Tuesday, April 27, 2010

"Intelligent Design" and Scotism

Now that I've prepared the ground a little bit, I feel more prepared to consider Dr Torley's claim that ID proponents are Scotists, or at least that Scotism is more sympathetic to their position than Thomism is.

I find little to complain about in Dr Torley's account of univocity and intelligence as a pure perfection in Scotus. It's accurate enough for his purposes. I'm less sure, however, about his attempts to apply these notions to ID.

First of all, the ID definitions of "intelligence" seem very unscotistic and unscholastic to me. Consider the following:

Intelligent Design. The study of patterns in nature that are best explained as the product of intelligence.

Intelligence. Any cause, agent, or process that achieves and end or goal by employing suitable means or instruments.

Design. An event, object, or structure that an intelligence brought about by matching means to ends.


These definitions seem to be designed to lead to the desired conclusion from the outset, but they're very odd. On the one hand, since any pattern in nature whatsoever is, insofar as it is patterned, intelligible, one might use a metaphysical argument to infer an intelligence: any intelligible in potency, the argument might go, implies a prior intelligible in act, that is, an understanding intelligence. But this applies as much to non-living things as to living ones. The ancients argued to God from the regularity of the motion of the heavenly bodies without examining their structure for complex specified functional information.

Furthermore, I'm not at all sure about the definition of intelligence as something that achieves ends by employing means. I suspect that one could argue that computers and elephants and Venus flytraps are intelligent by this standard. But traditionally intelligence is defined in terms of the act of understanding, that is, in the particular way that the forms of intelligible things are objectively apprehended by the intelligent thing, and in terms of intentionality. My computer neither understands nor intends, but it matches means to end in accordance with its programming - doesn't it?

By the way, every "scientific" attempt at defining intelligence in terms of some marker-activity like the use of tools, communication, problem-solving, etc., has always seemed wildly off the mark to me. These all may or may not be intelligent activity. But unless the activity is also intentional and objectively apprehensive, it's not indicative of intelligence per se. In my estimation the best marker for intelligence is still the capacity for abstraction.

Similarly, the definition of "Design" seems much too broad. When an angry monkey throws feces at me to make me go away or a hungry monkey throws a rock at a banana clump, isn't this matching means to ends? When a bird looks for twigs to build a nest, isn't it matching means to ends? But I don't think we want to call either the feces-volley or the nest indications that the monkey or the bird are designers. And, as I indicated earlier, when looking at the world of nature I'm not sure how the intelligibility and order in the world of physics is either more or less indicative of "design" than that in the world of life: both non-living physical objects and living things exist and come to be by nature, which exhibits a marvelous order and intelligibility.

Because of this I'm not sure the attempt to bring in Scotus does any good. Dr Torley writes:

Duns Scotus distinguished God’s intelligence from ours on the grounds that God’s is essentially infinite, while ours is finite. . . . The underlying idea here is that an intelligent enough Designer can create any effect, no matter how complex. That does not mean that the difference between God and humans is merely one of degree; it simply means that the term “intelligent” has the same meaning when predicated of God and ourselves, notwithstanding the infinite disparity between God’s intellectual capacities and ours, and the utter dissimilarity between the way God thinks and the way we do.


This is fine, but when we're talking about the ability to create any effect, we have to take into account what sort of thing the effect is, and a created tree, whether created in toto and ex nihilo, or put together out of pre-existing parts, or evolved over millions of years, or whatever, is still as much a natural thing as a star or a rock, and a created jet plane or laptop computer or water-driven mill is an artifact. And since the tree is natural, and since the parts of the tree are formed by the tree as a whole, therefore all its parts are natural, and the existence of all its parts is accounted for by the existence of the whole. It seems to me that any discussion of the origins of the tree must recognize this.

Anyway, the notion that we can think about God as a designer in the way that we can think of men as designers, because there is a univocal conception of intelligence as a pure perfection, seems to be the only actual attempt to use specifically Scotistic notions in Dr Torley's long post. It doesn't seem to me that it's very successful, not because I think he's necessarily distorting Scotus so far as univocity or pure perfections go, but because, as it seems to me, the notion of a living thing, or a part of a living thing, as machine-like in the relevant sense, is misguided. And if living things are not machine-like, then we cannot infer machine-like origins for them.

So what would a Scotist say about the origins of life? For myself, I would say nothing. As I said before, I'm no biologist. So far as I know there may well be natural laws which can be discovered scientifically which account for the origin of life on their own. Or perhaps there are no such laws and only a non-natural origin of these natural things can be posited. That seems to be an empirical question. If scientists can show life arising from non-life naturally, let them! If they can't, then don't take their stories for granted.

