Home » Intelligent Design, Philosophy » The superiority of the designer compared to his design

The superiority of the designer compared to his design

According to Richard Dawkins, Intelligent Design does not explain complexity because the designer must be even more complex than the design. In my opinion, it is like to claim that the car industry doesn’t explain cars because the mechanical engineers are more complex than cars. Dawkins says that because has a priori commitment to reductionism/evolutionism, according to which the explanation/cause must always be simpler than the data/effects. What for Dawkins is a supposed “defect” of ID, for me — who haven’t such commitment — is a great merit. Specifically it is a value of coherence with self-evident principles, first of all, the intuitive principle that more cannot come from less.

However Dawkins offers us the opportunity of asking some interesting questions. In what sense we can consider the designer “more complex” than his design? Or, similarly, what is the difference between the former and the latter? What is the relationship between simplicity, complexity, power, and causal sufficiency? In what follows I shall try to answer these questions. We can look at the designer vs. design difference from many perspectives. Among those there are at least the following:

(A) Causation. A first confirmation that the designer cannot be simpler than the design comes from causality. Since more cannot come from less, in general a cause is greater and higher (of course both adjectives meant in non spatial sense) than its effects. In fact while the designer is independent from its designs, in the sense that he doesn’t need them, it doesn’t hold the viceversa: his designs are dependent from him; they owe their existence to him, as effects owe their existence to their cause. Hence a designer is greater and higher than his designs in principle, because is the cause of them. Eventually to measure in practice the complexity of the designs may be easier than to measure the complexity of their designer. Indeed this is due to the qualitative nature of the designer. Quality is harder to measure than quantity by definition. In fact to measure is to quantify and exact quantification of what transcends quantity is impossible. Hence indeed the problems about “calculating” or “measuring” the designer confirm his superiority respect his products.

(B) Mind/matter. We can look at the thing also from a second point of view and we get the same conclusion. A designer must autonomously conceive the design in his mind before to construct it by assembling materials. This means that the Complex Specified Information (CSI) of the design ‘d’ is already inside his mind from the beginning and independent from its fabrication. This mind contains many other things beyond d, call them M (for example, all other potential designs that the designer, yet capable to construct, doesn’t actually construct). So we have that the designer is at least (M+d) > d. The designer is more than the design because mind is more than matter.

(C) Ontology. The higher rank of a designer respect his design can be argued also ontologically. The designer is the agent/knower and his design is the object/known. The agent is always more powerful than the object, because in their relation the former is active while the latter is passive. What is active is more powerful than what is passive. Since metaphysically real knowledge is identification (to know is to be –Aristotle) an autonomous knowledge/complexity internal to the designer must be counterpart of the design, which is the known thing and has complexity k. Let say P the power/complexity the designer has compared to the design (due to his agent/knower status). The designer complexity becomes at least (P+k) > k. The designer is superior to the design because the agent/knower is more than the object/known.

D) Abstract/material. Any designer hosts an abstract idea of his fabricated design. So in the design process we have a hierarchy abstract/material, where the former is more qualitative than the latter. Aquinas says that “numerus [= quantity] stat ex parte materiae” and that human designs are necessarily related to “ideas in the Divine Intellect”. Someone also adds along this line, with similar meaning, that “human designs are imitations of Divine prototypes”. So the abstraction in the designer is more qualitative than the materials of the designs. The designer is more than the design because the abstract is more than the material.

E) Intelligent vs. non intelligent. The designer must be intelligent while the design is not. Here I should open an entire sub-topic to define intelligence. However, there is a way to simplify the problem. We can define intelligence as a source of information. As in a functioning electronic circuit there must be a current/voltage generator and some passive loads or devices, similarly in the design process there is the generator of information (the designer) and the load (the design). Notice that the electric energy in the circuit comes from another more powerful source and this one in turn from another one until to an ultimate power source. Mutatis mutandis there is an analogous chain about the information sources, which leads us to an ultimate information source (see below my notes about the “simplicity” of human mind). Returning to the present dicotomy, the bottom line is that the designer is more than the design because the intelligent is superior to the non intelligent.

