|
Author
|
Topic: Darwinism, Design, and Ontological Gaps
|
John Bracht
Member
Member # 5
|
posted 05. May 2002 20:39
In a recent Brainstorms thread that got closed down, edmund raised some issues that are worth discussing. I want to discuss some of the issues, hopefully in a positive way that will stimulate a fruitful discussion and demonstrate some important distinctions and subtleties that help resolve the objections raised by edmund. This discussion is less concrete than others and focuses on some philosophical foundations for design theory; I've included a notion of "ontological gaps" at the end which I think is a useful positive notion for design theory and helps to clarify why intelligent design works the way it does.
There were two main issues raised by edmund's post: 1. the "identity of the designer" question, and 2. the suggestion that design and evolution essentially predict the same things and hence design is superfluous.
The refrain to objection 1 goes something like this: "In order to make testable predictions about design, we have to know something of the goals and desires of the designer. Since ID makes no official statement about the identity of the designer, there can be no useful predictions of ID theory."
The flaw is easily seen if we consider cases in which we do, in fact, infer design without knowing anything about the character or nature of the designer. When we drive past Mt. Rushmore and intuitively conclude design, we do so entirely without any knowledge of the sculptors who were responsible. We don't know precisely what their intentions were, or their desires, or what sort of character they may have had. Indeed, none of this is relevant to our conclusion of design. (Granted, we know historically who the designer was, but I'm talking about the average person who comes by the mounument without any such knowledge. I, for one, can't remember the designer's name or anything about them--but I know Rushmore was designed.) A clearer example comes from an exhibit at the Smithsonian which consists entirely of artifacts for which no purpose can be discerned?-but these artifacts are clearly designed. In this case, we don't even know what these artifacts do (even in biology we can figure out what different molecular machines do) but we can concretely identify design. Further examples of design identification without knowledge of the designer include Stonehenge, the Egyptian pyramids, or the mysterious stone heads on Easter Island. In each of these cases, we know virtually nothing about how the monuments were constructed, we know little or nothing about who did the design work, and yet we can reliably and conclusively infer design.
At this point the critic will insist that we do know that human engineers were involved, and hence our design inference is justified; in biology this is not the case. But is this really important? For one thing, we don't know for sure that humans designed Stonehenge; some theories (admittedly a bit wacky) suggest that aliens were behind the monument (I think aliens have been suggested for the Egyptian pyramids too). The design critics always claim that humans build large stone structures and statues, so it's reasonable to conclude that humans designed these monuments. But, as far as I can tell, no identical structures have been built by modern humans (I don't see people erecting huge stones in concentric circles, for example, like Stonehenge; certainly nobody has built solid stone pyramids for tombs at a similar scale to the Egyptian pyramids). The only reasonable comparison between what these ancient people did and what modern humans do is that both build things out of stone. It's worth pointing out, however, that for each of these ancient monuments we don't understand the technology these people used for their respective construction projects; indeed, it's been long a matter of amazement that they were erected by primitive peoples at all. So we can hardly say that people of that age, with that level of technology, are known to build things of that scale and precision. In short, much about these monuments is shrouded in mystery, from the intent of the designers to the method of construction to the identity of the designers themselves. And yet we can detect design in these artifacts.
Furthermore, it is a general principle that designed objects, in and of themselves, tell us little or nothing about the nature or character of the designer. That's just the nature of the game. If all we have are designed artifacts, all we can say is that the designer must have wanted to produce those artifacts. We can look at the function of the designed objects and suppose that function was at least part of the goal of the design. But that's about it. But is that a problem? I suggest that it isn't. After all, just knowing something is designed is a very important piece of information, and there are ways of doing further science with design once it's been detected, science that wouldn't have occurred to us if we don't first detect the design (Mike Gene has done an excellent job of showing this with his various biological systems?-see his "utility of IC" thread for a good example). Mostly, we need to learn to think like engineers and begin to explore the possibility of reverse-engineering these systems, detecting how they were built and exploring topics like the robustness of design, the quality of design, and the effects of natural processes upon that design over time. In fact, the whole intention of intelligent design theory is to detect design without direct knowledge of the designer (all we have are the artifacts left behind) and use that design inference to do good science. To insist that design cannot be fruitful until we know something of the nature of the designer is (1) flat out false, since we've seen examples where design is useful even though we don't know the designer's identity, and (2) an attempt to shut down discussion before it starts because it simply rules out intelligent design de facto (because we have no experimental evidence of the designer).
