|
Author
|
Topic: Teleology and Causal Paradigms
|
warren_bergerson
Member
Member # 262
|
posted 05. April 2004 10:46
In order to better understand teleological or goal directed causal relationships, I suggest, it is useful to understand the concept of a causal paradigm. As defined here, a causal paradigm exists when you have a deterministic or predictable relationship between cause and effect, but the causal processes and mechanisms connecting cause and effect are variable, changeable, or indeterminant. Causal paradigms are common phenomena in applied science and computer simulations, where predictable relationships between input and output or cause and effect are maintained or reproduced even though the internal or intermediate processes and mechanisms are changed.
Defining causal paradigms begins with a mathematical modeling or set theoretic concept of a causal relationship. Causes and effects, in this relatively standard approach, are modeled as values or properties of space-time locations. An occurrence of a causal relationship is defined by a pair or space-time locations (a cause location and an effect location) and by the associated cause and effect values or properties. A causal relationship is defined as existing if for all occurrences of pairs of cause and effect locations, the cause and effect values can be represented by a mathematical function.
In somewhat simplified terms, a real world or observable causal relationship which satisfies the above mathematical definition can be classified as predictive or deterministic if it is possible and practical to observe and measure values associated with cause and effect variables, if it is possible and practical to identify the mathematical function connecting cause and effect relationships, and if it is possible and practical to measure the value of causal variable and reliably predict the values of effect variables. [ I am ignoring the complicating impact of noise. The predictions described refer to predictions ‘under idealized conditions’.]
Included in the set or class of deterministic or predictive causal relationships satisfying the above criteria, are relationships where there is significant space-time distance between causes and effects. We can label such relationships as discontinuous. For discontinuous cause and effect relationships, it is reasonable to assume the existence of some type of continuous chain of causation.
For most causal relationships analyzed in the physical sciences, discontinuous causal relationships are assumed to be connected by a single unique sets of processes, mechanisms, or forces. The existence of a single set of causal processes or mechanisms provides a useful intuitive explanation of the observable relationship between cause and effect. A single unique set of causal processes may be useful in explaining deterministic causal relationships, but the single set of processes does not appear to be logical requirement for a predictive and deterministic causal relationship.
As defined here, a causal paradigm exists if the requirements are satisfied for a deterministic, predictive discontinuous causal relationship, but there is no evidence for a single unique set of causal processes or mechanisms. If a deterministic causal relationship involves a causal paradigm, there may exist evidence that different occurrences of the relationship involve different processes. There ‘may’ even be evidence that the predicted effect occurs even if one of the known connecting processes is intentionally disrupted. [As defined here, a deterministic discontinuous causal relationship is said to involve a causal paradigm if there is no evidence for a single set of causal processes. Evidence of multiple causal chains is not required.]
It would appear to be relatively easy to create examples of causal paradigms. We could, for example, create a large number of different machines or processes which start with the causal input ‘2+2’ and generate the effect or output ‘4’.
Although beyond the scope of the discussion here, it does not appear difficult to demonstrate the existence of deterministic, predictive, discontinuous goal directed or teleological causal relationships. It would also appear that these observed teleological relationships involve causal paradigms rather than unique sets of causal processes.
IP: Logged
|
|
Evan
Member
Member # 164
|
posted 05. April 2004 19:31
Warren writes,
quote:
Although beyond the scope of the discussion here, it does not appear difficult to demonstrate the existence of deterministic, predictive, discontinuous goal directed or teleological causal relationships. It would also appear that these observed teleological relationships involve causal paradigms rather than unique sets of causal processes.
As usual, it would be interesting to see this "not difficult" demonstration of the existence of this. In fact, it would be useful to just see an example - a real example of some sort, of what you are talking about.
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 06. April 2004 07:48
EVAN,
The point of this thread is the concept of causal paradigms. This phenomena, I suggest, arises when:
1. The space-time locations used to define and quantify the cause variables and the space-time locations used to define and quantify effect variables in a cause and effect relationship are separated by time and location.
2. There exists a deterministic or predictive causal relationship between the values of the cause variables and the values of the effect variables (based on the mathematical definition of causation.)
3. The chain of causation connecting cause and effect is dynamic or variable rather than controlled by a single set of causal processes or mechanisms.
As I stated in my presentation, it does not appear difficult to artificially create causal paradigms. As a simple example, we can create all sorts of different computer programs (causal chains or processes) than generate the same arithmetic output for a given arithmetic input.
If you accept the concept and definitions offered, and if you accept that causal paradigms can be artificially created, then we are in general agreement on the concept introduced and we can move on to the more interesting and complex questions such as 1)do causal paradigms occur in nature, and 2)can we productively use causal paradigms in scientific theories. It would not, however, be productive to move on to more complex topics until there is a meeting of minds on the basic concept discussed.
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 07. April 2004 08:28
The concept of causal paradigms as defined here, raises some interesting questions regarding the nature of causal relationships and predictive scientific theories.
For instance, a causal paradigm suggests the existence of multiple causal paths between a causal event and the corresponding determined causal effect. A deterministic causal process, by contrast, would seem to suggest a single causal path. Does this suggest that causal processes and causal paradigms are mutually exclusive? If scientists identify a predictive or deterministic causal relationship, must the relationship be either a causal process or a causal paradigm(but it can not be both)?
Viewed from a somewhat different perspective, if intelligent human behavior or evolutionary change are controlled by deterministic causal relationships, are these causal relationships in the form of traditional causal processes or are they in the form of causal paradigms?