As a metaphysician, I would insist that act is absolutely prior to potency, and that being does not arise from non-being. So any account of life's origins would have to cope with that. From my layman's understanding it appears that naturalistic Darwinism fails in this regard. But to say that doesn't automatically mean embracing ID as Torley presents it. It seems to me that one can admit as a point of metaphysics and theology that God is the creator and architect of the world without taking any stance whatsoever on the manner in which the first instances of certain kinds in the world came to be. God is the first cause of all that is, and in his Providence governs every event in the universe without exception; but does this mean that a given living cell was made by natural forces, like Mt Sinai, or by the immediate writing of the finger of God, like the tables of the commandments? Or neither? In any case it's not clear to me that a thing that, if it is to be at all, is a natural kind, like a tree and not like a jet, must be understood as having been (at least in its first instance) assembled, as Dr Torley says.

None of this is to say that I side with everything Dr Feser says in this debate. I do not admit, for instance, that when God makes something this is to be understood as God "conjoining an essence with an act of existence", since I have severe doubts about the real distinction and composition of essence and existence. Perhaps more germane to the subject at hand, I have a lot of sympathy for interlocutors like Lydia McGrew who object to the Thomistic account of generation, in which the new thing annihilates and replaces the substantial form of the old. It seems to me that there is a real case to be made here that the Thomistic doctrine of the unicity of substantial form is difficult to reconcile with many facts about the natural world; I would argue that a Scotistic plurality of substantial forms is much more consistent with both observation and common sense. That's a metaphysics debate for another day, however, and doesn't seem to have any bearing on the critiques presented either by Dr Feser or myself here.

Finally I want to emphasize once more that I am a novice to this debate and have been working out my thoughts in these posts for the first time. I've never read an ID book and can't claim to be conversant with the range of positions. This is a response to Dr Torley's recent post alone. My views on ID, evolution, etc., are subject to refinement and revision. Just today I was at the local used bookstore and picked up a copy of Mere Creation: Science Faith & Intelligent Design, a collection of essays on ID by various contributors and edited by William Dembski. So I may read it and have something more intelligent to say about the subject in the future. Or perhaps not!

22 comments:

David said...

"The ancients argued to God from the regularity of the motion of the heavenly bodies without examining their structure for complex specified functional information"

I dunno — sure, they didn't talk in those terms, but the recognition of order is acknowledgement that that motion specifies some kind of "function". They just had lower — or should I say, higher — standards than we do today. Centuries after Newton, we've come to take boring old elliptical motion for granted; modern scientists just start with these "fundamental laws of physics" as givens. To a philosopher, that's begging all sorts of questions, and so philosophers can still conclude an intelligence behind even the "simple" order of physics. It's somewhat understandable that scientists qua scientists aren't metaphysicists, so making the connection on a more complex level (e.g. biological patterns that are too complex to be specified in terms of physical functions alone) is not an unreasonable challenge.

"These definitions seem to be designed to lead to the desired conclusion from the outset, but they're very odd."

I agree that defining "ID" is itself a mighty challenge, but then you just used the word "design" yourself there in the sense of "matching a means to an end". In fact, I'm not sure you didn't use it deliberately... which would be yet another example of design in that sense. So I don't think the definitions are too odd! (Just still not rigorous enough for my liking.)

"I suspect that one could argue that computers and elephants and Venus flytraps are intelligent by this standard."

One could, and why not? Not that they possess intellects themselves, but in the indirect sense that we say a piece of writing is intelligent — in other words, "understanding" was required by whatever caused these things. (Or perhaps that "abstraction" is necessarily required to come up with these things, which I guess is another way of saying that the cause must at some point have included acts of intelligence.) How much abstraction animals are actually capable of is an interesting diversion, but the bird or the monkey need not be intelligent itself, as long as its ability to match means to ends can be explained by, say, the direct or indirect intelligence of God.

And again, I would agree that the physical world itself is "intelligent" in this indirect sense; as ID also argues, based on cosmological examples such as our uniquely suitable position in the galaxy, or the moon's perfect size in regard to make solar eclipses possible, etc.

Michael Sullivan said...

David,

it seems to me that you're conflating intelligibility with intelligence.

David said...

Well, I'm just trying to give the best possible interpretation to Torley's definitions! I think it's fair to read it as "Any ultimate cause that achieves ends via means", meaning a bird can exhibit intelligent behaviour without necessarily being the source of that intelligence. But I'm still not sure that avoids circularity....

I do see the distinction Torley is trying to get at; there is a difference between the design of a nest-building instinct and the design of planetary motion, say. Although the way planets move depends on, and instantiates, intelligible forms, it doesn't seem directed to an independent end the way nest-building is. That is, the end of planetary motion is just to move planets elliptically, isn't it? The obvious goal of nest-building is not to engage in nest-building activity, but to end up with a nest. If we could get the nest without the activity, then the activity would be superfluous. But I don't see what we get from, say, planets moving according to precise gravitational patterns other than the planetary motion itself. So while both are evidence of a causing intelligence (at least to those of us who are scholastically disposed), the kind of design that Torley is trying to define is on a different level. Planets and birds both have final causes, but the birds' final causes have their own final causes, or something like that.