Be that as it may, I do agree with the design critic that there is one key assumption in intelligent design theory about the nature of the designer, namely, that in at least some aspects, the designer operated in human-like ways. Why is this? Because the only design we're intimately familiar with is human design. Thus, we are good at detecting this design, and any design we are going to be able to detect needs to reflect certain aspects of this design we're already familiar with. In other words, detectable design needs to fit patterns that humans would recognize as a specification (consider the alien sequence of prime numbers from the movie Contact). Otherwise, the design will be "over our head" and slip right through our design-detection filter. Another way to state this assumption is that humans are a type of intelligent agency (we don't know if there are other types) and if humans are to detect the actions of another intelligence, that other intelligence needs to operate in ways that are in some ways human-like.
But this requirement is really not a problem for intelligent design theory as it applies to biology. The fact is, biology reflects the type of design that humans (really advanced humans) might do. Even Richard Dawkins admits "Biology is the study of complicated things that appear to have been designed for a purpose." Francis Crick cautions biologists to keep reminding themselves that what they are looking at wasn't designed, it evolved. Biology already tickles our "design intuition" just as a series of prime numbers from outer space would. The stuff of life looks suspiciously like the best nanoengineering of some advanced human technology; molecular machines that operate in ways precisely analogous to human designed ones. Clearly, the patterns of biological systems fit with what humans already understand as "design." In short, humans will only be able to detect design if it is in some ways human-like; but this criteria has already been fulfilled in biology.
This leads us to point number 2 from Edmund, namely, that design and evolution essentially predict the same things and hence design is superfluous. This is where the waters get tricky. The reason is that Darwinism embraces a form of design: apparent design. This apparent design is, for all intents and purposes, indistinguishable from real design. Furthermore, many predictions of intelligent design theory would seem to overlap with predictions from Darwinian theory. For instance, both theories suggest that organisms should be highly adapted to their environments and good at surviving and reproducing. Survival and reproduction are cases of good design as well as being the results of a Darwinian struggle for survival. So how do we resolve this problem?
First, I want to say that it doesn't have to be an either/or proposition. I personally believe that good design will incorporate Darwinian evolution in strategic ways. There is no argument: evolution happens. The only debate is over whether evolution can do all the engineering work that is attributed to it.
Thus, we have a rather complex situation where both design and evolution predict some similar things. The key fallacy to be avoided is to assume that either one of them is true, a priori. Edmund wanted to assume that since Darwinism predicts the same things as design, design is superfluous. However, we need to distinguish between 1. predicting something, and 2. actually being capable of doing the thing predicted. In other words, both theories need to be carefully examined and careful attention paid to the subtle clues that may allow us to distinguish one from the other. There is an interesting asymmetry here that should also be noted: Design is in-principle capable of explaining everything, while Darwinism is not. To balance this out, we also invoke the asymmetry that when Darwinism can explain something, design is not needed (is superfluous, as Edmund said). In other words, we discipline our design inferences to areas where mechanism is insufficient (hence the eliminative nature of the explanatory filter). What this shakes out to mean is that, like it or not, we need to be very careful to characterize the limits of the Darwinian mechanism. It is in the things that Darwinian theory cannot explain (and where that inadequacy is inherent in the object and not merely an artifact of incomplete knowledge) that design may be legitimately inferred. Like it or not, it seems that the inadequacies of Darwinism are a real part of this debate. Notice that this isn't an argument from ignorance or a "design of the gaps" argument; it is an argument from knowledge of the limits of Darwinism and the abilities of engineers to do things that Darwinism cannot. We have seen engineers put together car engines and myriad other machines that have irreducibly complex interactions of parts; we know that the Darwinian mechanism cannot exercise foresight and planning. If a system requires foresight and planning (as do certain irreducibly complex systems), design is legitimately invoked as a best explanation.
