warren_bergerson
Member
Member # 262
|
posted 07. November 2002 07:02
OVERVIEW AND BACKGROUND
The future development of ID as a science can follow one of three sets of standards: metaphysical standards, peer-review standards, or engineering standards. Apparently, based on a comment of Dembski’s from the discussion of No Free Lunch, one of the roadblocks to developing ID on the engineering standard is the lack of a materialistic/reductionist/deterministic model or explanation of ‘intelligent causation’. How, in effect, are organisms capable of producing complex intelligent reactions to the problems of survival. How, are organisms capable of producing complex solutions like the flagella.
The inadequacy of evolutionary algorithm explanations of intelligent causation may not be apparent using peer review standards, but the inadequacy is obvious using engineering standards. No one can actually demonstrate more than trivial levels of intelligent causation using evolutionary algorithms. There are, however, alternatives to EA’s called ‘information generating machines’ or IGM’s which do appear capable of generating intelligent causation.
To understand IGM’s, one must first recognize the distinction between 1)abstract mathematical/logical machines or processes and 2)real world manifestations of logic machines. This is the distinction, commonly used in engineering, between 1)an idealized, optimal or perfectly efficient machine or process operating in an idealized environment and 2) a less than perfectly efficient machine or process operating in a complex, less than ideal real world. An example of this distinction would be the difference between an idealized errorless logic machine with unlimited processing and storage capacity, and 2)a real world computers that on rare occasions malfunctions and has finite processing and storage capacity.
Second, in order to understand intelligent causation, using an engineering approach, one first identifies a set of abstract mathematical processes or machines that can exhibit intelligent causation. Once idealized logic machines capable of intelligent causation have been defined and analyzed, then the practical ‘engineering’ problems of either creating a real world manifestation or demonstrating that an existing phenomena is a real world manifestation can be addressed.
Third, a key concept of the engineering approach is the concept of complex/compound causation. A machine, including a mathematical logic machine, is, or can be defined as, a structured set of compound/complex causal relationships. It is important to note that while the relationships which are characterized as ‘intelligent causation’ or ‘teleological causation’ may be relatively simple, the processes/machines that generate these ‘complex’ causal relationships are not necessarily simple.
DEFINING INTELLIGENT OR TELEOLOGICAL CAUSATION
A causal relationship can be defined as a relationship of the form "S causes R" where S denotes some type of ‘cause’ or ‘input’ or ‘stimulus’ and R denotes some type of ‘effect’, ‘result’, ‘output’ or ‘response’. Causal relationships are defined/modeled in set theoretic terminology by functional relationships of the form f(S)=R where S and R are properties of phenomena meeting certain space-time constraints.
An intelligent or teleological causal relationship is a relationship of the form "S causes R which increases the likelihood of G" where G is some purpose or goal. There are at least three common interpretations of teleological causation. It is useful to briefly consider these alternative interpretations, because only one of the three interpretations is acceptable from a deterministic/materialistic perspective.
The first common interpretation of intelligent causation is that ‘S causes R’ occurs in order to or with the intent of achieving ‘G’. This suggests that ‘S causes R’ occurred with the ‘knowledge’ that some future event would result from ‘S causes R’. This implies, or can be interpreted to mean that, some future event G causes a prior event ‘S causes R’. From a materialistic/deterministic perspective, such an interpretation is not permissible because it implies causation operating backward in time. [The ‘proper’ materialistic/deterministic is that ‘S causes R’ is produced by ‘the expectation of G’. Expectation is a very different concept from intent.’]
The second interpretation of intelligent causation is that there is a inherent or natural relationship between ‘S causes R’ and G. This suggests that 1) that ‘S causes R’ could take many possible forms, and 2)there is some natural law which produces the form of S causes R which is most likely to produce G. This interpretation appears to included in certain metaphysical approaches to design (such as CTMU ) and in the approaches advocated by certain physicists. To my knowledge, no one has demonstrated that this metaphysical interpretation is compatible with the materialistic/deterministic approach.
The third interpretation, like the second, assumes that the relationship ‘S causes R’ is a member of a set of different possible causal relationships. Specifically, this interpretation assumes that S is from set of possible inputs Sx = s1,s2…sn, and R is from some set of possible outputs Ry = r1, r2,…, rm. There is a set of possible functions Fxy which can be constructed with domain Sx and range Ry. At any given point in time t there will be some subset fxyt of Fxy which will produces the form of ‘S causes R’ which increases the likelihood of G. This interpretation suggests that a relationship is intelligent or teleological if there exists some process capable of 1)finding fxyt in Fxy, and 2)changing the form of the causal relationship ‘S causes R’ to a member of fxyt. This is the (my) deterministic/materialistic definition or interpretation of intelligent or teleological causation.