"But unless the activity is also intentional and objectively apprehensive, it's not indicative of intelligence per se. In my estimation the best marker for intelligence is still the capacity for abstraction."

Hm, don't we need more than that? Otherwise, we're open to calling computers intelligent; they can certainly manipulate data formally and spit out a representation of an abstracted form. But the computer doesn't understand anything about that form. So abstraction+understanding?

Malcolm Chisholm said...

An enormous tract of ground is covered in this post. A problem right at the beginning is the definition of "intelligence". This is an equivocal term. My friends in the Intelligence Community have an open project on how to define it from their perspective (http://tinyurl.com/2alz3cu), and lack of a clear definition is enormously problematic for them. Of course, this has nothing to do with Intelligent Design (ID), but that is precisely the point. Without a clear definition if "intelligence" we are going to have problems with "intelligent design".

The definition of "design" presented is shallow. I do design for a living so I have to think about it. There is no design without requirements, but these are not always ends. In software development, we have functional requirements (what the intended system is supposed to do) and non-functional ones (e.g. it must be developed with a certain budget). Also design is very much about throwing away things and foreclosing possibilities. I cannot have a car that has high acceleration and gets high gas mileage. Animals that reproduce quickly do not live long.

And then we have "information". There seems to be no adequate definition for it in the ID debate. Oddly, the debate seems to be trying to pull it from the Conceptual Order into the Real Order - from logic into metaphysics. What "information" is may be the key to finding a resolution to the whole ID debate.

I liked the comment on abstraction as a marker for intelligence. As a side note, I am annoyed that the modern students of scholasticism consistently refuse to provide any practical guidance to people in my profession - information management. On a daily basis we have to deal with problems that I know the scholastics contributed to. Take abstraction for instance. What kinds of attributes should I remove from a database design when I perform abstraction? Are there good and bad patterns of abstraction? And just what is abstraction - is it a top-down or bottom-up activity? There are enormous sums of money involved in these decisions and we get no help from anyone. I suppose the only answer is for information management to gradually understand the scholastic landscape. I have been doing this personally and have just published a book on definitions (www.data-definition.com). It would not have been necessary if you guys even dimly appreciated what the Information Age is and that perhaps you hold the keys to it.

eric said...

Malcolm asked about the definition of "information".

In regard to biological information, this includes encoded symbolic information, which requires translation to decode into the functional counterpart that it represents.

Since you are familiar with computing, you deal with this all the time. The one's and zero's of binary information do not have any intrinsic meaning at all. Symbolic meaning is always extrinsic. The same binary word of one's and zero's might be found as a piece of a document, a song or sound file, an image, or within the instructions of an executable program.

When developing software, one also has the example of higher levels of symbolic information, such as symbolic code that must be translated by a compiler, interpreter, or assembler into executable instructions.

Symbolic information is the type of information we find in DNA that holds encoded instructions for the directed construction of functional proteins. (Notice, cells do "build" proteins, through translating instructions, assembly and folding.)

It cannot be overemphasized that the meaning for this symbolic information is extrinsic to the sequences in the DNA. We have known for decades that the genetic code is not universal. Some organisms use variant genetic codes to decode their encoded information. The particular coding scheme that assigns meaning to the DNA information is expressed by the particular set of transfer RNA used in the ribosome of that organism.

I hope that helps clarify the sense in which information is foundational to biological life. You are quite right that it is a very important concept for the whole topic.

Michael Sullivan said...

Malcolm,

I definitely appreciate your comments.

Eric,

It doesn't seem to me that you have adequately defined information, since the term to be defined is included in the definition.

I am also doubtful about your claim that DNA contains symbolic information. People I know who are a lot more familiar with the biological facts than I am are very skeptical about the presence of semantic or symbolic content in DNA. If it were generally agreed on by biologists that the genetic structure actually contains a language, not just metaphorically, then we might be getting somewhere. But to my knowledge there is no such agreement. It seems at least possible to me that "code" is a code only in a metaphorical sense. The same thing goes for your comment in another thread about the bacterial flagellum as a nautical robot. I'm not at all sure that your picture of the process isn't tainted with anthropomorphization and metaphor there too. One thing that I find suspicious is that the language ID people use to describe living structures, in terms of codes, computers, robots, and so forth, are suspiciously derived from the most distinctive technological aspects of our own day, the innovations most indicative of our own time. It's just like previous ages in which scientists talked about life in terms of clockwork, when clockwork was the most impressive and complex machinery we had. But the ID proponents are claiming that their language is not metaphorical, that living things are actually computers and robots. This is, forgive me, just a priori improbable and suspicious. If our interests or technology change it seems all too likely that our concepts of living things as computers and robots will be scrapped and replaced with whatever the new flavor of the month then will be. We have to be on guard against interpreting the natural world in terms of those of our creations of which we are most proud and of which we think most often, lest we fashion the world's and our creator in our own image.

Malcolm Chisholm said...

