains, or sentences, whose structure is defined by a grammar and whose meaning is realized by semantic agreement; and thirdly – a requirement often tacitly ignored – an apparatus for the reading (and, if necessary, for the translation) of the message contained in the symbol sequence. All of these three requirements for the use of nucleic acids as information stores are fulfilled …

    – Manfred Eigen. Steps Towards Life. p 65

  37. Allan,

    Whether or not there is a ‘prior-map’ is the Quod-Erat-Demonstrandum of the whole Argumentum. Get Demonstranding!

    Amen.

  38. Allan Miller:

    The genetic code is a ‘real code’ when it is a map of the system – a post-map. The system, however, is not a code but ‘a thing’. Whether or not there is a ‘prior-map’ is the Quod-Erat-Demonstrandum of the whole Argumentum. Get Demonstranding!

    This is the same bait-and-switch the IDiots use when they crow about DNA having lots of specified information. All they have is an after-the-fact description of the DNA and amino acids, not a prior specification.

  39. The genetic code (sensu stricto) is the rule which prescribes, given polynucleotide sequence, the corresponding polypeptide sequence.

    – Jacques Monod. Chance and Necessity. p. 181.

    And it has nothing to do with whether or not someone scribbles some symbols on a piece of paper.

  40. Mung:
    The genetic code (sensu stricto) is the rule which prescribes, given polynucleotide sequence, the corresponding polypeptide sequence.

    – Jacques Monod. Chance and Necessity. p. 181.

    And it has nothing to do with whether or not someone scribbles some symbols on a piece of paper.

    It has nothing to do with symbols used as abstract representations at all. 🙂

  41. The map arose at the same time as the code. That’s what the code is. It’s only this idea that you all have that associates codes with mental activity that calls for a ‘prior-map.’

    See here for an intro to code-makers.

  42. Mung:
    The map arose at the same time as the code. That’s what the code is. It’s only this idea that you all have that associates codes with mental activity that calls for a ‘prior-map.’

    See here for an intro to code-makers.

    Shorter Mung:

    “Here’s some more new age woo about codes I Googled up today. Take that evos!”

Leave a Reply