All the above five fundamental reasons are consistent and point to eminently qualitative differences between the designer and the design. “Qualitative” means that quantification statements of the sort: “a designer is n times more complex than the design“ would be simplistic. Conversely, we cannot obtain the designer only by means of whatever quantitative operations applied to the design. Of course the formulas in B and C are minimalist. Nulla osta that the designer be something even greater and higher than the sum of above two things (this is in fact the situation of any human and non-human designers). Moreover one should note that ABCDE apply to any designer, human or non-human. Any designer meets all those five conditions and properties in the same time.

But after all Dawkins, with his criticism, had in mind ID, and ID theory tries to measure the complexity in a system with the amount of CSI it contains, expressed in bit. So one could ask: “the designer does or does not contain somehow such CSI?” If the designer does not contain it then ID theory might argue that the designer is simpler. Indeed according to the above qualitative arguments, we know that the designer must autonomously contain such CSI in abstract form. In fact from A: effects are virtually in their cause; from B: material designs must exist as conceptions in a mind; from C: the knower “contains” the known object; from D: abstraction/formalism precedes matter (what David Abel calls the Formalism precedes Physicality, F > P principle — The First Gene, chap.12); from E: the intelligence of the designer must account for the non intelligent design. So also from the mere ID point of view the designer is more complex than the design.

Some have proposed to compute the complexity of a system based on the number of its parts. This way they claim that sometimes a designer seems to have fewer parts than his designs, then he is simpler. It is easy to see that this metric is unuseful and too equivocal. In fact the definition of “part” is fully arbitrary and confusing. Any system can be described at many different levels. Do you define the system at the top level, at the architectural level, at the block level, at the functional level, at the component level, … or finally at the level of atoms? At every level there is a different number of parts. In such hierarchy the number of parts increases from top to bottom. What is the standard level to adopt? I see no reason to choose one. Most important, the parts comparison between the designer and the design is entirely meaningless. For example, a mechanical engineer shares no parts whatsoever with a car. How to compare those entirely different functional hierarchies composed of fully different parts? If we anyway compare them at different levels, e.g. the engineer at the block level while the car at the bottom level (atoms), should we should infer that the car has more parts and is more complex than a human? It is not this way we can compare systems, let alone designer and design.

- – -

Is human mind simple?

Given I have spoken of human designers, there is another misunderstanding to clear, which can be expressed somehow this way: “Human reason/mind is a simple thing, since it can generate a potential infinite number of new ideas, yet we are not infinite beings. So, compared to its output, reason/mind is simpler.” Since mind is what allows human engineers to design it is easy to see how this has a lot to do with our present topic. Reason-is-a-simple-thing is a concept that evolutionists like very much. If we IDers eventually agree, they have way to go to repeat their “ritornelli”: “analogously, chance and necessity, which is a simple thing, can create the complexity of life”; “brain and intelligence are simple after all and can be produced by natural processes” and so on. “An infinite number of new ideas” are much CSI. How can a “simple thing”, (= low CSI) to generate high CSI? And how does that agree with the ID law of conservation of information (“The active information cannot be gotten on the cheap but must always be paid for in kind”, Dembski, Marks, The nature of nature, III, 16)? If a thing has the potentiality to generate much CSI, this potentiality must involve at least as much CSI. I admit it is relatively easy to measure the complexity of the outputs, while is difficult to measure the complexity of the potentiality, the producer. But again the difficulty of calculating the producer is sign of superiority. It is sensible that a higher thing is more difficult to calculate than a lower thing (analogously in mathematics, in calculus integrals are harder to resolve than derivatives).

A finite mind per se has not the power to create infinite CSI. If the human individual mind can potentially do that is because the over-individual intellect interfaces it directly to the ultimate Source of information (God). As a consequence of this unbreakable communication one cannot consider reason/mind as a simple and isolated thing generating infinite complex things, let alone a “less” outputting “more”. Whatever be the “band-width” of the communication channel (depending on the skill of a specific individual mind), information cannot come from nothingness, and it must exist a first top information source powering all the lower information repeaters and destinations. Reason only illusorily seems simple.