Michael,
I liked your comments about the mood of the age predisposing us to interpret living things in terms of robots etc. A major problem is that we do not understand what the Information Age is. We try to define it in terms of technology, like hardware, operating system software, and so on. These are today's clockwork, and then we try to analogize it to living things. The problem of the Information Age is that we have used technology to build giant ecosystems that we have populated by entia rationis (sorry - I don't speak Logician's Latin). These beings of reason are things like mortgages, overdrafts, ownership, credit default swaps, employee actions, asset backed securities, and even money. They are all ideas, but the Information Age has given us the technology to manage them on a scale never before imagined. And, if you judge by recent market events, it is all totally out of control.

Yet, we seem to have to reflexively think in positivist terms, such that all facts are external to us, there is no real connection between knower and known, and that everything is based on material existence. Tables and roses can be perceived, and we can form images of them in our minds. But an overdraft cannot be perceived - it can only be conceived. Nobody can form an image of it because it does not exist in the material world. The piece of paper that comes in the mail which tells me that I have an overdraft is not an overdraft. If I thought it was, I would burn it and thereby get rid of my overdraft. But the overdraft is an idea. The record of it - the data about it - is what we can now so easily manage in the Information Age. Yet the overdraft only exists because I, the bank, and the law all agree that it exists as an idea.

From this we can see the inadequacy of trying to implicitly define the Information Age in terms of the Real Order when it is so much more about the Conceptual Order, and then trying to analogize living things in terms of the technology that supports the Information Age. It is to almost entirely miss the point. DNA, I would submit, is essentially data. It is data in the same way as the record of my overdraft is data. But the record of my overdraft corresponds to an idea, not to a real entity. If human data can ultimately correspond to ideas, why should the data in DNA not also (admittedly indirectly) correspond to ideas. We can then begin to see how ID might be possible.

Michael Sullivan said...

Malcolm,

These are some very good comments and I'm largely in agreement. I've said to my friend Faber before now that I have a suspicion that much of what is wrong with modern analytic philosophy (which lies behind the origins of computer science), going all the way back to the formulation of set theory in the 19th century, is that it doesn't adequately know how to distinguish between real beings and beings of reason (entia rationis is correct, by the way!).

On the other hand, I'm not sure that the notion of "data" is any more clear than the notion of "information". It's easy (for experts, not for me!) to look at a string of ones and zeros in a computer and tell what concept they correspond to, and how the manipulation of the symbols can produce a corresponding manipulation in my concepts at the end of the computation. While there is any analogy with DNA in the sense that strings of particles undergo operations which have determined end results, what I'm not clear about is the reasonableness of saying that the strings of particles correspond to concepts. And if they don't, then strictly speaking there isn't a language and there isn't a code.

eric said...

Dr. Sullivan,

It would be difficult to respond to those people "who are a lot more familiar with the biological facts" than you are, without more details about the basis of their skepticism. Is it mere subjective, psychological, or ideological bias (some people simply do not like this idea for ideological reasons), or else something more? If so, what? You can see I have no way to know.

Not knowing more, I will say that just as an oak tree is not merely "like" a tree, but rather is a tree, in the very same way, a genetic code is not merely "like" a code. It is a code that maps from triplets of bases to amino acids. Likewise, it is common knowledge and a long established fact that the information in DNA needs to go through a step of translation using the appropriate code for that information.

Furthermore, it is an observable plain fact that the interpreted meaning of the sequence is extrinsic to the sequence itself. (The transfer RNA that express a given genetic code are not even in the same location as the stored DNA, hence the need for messenger RNA to transport the information.)

How is this in any sense essentially different than, say, Morse code?

By any reasonable definition of code and symbol, the genetic code is functioning as a true code, and the codons are indeed functioning as symbols.

Likewise, the reality of the assembly of molecular machines has been documented beyond reasonable denial. Science has uncovered not only this reality, but additional layers of incredibly complex details -- assembled machines building other machines.

Please note also what I said under nature-artifacts-and-machines-2 about how this is an essential aspect of the properties of cells that is necessary to their ability to function. They rely upon preprogrammed, on-demand assembly. "Growing" copies from originals would not work.

Most importantly, there is a question I raised under nature-artifacts-and-machines-2 that I don't think you've yet touched on. Even if you are skeptical and hesitant to believe me, nevertheless, what is the case if what I am saying is completely and accurately true?

If I am describing the true situation (including the fact that the cell not only can but must depend on programmed assembly of machinery), that shouldn't cause metaphysical philosophies or your position to crumble, should it?

If your entire response rested heavily on appearing to deny the reality of what biology has discovered, that would clearly put you and your philosophy into an apparently very weak posture. (I think you would agree this would be doubly so if the basis for denying these facts rested on the unspecified skepticism of nameless people who are said to know more than you do about biology.) I don't think you really intend to base your case on denying biology.

Since I believe you readily recognize this, I think you see plainly that a robust response has to be able to consider the possibility that what I am describing is really true.