Manmade goal directed or teleological systems, such as search processes, appear to have the characteristics of causal paradigms. At least some of these systems have the ability to follow different causal paths to a predetermined effect or goal. Does this suggest that deterministic or predictive teleological causal relationships involve causal paradigms rather than deterministic causal processes?
At least IMO, the concept of causal paradigm ‘explains’ some of the issues that arise in analyzing the behavior of biological systems. Most existing analysis appears to be based on the assumption that if there are causal laws or relationships governing behavior then these causal relationships must take the form of deterministic causal processes. If the deterministic causal paradigm is an acceptable alternative to deterministic causal processes, and if deterministic teleological causal relationships involve causal paradigms, then the ‘laws’ or causal relationships governing biological behavior may involve a form of causation not addressed by most existing attempts to formulate theories of biological behavior.
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 08. April 2004 14:06
The concept of causal paradigms suggests that a single causal relationship can have multiple causal pathways. While different causal pathways may produce the same relationship between cause and effect, the performance characteristics can be different for different pathways. Some pathways may be faster, more efficient, or more reliable than others. We might reasonably expect that causal processes such as gravity will predictable performance characteristics. Such would not be the expectation for causal paradigms.
To illustrate this concept consider a set of class of search routines designed to find a maximum fitness value in a particular type of fitness landscape. The members of this class may all be capable of producing the same effect – identify of the maximum fitness value, for the same cause- instruction to find maximum fitness value, but some routines will perform the task faster and more efficiently than others.
If intelligent human behavior, or developmental processes or evolutionary processes are controlled by causal paradigms rather than causal processes, then we would expect that there will be different causal pathways involved. We would further expect that some causal pathways would be faster or more efficient than others (at least with respect to a particular type of behavior, development or evolutionary change).
If human behavior, developmental processes, and evolutionary processes involve causal paradigms, then we might reasonably expect the existence of a process capable of creating and modifying causal pathways. The process responsible for modifying causal pathways might be defined as involving or requiring intelligence.
The formulation, testing and use of scientific theories can be viewed as involving mathematical modeling of causal relationships. If the causal relationships associated with intelligent behavior, developmental processes and evolutionary processes involve causal paradigms rather than deterministic causal processes or chains, then there are significant implications for the form of scientific hypotheses produced. Causal paradigms also have significant implications for the type of testing required.
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 11. April 2004 09:19
In order to logically justify scientific ‘causal’ predictions, you need 1)a predictive algorithm, 2) a set of observed occurrences of the causal and effect relationship which are compatible with the predictive algorithm, and 3)you need a scientific determinism premise or assumption compatible with both the algorithm and the observed data. For complex ‘intelligent’ behavior associated with biological systems, we develop models based on teleological algorithms, and we can fit these algorithms to sets of observed behavior. In order to create scientific ‘teleological’ predictions, we also need a teleological determinism premise or assumption. In greatly simplified terms, a teleological determinism assumption states that a system whose behavior is currently goal directed or teleological will be goal directed or teleological in the future.
I bring up the subject of teleological determinism because there is one well known form of form of scientific determinism, genetic determinism which appears to clearly fit the definition of teleological determinism (and not the more traditional permanent and universal determinism assumption or the stochastic determinism assumption).
There is clearly some type of deterministic or predictable relationship between genetic material and mature phenotypes. Horse genes develop into horses and bird genes develop into birds. However, it is also reasonably clear that there is no simple deterministic process, mechanism, or mapping responsible for transforming from genes to mature forms.
Developmental processes would appear to clearly fit the definition of a causal paradigm. Within limits, the process or path from horse gene to horse can be disrupted and alternative pathways will still result in a horse. There are clearly multiple or redundant causal pathways in developmental processes.
If, as seems reasonable, genetic determinism is a form of teleological determinism, then there should be teleological algorithms which would fit the data associated with developmental processes, and it should be possible to create predictive teleological models and hypotheses relating to the developmental processes.
IP: Logged
|
|
warren_bergerson
Member
Member # 262
|
posted 12. April 2004 10:19
It is probably not obvious, but causal paradigms appear to offer a scientifically acceptable alternative to the traditional materialistic or mechanistic or reductionist approaches to science.
What might be labeled the traditional reductionist approach to science appears to imply that if you have a causal relationship defined or modeled by a scientific theory, then you will have a continuous deterministic set of causal processes connecting cause and effect. If, in traditional science, you have a cause and an effect separated in time and location, then you should be able reduce the relationship to smaller and smaller intermediate steps and find the processes responsible for the observed or hypothesized relationship between cause and effect.
The causal paradigm concept offers an alternative to the mechanistic or materialist connection between cause and effect. The causal paradigm concept does not deny the existence of deterministic causal paths connecting cause and effect. Instead, the causal paradigm concept suggests the existence many different causal paths responsible for a cause and effect relationship. The causal paradigm concept suggests that the number of different causal pathways is so large that it is not useful or practical to attempt to identify them all. The causal paradigm concept also suggests that reductionism is not always a useful form of analysis since reductionism can not by itself deal effectively with very large numbers of very different pathways.
It may be useful to repeat that the causal paradigm concept is not new to science since it is a standard feature of applied science. Engineering or applied science might even be characterized as the use of human intelligence to modify and improve causal paradigms. All that is new here is the suggestion that the concept of causal paradigms may be useful in formulating predictive theories. It is not unreasonable to suggest, that if human intelligence can modify and ‘engineer’ causal paradigms, then possibly other forms of biological intelligence can also modify and engineer causal paradigms.
IP: Logged
|
|
|
Printer-friendly view of this topic