Notice another subtlety: there is a danger of invoking an "evolution of the gaps" argument wherein the real inadequacies of the theory are papered over by definitions and assumptions. It is a real possibility that there may be ontological gaps in nature, gaps that cannot be crossed by the Darwinian mechanism yet are regularly crossed by living things. When the Darwinist accuses the design theorist of an argument from ignorance and tells him to "do some science" and "get back in the lab" to search for the pathway by which the Darwinian mechanism could cross these gaps, he has moved from a scientific position to a position based on blind faith. Furthermore, he has effectively insulated Darwinism from legitimate criticism by simply defining it as "science." In testing a mechanism, it is the nature of science to seek limits on what that mechanism can do; yet many Darwinists assert that their mechanism has no limits and can account for all biological change, even if there is no evidence for the changes in question (or strong evidence against such change). In other words, the burden of proof is on the Darwinists to provide evidence of many of these transitions--they cannot just accuse skeptics of not "doing enough science" or "arguing from ignorance." To simply assert (as is often done) that Darwinism is capable of doing all the engineering work is to beg the question; we really need to be asking whether, in fact, the evidence supports that assertion. That is why intelligent design must involve 1. a careful look at the limitations of Darwinian theory, and 2. a positive observation of the characteristics of intelligences and the hallmarks of design. A fruitful science of intelligent design will arise once we get a good handle on the ontological gaps that can only be crossed by foresight and planning.
John Bracht
[ 06 May 2002, 13:53: Message edited by: John Bracht ]
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 06. May 2002 09:53
You raise two important issues-1)the nature of the designer, and 2)the relationship of ID and Darwinian theory.
THE NATURE OF THE DESIGNER
I think, as you mention, it is a good starting point to assume that a designer must possess human-like intelligence and design capabilities. This starting point does not, however, imply that a designer must have all the features that might typically be associated with human intelligence, such as consciousness. Until human intelligence is more completely understood, it could be misleading to jump to conclusions as to what is incidental and what is essential to intelligence.
One interesting issue, is whether the designer is external or internal. Is the designer like a watchmaker, operating on and creating some external to itself? Or alternatively, is the designer some internal entity or process? Is human intelligence being created by humans? Were complex organisms created by intelligence which exists inside cells?
Although it might be easier to visualize an external designer, the concept of an internal designer can not be eliminated as a possibility. If intelligence can create not only complex designs, but also new forms of intelligence, then it is at least possible that some type of simple chemical intelligence created cell intelligence which created nervous system intelligence which created human intelligence, and human intelligence may someday create artificial forms of intelligence.
A second feature of the designer to be considered is power. How creative, and how powerful are designers. As anyone who has worked with computer simulations knows, even very simple design problems are a lot more complex than they appear on the surface. Design, in computer terminology, requires huge amounts of computing capacity. If designers are internal rather than external, there must exist internal processes with huge computing capacities.
DARWIN VERSUS ID
The terms Darwinian evolution and ID both refer to possible theories and models. If you use concepts that are broad enough and vague enough, either approach becomes untestable and logically compatible with everything that has an could happen in the universe. The relative value of the two approaches can only be determined when specific forms of each approach make a testable prediction on the same subject, and the predictions can not both be true. The issue, as you point out, is not Darwin versus ID, but precisely defined versions of each.
One of the interesting areas where precision becomes important is ‘within life time intelligence and intelligent design’. Is a complex organism ‘designed’ when the male and female cells combine, or do ‘intelligent processes continue to operate to design the organisms during its lifetime"? Within life time design involves a practical issue- "How do you demonstrate within life time design?"- and a theoretical issue- "Which approach does or does not predict within life time design?"
In order to predict within life time design, Darwinian evolution needs to be interpreted as operating on ‘design processes’ rather than as operating on the design of the organism. Can Darwinian evolution be stretched to include evolution of design processes?
If the designer in ID is viewed as an internal process, then ID would predict within life time design. If the designer is assumed to be external, then, within life time design is possible, but it would seem less likely.
If we accept that humans are capable of intelligent design, then we accept the concept of some within lifetime design. Are there other forms of within lifetime design? Are trees capable of within lifetime intelligent design?
[ I hope the above observations are not out of line with the rules for this forum. My comments on the nature of the designer might be viewed as a question-Can the concept of designer be stretched as I suggest? The differences between Darwin and ID are in the specifics not the generalities. Within lifetime design, in my opinion, provides a useful place to define and test specifics.]