If it really is just as I have described, then what?

eric said...

"A structural identity has been discovered between the genetic messages on DNA and the written messages of a human language. This discovery opened the way for the application of information theory to biology. Information theory applies to any symbol system, regardless of the elements of that system. The so-called Shannon information laws apply equally well to human language, Morse code, and the genetic code. Hubert P. Yockey notes in the Journal of Theoretical Biology:

'It is important to understand that we are not reasoning by analogy. The sequence hypothesis [that the exact order of symbols records the information] applies directly to the protein and the genetic text as well as to written language and therefore the treatment is mathematically identical.'

"There is an identity of structure between DNA (and protein) and written linguistic messages. ... We are not dealing with anything like a superficial resemblance between DNA and a written text. We are not saying DNA is like a message. Rather, DNA is a message."

-- Dr. Charles Thaxton

eric said...

Dr. Sullivan: "It's easy (for experts, not for me!) to look at a string of ones and zeros in a computer and tell what concept they correspond to, and how the manipulation of the symbols can produce a corresponding manipulation in my concepts at the end of the computation."

Actually, not so, even for an expert. The reason is important and relevant.

Because the "string of ones and zeros" is a string of symbols, the meaning is extrinsic. It is not contained in or intrinsic to the symbols themselves. There is an external association between each symbol and the reality it represents. That is what makes it symbolic information. The same ones and zeros might represent text, or sound waves, or image colors, or executable instructions, or a protein's amino acid sequence, or something else.

In exactly the same way and for exactly the same reason, a string of bases of DNA or RNA do not intrinsically represent anything at all. Some sequences aren't even meant to be interpreted as representing proteins. (Many DNA sequences instead represent RNA ribozymes.)

Even when they do represent protein sequences, the meaning of the bases is extrinsic to the sequences themselves -- they are symbols -- and cannot be translated into their associated meaning apart from having knowledge of the particular corresponding code -- a true code.

I know some people are uncomfortable with the idea, but I know of no logical basis upon which to deny the reality that the cell depends on symbols that encode information which must be decoded according to a true code. This is now basic cell biology.

eric said...

Regarding clarification of concepts...

In the context of codes, a symbol represents something other than itself according to an association that is extrinsic to itself. To say that this association is implemented by an external convention is another way to express that there is nothing internal to the symbol that requires or specifies what it represents. A code is a coordinated set of such associations between symbols and their meaning.

In Morse code, three dots represents the letter "S" while three dashes represents the letter "O", but it could have been the other way around, or either could have represented something else, such as a period to end a sentence.

The same is true in a genetic code. Each genetic code maps from a set of 64 possible triplets of bases (i.e. codons) to 20 different amino acids plus a STOP code indicating the end of a protein, similar to a period ending a sentence.

Every genetic code has a built in redundancy such that for each amino acid there are on average about three different codons that each represent it. (This redundancy is arranged in a way that can tolerate the most common coding errors far better than a random assignment of associations.)

Just as an amino acid (or STOP code) need not be represented by a particular codon, when we compare different genetic codes we find that a given codon need not represent a particular amino acid. In one code, a given codon might represent one amino acid while in certain other organisms the very same codon might represent a different amino acid or else might represent the STOP code.

We also know that certain parts of the genome don't represent amino acids for proteins at all. The sequence of bases might instead by used to specify the sequence needed for an RNA strand that functions as a ribozyme (an RNA enzyme).

The idea that genetic codes are "just an analogy" to codes, not actual functioning codes, does not hold up under scrutiny. It is essential that the information encoded within DNA according to that organism's genetic code must be replicated during reproduction. This information is real, symbolic and functionally necessary to the nature of cellular life. Beyond doubt, it is a fundamental aspect of understanding what cellular life is.

eric said...

Perhaps it will help if I address a couple of your particular concerns.

Dr. Sullivan: "One thing that I find suspicious is that the language ID people use to describe living structures, in terms of codes, computers, robots, and so forth, are suspiciously derived from the most distinctive technological aspects of our own day, the innovations most indicative of our own time."

When encountering sophistication that is far beyond anything we have yet produced, I think you would agree we cannot very well describe it in terms any technology we do not yet know of, and it would hardly help to use terminology from more primitive technologies. So, in exactly what way is it "suspicious" to use the most fitting terms now available to us? What are examples of what you would propose instead that would not be "suspicious" but that would be at least as accurate?

The implied charge seems to be that the terms are not appropriate. Yet so far, that charge has not been substantiated in any way. In what way do we see clearly that the terms are not appropriate, if indeed they are not?

Dr. Sullivan: "But the ID proponents are claiming that their language is not metaphorical, that living things are actually computers and robots. This is, forgive me, just a priori improbable and suspicious."

[Keep in mind this does not mean that this is all that life consists of. But science must describe what it has discovered, and it has.]

Exactly how is it "improbable"? Compared to what exactly? Is there are more "probable" way that cells should operate? Are you completely certain that the "a priori" judgement you expressed is not in any way a prejudgement, i.e. a prejudicial feeling or expectation?