This interpretation of intelligent causation requires the existence of dynamic or changeable causal relationships, and the existence of complex relationships, processes or ‘machines’ capable of producing changes in causal relationships. It is readily demonstrated using set theoretic mathematics that sets of ‘permanent and universal’ causal relationships can be transformed to form these complex types of causal relationships. Dynamic causal relationships and complex processes or logic machines are part of a deterministic universe. I will be glad to discuss mathematical transformations of causal relationships in more detail.
DEFINING THE ‘DESIGN PROCESSES’ WHICH CREATE INTELLIGENT CAUSATION
The materialistic/deterministic definition of intelligent causation requires the existence of a design process capable of finding and implementing changes in dynamic causal relationships that will make the causal relationships teleological or intelligent. Using the engineering approach, the first step in analyzing these design processes is to identify abstract or idealized mathematical processes or ‘logic machines’ capable of generating intelligent teleological causal relationships. For the discussion here, these complex processes are defined or viewed as logic machines and the operation of the logic machine is defined/described in terms of a logic machine or computer programs. [The use of logic machine programs to express complex processes or complex causal relationships is not new.]
The abstract logic machines identified here as capable of generating intelligent causal relationships are called information generating machines or IGM’s. In order to generate intelligent causal relationships these logic machines must be capable of performing the following operations[This is a greatly simplified list of operations. As mentioned on other threads, I will be glad to provide a copy of pseudo code that defines the required operations.]
1. Read input from external environment and classify current environmental conditions st as a specific member of Sx.
2. Generate each of the reactions or outputs in Ry.
3. Given input st identify the causal operator fxyt and identify the appropriate response or output to be generated.
4. Generate alternates fxya to ft from the set Fxy of possible functions.
5. Determine (select) based on input if fa is more likely to achieve goal G than ft.
6. If fa is more likely than ft to achieve goal G then replace.
Generally, it is more convenient to express Fxy and fxyt in terms of multiple causal relationships rather than a single very complex relationship. As should be obvious, breaking fxyt into a number of smaller components is simply a mathematical transformation. If you want the IGM to have the ability to generate creative solutions, you include in the IGM the abilities to:
1. Increase the number of inputs or stimuli in the set Sx
2. Increase the number of potential outputs in the set Ry. And
3. Increase the number of causal relationships.
[It will be recalled that an IGM is a complex logical process consisting of or defined in terms of sets of complex causal relationships. If the IGM operates on itself, (i.e. on the causal relationships making up the IGM), then an IGM can ‘evolve’. This is outside the scope this thread, but it is useful to note that IGM’s address not only the issue of intelligent causation but also the issue of creating living systems from non-living systems.]
I realize the above description of IGM’s is very simplified. I will gladly discuss the subject in more detail.
PERFORMANCE ISSUES
If an IGM is assumed to have unlimited processing and storage capacity/speed, then even a very simple IGM design can produce all the intelligent causation associated with any living system. However, an IGM based on a simple design would require very, very large processing capacity/speed to simulate/model even a relatively simple sets of teleological causal relationships (or simple life forms). [The calculations are relatively simple and I will be glad to discuss them. I will also be happy to discuss the assertion that significant intelligent causation can not be produced by an RM&NS design.]
Simple design IGM’s aren’t very useful in describing how real world biological systems produce intelligent causation, but they are useful in quantifying the ‘brute force’ storage and speed capacities that would be required. These brute force measures provide useful benchmarks for evaluating the feasibility of specific models and theories (like neo-Darwinian theories).
As anyone who has worked with computer systems knows, there are all sorts of techniques, approaches and gimmicks which can be used to reduce the need for storage and processing capacity. Clearly if biological systems produce intelligent or teleological causation by materialistic/deterministic processes such as IGM’s, the real world processes are based on some highly effective techniques, approaches and gimmicks.
SUMMARY
ID is based on the observation that biological systems exhibit intelligent or teleological causation. If ID is to develop as a science based on engineering standards, then there must be some materialistic/deterministic/engineering approach to modeling, simulating and explaining intelligent causation. As outlined above, such an approach does exist.
IP: Logged
|
Printer-friendly view of this topic