IP: Logged
|
|
Janitor@MIT
Member
Member # 125
|
posted 08. May 2002 15:01
Food for thought: Lewens, Tim, “Adaptationism and Engineering,” Biology and Philosophy, Jan. 2001: 17 (1), pp. 1-31.
IP: Logged
|
|
edmund
Member
Member # 206
|
posted 08. May 2002 23:33
Much to my embarrassment, I seem to have confused people rather than shedding light on the topic. I'll try to clean up my mess:
John Bracht:
quote:
The refrain to objection 1 goes something like this: "In order to make testable predictions about design, we have to know something of the goals and desires of the designer. Since ID makes no official statement about the identity of the designer, there can be no useful predictions of ID theory."
The flaw is easily seen if we consider cases in which we do, in fact, infer design without knowing anything about the character or nature of the designer. When we drive past Mt. Rushmore and intuitively conclude design, we do so entirely without any knowledge of the sculptors who were responsible.
I did not make my original point sufficiently clear. My focus is specifically on the generation of testable hypotheses by ID, rather than the question "Is ID useful?" or "Does ID have explanatory power?" In the Mount Rushmore example, ID is used as an explanation for an observed phenomenon. This is certainly legitimate. However, using ID as an explanation does not necessarily allow us to make testable predictions about other phenomena.
Informally, the process of reasoning used to infer design in the case of Mount Rushmore can be eliminative rather than comparative. The strength of the eliminative approach is that it does not need to assume anything about the designer. Its weakness is that there is no evidence which is not compatible with its conclusion; therefore a conclusion reached by the eliminative approach can't make predictions.
This is not to say that it is useless (and I apologize if it sounded as though I were claiming that). But a theory which generates testable hypotheses is definitely stronger. After poring over a number of the threads on this board, I came to the conclusion that ID is perhaps saddling itself with unnecessary difficulties by avoiding theorizing about the attributes of the designer. I say this because, in several threads, contributors to this board are suggesting hypotheses about the designer's abilities, goals, limitations, etc. These hypotheses aren't usually being made explicitly, however. My opinion is that ID arguments could be made stronger if these hypotheses about the designer were made explicitly. The question could then be asked, "If the designer had these attributes, what characteristics should we expect to see in the empirical world?" Once we get to that point, we can do empirical tests. John Bracht is correct in noting that design doesn't necessarily tell us anything about the designer. But we can start with even blind guesses about the designer, test those, and begin narrowing things down.
My second point was that the hypotheses we make about the designer should be of a specific sort, or that we should focus on specific details of those hypotheses. Many of the informal or implicit hypotheses about the designer, as I noted, end up predicting much the same sort of product that evolutionary processes should generate. Of course, this makes it difficult to distinguish evolution from design in these cases. I should note that the idea that design should be considered "superfluous" in these cases is not mine. It's actually Dembski's-- I am not so sure that design should be dismissed in this fashion. However, we can all agree that a stronger case can be made for design when its predictions differ from the predictions of evolutionary theory.
Which was exactly my point. Making hypotheses about the designer is most fruitful if we look for ways in which a system made by that designer would differ from an evolved system. John Bracht points out an excellent example: evolution cannot use foresight or planning. Therefore the hypothesis of a designer which uses foresight to build complex structures would yield hypotheses which predict different biological systems.
There's a potential headache surrounding "complexity" which I think ID could avoid. It's very hard to determine what evolution is and isn't likely to produce in terms of complexity. We are still figuring out what the 'rules' of evolution are, in many cases. Therefore it's very difficult to say for sure that "this complex system absolutely could not have evolved". Much of the debate over complexity revolves around trying to figure out what evolution can't do-- and this question may be intractable.
There is an easy way around this quagmire. Designed systems usually have other distinctive attributes besides complexity. Many of the attributes we often find in design-- beauty, symmetry, order, or a clear task which the designed object is to carry out-- are attributes which evolution would not create (except as epiphenomena in certain cases, and often these cases can be quite clearly defined). The task of ID might be made easier by looking for these less equivocal attributes in biological systems.
While hypothesizing about the nature of the designer is delicate, it seems as though the contributors to this board are already doing it. Is it possible to construct, from various peoples' insights and hunches, a coherent provisional picture of what the designer is like and which attributes it is likely to put in its designs that evolution wouldn't?
--edmund
IP: Logged
|
|
|
Printer-friendly view of this topic