The right thing to do is to define terms (as I just did with regard to the codes and symbols of translation) and see if the terms fit or not. I can indicate why the terms are appropriate. Can you indicate clearly why they are not?

I will address one other concern of yours, perhaps the key one, in the next post.

My question to you remains this:

What if it is true and correct that the cell is this way -- not as a metaphor? Is that a problem for your philosophy? I wouldn't have expected it to be, but perhaps it is.

Suppose for discussion this description is the truth. What are the consequences if that is so?

eric said...

[Is this your key concern? Or something else?]

Dr. Sullivan: "While there is any analogy with DNA in the sense that strings of particles undergo operations which have determined end results, what I'm not clear about is the reasonableness of saying that the strings of particles correspond to concepts. And if they don't, then strictly speaking there isn't a language and there isn't a code."

Let us clarify what kinds of languages the cell uses and what kind of concepts are represented in its information and programming.

The languages of the cell are for the purpose of programming and controlling the system of the cell. They are programming languages. Remember that the cell would not work by "growing" duplicates of what it needs via directly copying existing items. It must assemble them on-demand, based on stored instructions and control programming.

There are many levels at which we can see the necessary foresight and associated concepts embodied in its languages and programming.

1. The instructions that define how to build a protein do not themselves have the properties of the protein. Rather, they store the knowledge of what is required to make various functional proteins. Functional proteins are too rare to guess at. Stored knowledge is essential.

2. The machinery of the translation system itself would have no purpose unless and until appropriate symbolic information to translate exists. A specific DVD player implies the concept of a suitable DVD that matches the system.

3. The proteins assembled by that system often cannot function in isolation. The stored DNA information represents not only knowledge about individual proteins, but also knowledge about systems of proteins that must function together to work, often combined into molecular machines.

4. These complex machines may require assembly by other molecular machines specialized for that purpose (and which must also be assembled). The existence of a system of worker machines organized and selectively scheduled for staged assembly a bacterial flagellum implies the concept of the bacterial flagellum.

[Part 2 next]

eric said...

5. There exists error detection systems, e.g. to stop assembly if there has been a failure. The existence of those safeguard systems also imply the concept of an intended result that the actual process must conform to even during assembly, or else preprogrammed consequences ensue.

6. There are also systems for the correction of certain kinds of errors. Such systems imply not only the recognition of a way the system should be, but they also implement means by which it can be restored to that intended state.

7. The genetic code is not the only level of coding. We have now begun to uncover more sophisticated levels of coded programming. We've known that there are far, far more proteins produced by the human genome than we have genes to produce them from. We are now learning how the cell is programmed to reorder that information on the fly as needed. One gene may have packed into it the information for numerous different proteins. The higher level programming knows when to keep parts of the gene's recipe while leaving out others. Sometimes it reorders them as a step in producing the right protein for what is needed at that time. There are other still more advanced techniques I won't attempt to describe here.

This higher level of programming incorporates a preprogrammed understanding that is not simply a simple, "push this button, copy this gene" but rather a conceptually rich repertoire of capabilities that each must be employed when it is appropriate.

8. Throughout the system, capabilities and needs are physically separated. For example, the need for a given protein is for a use that comes after translation, while the decision process that determines which protein's recipe to assemble must happen before translation. This level of coordination is impossible apart from a system that has been conceptually integrated at a high level, one that is above the process of symbolic translation.

Michael Sullivan said...

Eric,

When encountering sophistication that is far beyond anything we have yet produced, I think you would agree we cannot very well describe it in terms any technology we do not yet know of, and it would hardly help to use terminology from more primitive technologies. So, in exactly what way is it "suspicious" to use the most fitting terms now available to us? What are examples of what you would propose instead that would not be "suspicious" but that would be at least as accurate?

What I'm objecting to is the notion that any technological language at all ought to be used until it has already been established that the object or system is a product of techne rather than physis, ars rather than natura. My point all along has been that ID is eliding this distinction, and it seems to me that you still are.

I'm afraid I'm still not convinced that a living body is a giant computer and that its genetic structures are encoded software. I'm aware that some biology sources talk like this. The Wikipedia article on genetics, for instance, is full of this sort of talk in a totally unexamined way. But it seems to me that it's not just an accident that computer science and knowledge about DNA arose at about the same time, and there are a lot of contingent cultural reasons why they've been associated. It also seems to me that this association hasn't been thought through very carefully.

It's not sufficient to say that biology has proved beyond the shadow of a doubt that there is a real language and code in the cell. Despite how some biologists express themselves, scientists are by no means unanimous in affirming this unanimously. This is plain by a simple google search.

Michael Sullivan said...