- – -

The “simplicity” of God.

Now, a digression is useful to avoid a possible misunderstanding or disagreement with some theological views. Given that the Designer of the universe is God, many theologicians of orthodox traditions claim that God is “simple”. This doesn’t refute my above arguments because God’s “simplicity” has to be meant as “indivisibility”. About the metaphysical Unity [the Being, God] as “simple” here is what the French metaphysician René Guénon writes in “The Reign of Quantity and the Signs of the Times” (chap.11, “Unity and simplicity”):

“That the very unity can be defined as “simple” is true in a very different sense and only because is essentially indivisible. This necessarily excludes any “composition” and implies for it the absolute impossibility of being conceived as composed of parts. […] But in the same time, the principial unity, although absolutely indivisible, is nevertheless of extreme complexity, so to speak, because it contains “eminently” all those things that, descending to the lower levels, constitute the essences or qualitative aspects of the manifested beings”.
[…]
“One doesn’t understand why the things should always begin simple first, then become complex after; on the contrary, if we think that the germ of a being must necessarily contain the virtuality of all that the being will be in the future, and that all the possibilities that he will develop during his existence are already included in him, one is led to think that the origin of all things [the Being] must really be extremely complex, and in fact it is indeed this the qualitative complexity of essence”.
[my translation]

By the way, we could paraphrase Guénon and say: any designer, although constituting a principle of relative unity respect his many designs, is nevertheless more complex than them, because he contains all those things that, descending to the level of matter, constitute the essences or qualitative aspects of his designs. In other words, what holds absolutely for the Great Designer can also be relatively stated for any intelligent designer.

So I think it is appropriate to speak eminently of the “complexity” of God if we carefully distinguish two different points of view. From the absolute viewpoint God is One, simple and without parts, also if He accounts for the multiplicity of all the created things. In a sense in Him all things are “fused but not confused” — as Meister Eckhart said — and in Him “all things are the same and nevertheless distinct” — as Plotinus puts it. Things are perfectly integrated in God as a true Whole, a priori respect parts. From the relative viewpoint, which is ours, God is complex because must account for the complexity of the universe, His design. In this second sense, the complexity of the cosmos is indeed an effect of the complexity of its First Cause. Human reason can deal only with multiplicity and, as a consequence, can have only a relative and partial view, made of distinct parts, of the Unity, when Unity really is not composed of them. It is the analytic understanding, the only one that we humans can have. In the jargon of Eckhart and Plotinus the things “fused” or “the same” in God is related to His “simplicity” (His being One), while the things “not confused” or “distinct” is related to His “complexity” (His being the First Cause of the universal existence). I see a danger in speaking of God’s “simplicity”, without specifying the teological context where it means indivisibility and no composition, because one could erroneously believe that more comes from less, that in the hierarchy of reality the top has less power than the bottom, the worse fallacies of all. Obviously what said responds also to the contradictory view of who claims that higher the intelligence lower must be its complexity. God is the highest Intelligence and has the maximum of complexity.

- – -

The conclusion is that a designer can be less than his designs only in quantitative aspects. For example, a designer can well weight less than the design (e.g. a mechanical engineer is smaller and lighter than a car). But apart from quantity, the designer, from whatever point of view we assume, and whatever aspect we consider, is always superior and more complex than the design.

  • Delicious
  • Facebook
  • Reddit
  • StumbleUpon
  • Twitter
  • RSS Feed

8 Responses to The superiority of the designer compared to his design

  1. It is certainly about time that this answer to Dawkins is given!

    The traditional God of theism can never be simple in the sense that Dawkins demands of any explanation. In fact, that God is taken to be infinite. Necessarily infinite, even. And if that infinity is to mean anything more than in name only, there must be a sense (a ‘potential’ or intellectual sense if necessary) in which God must contain infinite things!