But let's say that you're right and that there is a real language and a real code in the cell. My question is, where is the conceptual content? The fact that a codon corresponds to an amino acid is not enough to say that is represents that amino acid, or that it means it. You also have to show that, in addition to correspondence, there is intentionality. I don't see that at all. If there is a "language" in the cell, then something must be said. But nothing is said in a plant or in the production of a plant except the plant itself. As far as I know the "language" of the cell cannot represent abstract content. The fact that you have a very highly ordered form which undergoes regular transformations to produce other forms is not enough to call the first a "symbol" for the second. If there is a code, what is being encoded? What is the message? Can it be deciphered, interpreted, translated into English to Latin? No, because the "code" does not "represent" conceptual content, but physical forms. Your claim sounds to me a little like saying that in a Transformers toy the car "represents" or "encodes" or "stands for" the robot - but that's not right, even though a similar end-robot may have been folded into a different sort of initial-car. Perhaps this isn't a good analogy, however.

The fact that you have some quotes claiming that there is a message and a language in DNA is not convincing to me unless you can show me, not only that the "information" (a "scientistic" word which fails to distinguish between abstract intentional content and concrete complex formal but non-intentional content) is highly ordered and productive, but also that something can be communicated in this language. For a cell to produce a cell or a tree to produce a tree is not communication in the relevant sense.

In the context of codes, a symbol represents something other than itself according to an association that is extrinsic to itself.

I don't think this is good enough either. The songs of the birds outside my window right now are correlated with effects which are not intrinsically tied to their specific form - that doesn't make them symbolic expressions. The birds are singing, and there's a reason for it, and the "tune" of the song all on its own can't give me the reason. But there's nothing encoded in their song. There's also nothing conventional in it, even though the same effects could be produced by different songs.

What if it is true and correct that the cell is this way -- not as a metaphor? Is that a problem for your philosophy? I wouldn't have expected it to be, but perhaps it is.

No, it's not a problem per se. God could very well have created a universe consisting only of preprogrammed robots carrying a lot of encoded plaintext in their circuitry. However, it seems that he did not.

To the extent that anything is "said" in a living thing, what is said is that living thing itself. The "information" contained deep inside the tree is not "about" anything. Instead the complexity of the DNA is part of the overall act of the tree's energeia preserving itself and its form in being. Everything about a living this needs to be understood in this context, I think, and "technologizing" the parts without reference to the whole will inevitably lead to substituting some outside agent to explain what the tree is doing all by itself, by nature.

eric said...

[In this comment I'll clarify some relevant technical issues, and then respond to some other key points in a separate comment.]

Dr. Sullivan: "The fact that you have a very highly ordered form which undergoes regular transformations to produce other forms is not enough to call the first a "symbol" for the second. ... Your claim sounds to me a little like saying that in a Transformers toy the car "represents" or "encodes" or "stands for" the robot - but that's not right, even though a similar end-robot may have been folded into a different sort of initial-car."

To picture translation as folding and regular transformations would be incorrect and misleading in important ways. Messenger RNA is never transformed into a protein. Rather it provides coded instructions that must be interpreted to guide protein assembly. Throughout translation it remains itself.

In fact, the codons do not even touch the amino acids that they symbolically represent. This relationship is mediated by the genetic code, as implemented by a coordinated set of transfer RNA. Even within each transfer RNA, the anticodon that does pair up with a corresponding codon is logically independent from the distinct region that determines the associated amino acid. This is why alternate genetic codes can and do exist.

Here are just two documented facts through which one can see that this is a symbolic association.

A) Within an organism, multiple physically distinct codon patterns can will be used to represent the very same meaning, i.e. an instruction to add a particular amino acid or to STOP the process of adding amino acids to the protein.

B) Between organisms, the very same codon sequence can have different meanings, i.e. may represent different amino acids or may represent a STOP instruction rather than an amino acid.

Dr. Sullivan: "If there is a code, what is being encoded? What is the message?"

There are actually multiple levels of codes (e.g. see my item 7 in my previous comment). At the lowest and simplest level of messenger RNA, it is the necessary program for specifying the proper assembly of a protein. This is a programming language, such as we use for directing the operation of programmable devices.

Do this. Then do this. Then do this. ... STOP (end of this phase).

It does so by symbols that do not intrinsically indicate these steps. The association is extrinsic and therefore symbolic.

Dr. Sullivan: "Can it be deciphered, interpreted, ..."

Yes, that is exactly what the ribosome does do -- every day in every living cell. That is why it is recognized as translation, (not as transformation).

Dr. Sullivan: "Can it be ... translated into English to Latin? No, because the "code" does not "represent" conceptual content, but physical forms."

On the contrary, we can indeed translate it into English, either as verbose wording or more succinctly using letters from the Latin/Roman alphabet as corresponding equivalent symbols. The same information can be converted to and through expression in other mediums.

Notice that the information expressing the knowledge of the necessary sequence for a particular functional protein does represent conceptual content. The physical form of the information (i.e. the medium) is not relevant to that concept because the properties of the physical form of the information are not the properties of the realized concept it represents.

Just as a kitchen recipe card would not taste like the food it describes, and "H2O" is not wet, so also the coded instructions to make hemoglobin do not have any ability to transport oxygen.

Clarifying Question:
In the past, computers might be given instructions as patterns of holes punched into paper tape according to a code. You would acknowledge that the computer's instructions are symbolic, correct? The same is true for the cell's coded instructions.

eric said...

Dr. Sullivan: "God could very well have created a universe consisting only of preprogrammed robots carrying a lot of encoded plaintext in their circuitry."

Keep in mind, though cells do rely upon a lot of programming and switching, that fact in no way implies that people or organisms as a whole are merely "robots".

Dr. Sullivan: "To the extent that anything is "said" in a living thing, what is said is that living thing itself. The "information" contained deep inside the tree is not "about" anything. Instead the complexity of the DNA is part of the overall act of the tree's energeia preserving itself and its form in being. Everything about a living this needs to be understood in this context, I think, and "technologizing" the parts without reference to the whole will inevitably lead to substituting some outside agent to explain what the tree is doing all by itself, by nature."

The idea that a tree is "preserving itself and its form in being", or that "the DNA is part of" this, or that the information is about "that living thing itself", or that the tree is doing this "all by itself, by nature" -- none of these have been in question at any time. They are non-issues.

The immediate questions concern what has been discovered about the means by which all this happens and the nature of what a cell is.

In the nature-artifacts-and-machines-2 thread, you expressed this view (emphasis added):

"I'm not considering the question of origins here. I'm pulling Socrates' card from the Meno and insisting that we can't ask how something comes to be with any expectation of success before we first clarify what that thing is."

Yet in a comment above you said (emphasis added):

"What I'm objecting to is the notion that any technological language at all ought to be used until it has already been established that the object or system is a product of techne rather than physis, ars rather than natura. My point all along has been that ID is eliding this distinction, and it seems to me that you still are."

Here you seem to be claiming that we must first establish that the object or system is a product of techne rather than physis, ars rather than natura -- clearly questions of origins -- before we are entitled to observe that the genetic code is a true code, or that DNA holds true information that is translated to direct a process of assembly of parts for molecular machines, some of which serve the sole purpose of building even more elaborate molecular machines.

These latter observations are discoveries about aspects of what a cell is, not claims about the origin of cells. The origin question is distinct. These facts about the cell do not directly specify whether cells were created ex nihilo, or were the product of intelligently directed manipulations of preexisting matter, or were the outcome of an undiscovered property of prebiotic physical and chemical processes.

Having distinguished between nature and origins, by your own standards it would be inappropriate to insist on determining the question of origins before clarifying what a cell is and whether the terms I have been using are accurate.

In the nature-artifacts-meaning-and-providence thread, you said:

"... you can't define your terms in such a way that you get your desired conclusion automatically."

By your own standards, we cannot define terms in such a way that you get a desired conclusion of yours automatically. We cannot artificially define terms in such a way as to automatically exclude an ability to recognize such facts as true concerning what a cell is.

For example, if you do not accept that the cell assembles parts and molecular machines, what neutral and reasonable definition of terms (without regard to origins) shows this can be true for preprogrammed robotic assembly, yet is not at least as true for a cell?

eric said...

Here are some relevant thoughts on terminology from a surprising source -- Thomist philosopher Edward Feser.

"…the nature and structure of DNA is exactly the sort of thing we should expect to exist given an Aristotelian metaphysical conception of the world, and not at all what we would expect if materialism were true. The reason is that the notions of DNA, of the gene, and so forth are utterly suffused with goal-directedness and potentiality. It is no accident that terms like "encoding," "information," "blueprint," and the like are often used to describe the workings of DNA, for there is no other way coherently and informatively to describe these workings; and yet the notion of being encoded, or being information, or being a set of instructions, or being a blueprint all involve directedness of something toward an end beyond itself, and thus final causality."

p. 129, The Last Superstition, by Edward Feser

Notice that he mentions both the essential appropriateness of such terminology, and also the aspect of final causality within the cell as seen in the many examples of "directedness of something toward an end beyond itself".

eric said...

[p.s. Special thanks to Daniel Smith for providing that quotation.]

Malcolm Chisholm said...

As someone who designs databases for a living, I would like to add an additional perspective on "design".

When I design something it is not in a vacuum. It is in response to requirements that I am given. There is no design without requirements. Just in case you might think that you could infer the requirements from the design that is produced, I am skeptical based on my experience. The requirements may be quite illogical and arbitrary. The observer of the design cannot be guaranteed to think in the same way as the intelligence providing the requirements. Thus the observer cannot necessarily go from design to requirement. In fact, in my work I am forced to document the chain from requirement to design to prove that I am trying to meet each requirement in my design.

Furthermore, every design decision closes the door on possibilities. When I make a design decision, there is a huge loss. There will be much that my database cannot do. I believe that this principle of foreclosure is fundamental to design.

I think that these aspects of design are worthy of attention in the current debate.