    What we conclude overall, is that the demand that “any cause or designer must be simpler that the product” joins the long line of premises for (apparently) proving materialism, but which are actually equivalent to materialism, and so should be rejected. Other examples are “the cause of everything physical must be physical” or “the physical world is causally closed”.

  2. OP quote: “In my opinion, it is like to claim that the car industry doesn’t explain cars because the mechanical engineers are more complex than cars. Dawkins says that because has a priori commitment to reductionism/evolutionism, according to which the explanation/cause must always be simpler than the data/effects.”

    The analogy is a terrible one (and a clear inversion of Dawkins’ argument).

    The “car industry” explains almost nothing about the process of car design apart from the obvious relation that car designers are likely to work in the car industry. To correctly analogise Dawkins’ comment, you would need to say something like “we can’t explain the complexity of cars unless we understand the complexity of car designers and the organisations that they work in”.

    The rest of the post sets out to define how we should measure the relative complexities of designer and designed – and (surprise!) you end up agreeing with Dawkins: any putative designer must be more complex than the designed object!

    I also note that Dawkins’ statement is not an a priori commitment to reductionism/evolutionism (whatever you mean by the terms). You seem to agree with the reductionist approach, since you then employ that methodology in your own analysis).

    Dawkins’ statement is an inference drawn from observing actual designers and the actual products of the design process. That is, in all cases where we have access to information about actual designers and their designed products, it is always true that the designer is the more complex of the two.

    I also note that Dawkins’ statement has nothing to say about the relative simplicity of causes to products in undesigned processes. He was clearly addressing cases where conscious intelligence is or is proposed to be at work. Which tells us nothing about whether simple, undirected processes are able to produce an appearance of design that can fool pattern-matching organisms like us.

  3. timothya,

    Our interpretations of Dawkins are different. Dawkins’ aim is to debunk ID, no news here. For him, the complexity of the designer is a pretest to deny scientific status to ID. His accusation is that ID doesn’t resolve the problem of the origin of systems because substitutes a complexity X with a complexity Y even greater and less explainable than X. The “gain” in scientific knowledge is null, then ID has to be discarded.

    I have no problem to admit that Dawkins is right about the complexity of the designer, but — differently from him — I don’t consider this a limit of ID theory, whose job is to detect the signs of intelligent agency in the systems, not to explain intelligence in toto.

    Reductionist approaches can be used sometimes, the problem is when reductionism is adopted as unique and absolute paradigma.

  4. The point Dawkins is making is that of infinite regression. The complexity of the object necessitates a designer, the designer is at least, if not more complex than the design.
    Therefore the designer necessitates another designer. And so on..
    His argument assumes materialism. Any magical designers along the chain would put a halt to the regress.

  5. While I’d disagree with the evaluation of ‘God as designer’ given here, I’d just point out some other possible replies.

    * There’s an infinite regress of designers. Dawkins isn’t really in the position to argue that infinite regress of causes fails to be an explanation, given typical atheist cosmological musings.

    * There can be a finite regress of designers, but ID only gets one to a generally inferred designer, whose exact nature or natures are broad in scope.

    * If the charges goes through, then design can never be an explanation or inference even in cases where it’s entirely obvious – say, a car. Because ‘well then you have to explain the designer of the car’, ultimately melting into ‘unless you explain the origin of the universe, you haven’t explained anything’.

    There’s other replies, but the key is this: ID, as I understand it, only intends to make inferences about a given particular, right-in-front-of-us design. It makes no attempt to explain all design that ever was.

  6. LYO: You have some pretty serious unfinished business of accounting for some loaded and unsubstantiated accusations you have made. Kindly, attend to this. KF

  7. Kairosfocus posted: “LYO: You have some pretty serious unfinished business of accounting for some loaded and unsubstantiated accusations you have made. Kindly, attend to this. KF”

    This thread has a topic under discussion. Please either contribute or remain virtually silent.

  8. TA: Pardon, I raised a serious disciplinary matter. KF

Leave a Reply